From d1a36572c531bfd341b938bd953952cd8f52f48b Mon Sep 17 00:00:00 2001 From: Florian Felten Date: Wed, 9 Jul 2025 12:13:00 +0200 Subject: [PATCH 1/5] Add example colab notebook --- example_hard_model.ipynb | 7011 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 7011 insertions(+) create mode 100644 example_hard_model.ipynb diff --git a/example_hard_model.ipynb b/example_hard_model.ipynb new file mode 100644 index 00000000..1bdb7538 --- /dev/null +++ b/example_hard_model.ipynb @@ -0,0 +1,7011 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Playing with a model\n", + "\n", + "In this notebook, we will use a Conditional Diffusion model to generate designs." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Preliminaries\n", + "\n", + "First let's make sure we have the necessary packages installed. Be sure to create a virtual environment, then install the required packages for this project in it." + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Obtaining file:///Users/ffelte/Documents/EngiLearn\n", + " Installing build dependencies ... \u001b[?25ldone\n", + "\u001b[?25h Checking if build backend supports build_editable ... \u001b[?25ldone\n", + "\u001b[?25h Getting requirements to build editable ... \u001b[?25ldone\n", + "\u001b[?25h Preparing editable metadata (pyproject.toml) ... \u001b[?25ldone\n", + "\u001b[?25hRequirement already satisfied: engibench[all] in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.0.1a0)\n", + "Requirement already satisfied: tyro>=0.9.2 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.9.2)\n", + "Requirement already satisfied: numpy in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (1.26.2)\n", + "Requirement already satisfied: torch>=2.5.0 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (2.5.1)\n", + "Requirement already satisfied: torchvision>=0.20.1 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.20.1)\n", + "Requirement already satisfied: wandb>=0.18.7 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.19.10)\n", + "Requirement already satisfied: tqdm>=4.67.1 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (4.67.1)\n", + "Requirement already satisfied: matplotlib>=3.9.2 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (3.9.3)\n", + "Requirement already satisfied: denoising_diffusion_pytorch in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (2.1.1)\n", + "Requirement already satisfied: diffusers>=0.33.1 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.33.1)\n", + "Requirement already satisfied: pymoo>=0.6 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.6.1.3)\n", + "Requirement already satisfied: ax-platform>=0.5 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.5.0)\n", + "Requirement already satisfied: scikit-learn>=1.6.0 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (1.6.1)\n", + "Requirement already satisfied: hyppo>=0.5.0 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.5.1)\n", + "Requirement already satisfied: kaleido>=0.2.1 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.2.1)\n", + "Requirement already satisfied: botorch==0.13.0 in ./.venv/lib/python3.12/site-packages (from ax-platform>=0.5->engiopt==0.0.1) (0.13.0)\n", + "Requirement already satisfied: jinja2 in ./.venv/lib/python3.12/site-packages (from ax-platform>=0.5->engiopt==0.0.1) (3.1.4)\n", + "Requirement already satisfied: pandas in ./.venv/lib/python3.12/site-packages (from ax-platform>=0.5->engiopt==0.0.1) (2.2.3)\n", + "Requirement already satisfied: scipy in ./.venv/lib/python3.12/site-packages (from ax-platform>=0.5->engiopt==0.0.1) (1.14.1)\n", + "Requirement already satisfied: ipywidgets in ./.venv/lib/python3.12/site-packages (from ax-platform>=0.5->engiopt==0.0.1) (8.1.6)\n", + "Requirement already satisfied: plotly>=5.12.0 in ./.venv/lib/python3.12/site-packages (from ax-platform>=0.5->engiopt==0.0.1) (6.0.1)\n", + "Requirement already satisfied: pyre-extensions in ./.venv/lib/python3.12/site-packages (from ax-platform>=0.5->engiopt==0.0.1) (0.0.32)\n", + "Requirement already satisfied: sympy in ./.venv/lib/python3.12/site-packages (from ax-platform>=0.5->engiopt==0.0.1) (1.13.1)\n", + "Requirement already satisfied: typing_extensions in ./.venv/lib/python3.12/site-packages (from botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (4.12.2)\n", + "Requirement already satisfied: gpytorch==1.14 in ./.venv/lib/python3.12/site-packages (from botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (1.14)\n", + "Requirement already satisfied: linear_operator==0.6 in ./.venv/lib/python3.12/site-packages (from botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (0.6)\n", + "Requirement already satisfied: pyro-ppl>=1.8.4 in ./.venv/lib/python3.12/site-packages (from botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (1.9.1)\n", + "Requirement already satisfied: multipledispatch in ./.venv/lib/python3.12/site-packages (from botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (1.0.0)\n", + "Requirement already satisfied: threadpoolctl in ./.venv/lib/python3.12/site-packages (from botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (3.6.0)\n", + "Requirement already satisfied: jaxtyping in ./.venv/lib/python3.12/site-packages (from gpytorch==1.14->botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (0.3.2)\n", + "Requirement already satisfied: mpmath<=1.3,>=0.19 in ./.venv/lib/python3.12/site-packages (from gpytorch==1.14->botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (1.3.0)\n", + "Requirement already satisfied: importlib-metadata in ./.venv/lib/python3.12/site-packages (from diffusers>=0.33.1->engiopt==0.0.1) (8.7.0)\n", + "Requirement already satisfied: filelock in ./.venv/lib/python3.12/site-packages (from diffusers>=0.33.1->engiopt==0.0.1) (3.16.1)\n", + "Requirement already satisfied: huggingface-hub>=0.27.0 in ./.venv/lib/python3.12/site-packages (from diffusers>=0.33.1->engiopt==0.0.1) (0.30.2)\n", + "Requirement already satisfied: regex!=2019.12.17 in ./.venv/lib/python3.12/site-packages (from diffusers>=0.33.1->engiopt==0.0.1) (2024.11.6)\n", + "Requirement already satisfied: requests in ./.venv/lib/python3.12/site-packages (from diffusers>=0.33.1->engiopt==0.0.1) (2.32.3)\n", + "Requirement already satisfied: safetensors>=0.3.1 in ./.venv/lib/python3.12/site-packages (from diffusers>=0.33.1->engiopt==0.0.1) (0.4.5)\n", + "Requirement already satisfied: Pillow in ./.venv/lib/python3.12/site-packages (from diffusers>=0.33.1->engiopt==0.0.1) (11.0.0)\n", + "Requirement already satisfied: numba>=0.46 in ./.venv/lib/python3.12/site-packages (from hyppo>=0.5.0->engiopt==0.0.1) (0.61.2)\n", + "Requirement already satisfied: autograd>=1.3 in ./.venv/lib/python3.12/site-packages (from hyppo>=0.5.0->engiopt==0.0.1) (1.7.0)\n", + "Requirement already satisfied: contourpy>=1.0.1 in ./.venv/lib/python3.12/site-packages (from matplotlib>=3.9.2->engiopt==0.0.1) (1.3.1)\n", + "Requirement already satisfied: cycler>=0.10 in ./.venv/lib/python3.12/site-packages (from matplotlib>=3.9.2->engiopt==0.0.1) (0.12.1)\n", + "Requirement already satisfied: fonttools>=4.22.0 in ./.venv/lib/python3.12/site-packages (from matplotlib>=3.9.2->engiopt==0.0.1) (4.55.0)\n", + "Requirement already satisfied: kiwisolver>=1.3.1 in ./.venv/lib/python3.12/site-packages (from matplotlib>=3.9.2->engiopt==0.0.1) (1.4.7)\n", + "Requirement already satisfied: packaging>=20.0 in ./.venv/lib/python3.12/site-packages (from matplotlib>=3.9.2->engiopt==0.0.1) (24.2)\n", + "Requirement already satisfied: pyparsing>=2.3.1 in ./.venv/lib/python3.12/site-packages (from matplotlib>=3.9.2->engiopt==0.0.1) (3.2.0)\n", + "Requirement already satisfied: python-dateutil>=2.7 in ./.venv/lib/python3.12/site-packages (from matplotlib>=3.9.2->engiopt==0.0.1) (2.9.0.post0)\n", + "Requirement already satisfied: cma==3.2.2 in ./.venv/lib/python3.12/site-packages (from pymoo>=0.6->engiopt==0.0.1) (3.2.2)\n", + "Requirement already satisfied: alive-progress in ./.venv/lib/python3.12/site-packages (from pymoo>=0.6->engiopt==0.0.1) (3.2.0)\n", + "Requirement already satisfied: dill in ./.venv/lib/python3.12/site-packages (from pymoo>=0.6->engiopt==0.0.1) (0.3.8)\n", + "Requirement already satisfied: Deprecated in ./.venv/lib/python3.12/site-packages (from pymoo>=0.6->engiopt==0.0.1) (1.2.18)\n", + "Requirement already satisfied: joblib>=1.2.0 in ./.venv/lib/python3.12/site-packages (from scikit-learn>=1.6.0->engiopt==0.0.1) (1.4.2)\n", + "Requirement already satisfied: networkx in ./.venv/lib/python3.12/site-packages (from torch>=2.5.0->engiopt==0.0.1) (3.4.2)\n", + "Requirement already satisfied: fsspec in ./.venv/lib/python3.12/site-packages (from torch>=2.5.0->engiopt==0.0.1) (2024.9.0)\n", + "Requirement already satisfied: setuptools in ./.venv/lib/python3.12/site-packages (from torch>=2.5.0->engiopt==0.0.1) (75.6.0)\n", + "Requirement already satisfied: docstring-parser>=0.16 in ./.venv/lib/python3.12/site-packages (from tyro>=0.9.2->engiopt==0.0.1) (0.16)\n", + "Requirement already satisfied: rich>=11.1.0 in ./.venv/lib/python3.12/site-packages (from tyro>=0.9.2->engiopt==0.0.1) (13.9.4)\n", + "Requirement already satisfied: shtab>=1.5.6 in ./.venv/lib/python3.12/site-packages (from tyro>=0.9.2->engiopt==0.0.1) (1.7.1)\n", + "Requirement already satisfied: typeguard>=4.0.0 in ./.venv/lib/python3.12/site-packages (from tyro>=0.9.2->engiopt==0.0.1) (4.4.1)\n", + "Requirement already satisfied: click!=8.0.0,>=7.1 in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (8.1.7)\n", + "Requirement already satisfied: docker-pycreds>=0.4.0 in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (0.4.0)\n", + "Requirement already satisfied: gitpython!=3.1.29,>=1.0.0 in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (3.1.43)\n", + "Requirement already satisfied: platformdirs in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (4.3.6)\n", + "Requirement already satisfied: protobuf!=4.21.0,!=5.28.0,<7,>=3.19.0 in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (5.28.3)\n", + "Requirement already satisfied: psutil>=5.0.0 in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (6.1.0)\n", + "Requirement already satisfied: pydantic<3 in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (2.11.3)\n", + "Requirement already satisfied: pyyaml in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (6.0.2)\n", + "Requirement already satisfied: sentry-sdk>=2.0.0 in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (2.19.0)\n", + "Requirement already satisfied: setproctitle in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (1.3.4)\n", + "Requirement already satisfied: accelerate in ./.venv/lib/python3.12/site-packages (from denoising_diffusion_pytorch->engiopt==0.0.1) (1.2.0)\n", + "Requirement already satisfied: einops in ./.venv/lib/python3.12/site-packages (from denoising_diffusion_pytorch->engiopt==0.0.1) (0.8.0)\n", + "Requirement already satisfied: ema-pytorch>=0.4.2 in ./.venv/lib/python3.12/site-packages (from denoising_diffusion_pytorch->engiopt==0.0.1) (0.7.7)\n", + "Requirement already satisfied: pytorch-fid in ./.venv/lib/python3.12/site-packages (from denoising_diffusion_pytorch->engiopt==0.0.1) (0.3.0)\n", + "Requirement already satisfied: gymnasium>=1.0.0 in ./.venv/lib/python3.12/site-packages (from engibench[all]->engiopt==0.0.1) (1.0.0)\n", + "Requirement already satisfied: datasets>=3.1.0 in ./.venv/lib/python3.12/site-packages (from datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (3.1.0)\n", + "Requirement already satisfied: pyarrow<20.0.0,>=15.0.0 in ./.venv/lib/python3.12/site-packages (from engibench[all]->engiopt==0.0.1) (18.1.0)\n", + "Requirement already satisfied: xxhash in ./.venv/lib/python3.12/site-packages (from datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (3.5.0)\n", + "Requirement already satisfied: multiprocess<0.70.17 in ./.venv/lib/python3.12/site-packages (from datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (0.70.16)\n", + "Requirement already satisfied: aiohttp in ./.venv/lib/python3.12/site-packages (from datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (3.11.7)\n", + "Requirement already satisfied: six>=1.4.0 in ./.venv/lib/python3.12/site-packages (from docker-pycreds>=0.4.0->wandb>=0.18.7->engiopt==0.0.1) (1.16.0)\n", + "Requirement already satisfied: gitdb<5,>=4.0.1 in ./.venv/lib/python3.12/site-packages (from gitpython!=3.1.29,>=1.0.0->wandb>=0.18.7->engiopt==0.0.1) (4.0.11)\n", + "Requirement already satisfied: cloudpickle>=1.2.0 in ./.venv/lib/python3.12/site-packages (from gymnasium>=1.0.0->engibench[all]->engiopt==0.0.1) (3.1.0)\n", + "Requirement already satisfied: farama-notifications>=0.0.1 in ./.venv/lib/python3.12/site-packages (from gymnasium>=1.0.0->engibench[all]->engiopt==0.0.1) (0.0.4)\n", + "Requirement already satisfied: llvmlite<0.45,>=0.44.0dev0 in ./.venv/lib/python3.12/site-packages (from numba>=0.46->hyppo>=0.5.0->engiopt==0.0.1) (0.44.0)\n", + "Requirement already satisfied: pytz>=2020.1 in ./.venv/lib/python3.12/site-packages (from pandas->ax-platform>=0.5->engiopt==0.0.1) (2024.2)\n", + "Requirement already satisfied: tzdata>=2022.7 in ./.venv/lib/python3.12/site-packages (from pandas->ax-platform>=0.5->engiopt==0.0.1) (2024.2)\n", + "Requirement already satisfied: narwhals>=1.15.1 in ./.venv/lib/python3.12/site-packages (from plotly>=5.12.0->ax-platform>=0.5->engiopt==0.0.1) (1.37.0)\n", + "Requirement already satisfied: annotated-types>=0.6.0 in ./.venv/lib/python3.12/site-packages (from pydantic<3->wandb>=0.18.7->engiopt==0.0.1) (0.7.0)\n", + "Requirement already satisfied: pydantic-core==2.33.1 in ./.venv/lib/python3.12/site-packages (from pydantic<3->wandb>=0.18.7->engiopt==0.0.1) (2.33.1)\n", + "Requirement already satisfied: typing-inspection>=0.4.0 in ./.venv/lib/python3.12/site-packages (from pydantic<3->wandb>=0.18.7->engiopt==0.0.1) (0.4.0)\n", + "Requirement already satisfied: charset-normalizer<4,>=2 in ./.venv/lib/python3.12/site-packages (from requests->diffusers>=0.33.1->engiopt==0.0.1) (3.4.0)\n", + "Requirement already satisfied: idna<4,>=2.5 in ./.venv/lib/python3.12/site-packages (from requests->diffusers>=0.33.1->engiopt==0.0.1) (3.10)\n", + "Requirement already satisfied: urllib3<3,>=1.21.1 in ./.venv/lib/python3.12/site-packages (from requests->diffusers>=0.33.1->engiopt==0.0.1) (2.2.3)\n", + "Requirement already satisfied: certifi>=2017.4.17 in ./.venv/lib/python3.12/site-packages (from requests->diffusers>=0.33.1->engiopt==0.0.1) (2024.8.30)\n", + "Requirement already satisfied: markdown-it-py>=2.2.0 in ./.venv/lib/python3.12/site-packages (from rich>=11.1.0->tyro>=0.9.2->engiopt==0.0.1) (3.0.0)\n", + "Requirement already satisfied: pygments<3.0.0,>=2.13.0 in ./.venv/lib/python3.12/site-packages (from rich>=11.1.0->tyro>=0.9.2->engiopt==0.0.1) (2.18.0)\n", + "Requirement already satisfied: about-time==4.2.1 in ./.venv/lib/python3.12/site-packages (from alive-progress->pymoo>=0.6->engiopt==0.0.1) (4.2.1)\n", + "Requirement already satisfied: grapheme==0.6.0 in ./.venv/lib/python3.12/site-packages (from alive-progress->pymoo>=0.6->engiopt==0.0.1) (0.6.0)\n", + "Requirement already satisfied: wrapt<2,>=1.10 in ./.venv/lib/python3.12/site-packages (from Deprecated->pymoo>=0.6->engiopt==0.0.1) (1.17.2)\n", + "Collecting pyoptsparse@ git+https://github.com/mdolab/pyoptsparse#egg=7e862db49bbc8bde2b49dee686cce68c8d2ab9f0 (from engibench[all]->engiopt==0.0.1)\n", + " Cloning https://github.com/mdolab/pyoptsparse to /private/var/folders/fw/69hmbxm946l1xycg8w29nw0800l9gd/T/pip-install-f89x5s5_/pyoptsparse_d30cd54894d5404ba70e7fae5d361a3e\n", + " Running command git clone --filter=blob:none --quiet https://github.com/mdolab/pyoptsparse /private/var/folders/fw/69hmbxm946l1xycg8w29nw0800l9gd/T/pip-install-f89x5s5_/pyoptsparse_d30cd54894d5404ba70e7fae5d361a3e\n", + " Resolved https://github.com/mdolab/pyoptsparse to commit 9d417c740693dc706fef8ff2c064837fea04ec50\n", + " Installing build dependencies ... \u001b[?25ldone\n", + "\u001b[?25h Getting requirements to build wheel ... \u001b[?25ldone\n", + "\u001b[?25h Preparing metadata (pyproject.toml) ... \u001b[?25ldone\n", + "\u001b[?25hRequirement already satisfied: cvxopt>=1.3.2 in ./.venv/lib/python3.12/site-packages (from engibench[all]->engiopt==0.0.1) (1.3.2)\n", + "Requirement already satisfied: seaborn in ./.venv/lib/python3.12/site-packages (from engibench[all]->engiopt==0.0.1) (0.13.2)\n", + "Requirement already satisfied: mmapy>=0.3.0 in ./.venv/lib/python3.12/site-packages (from engibench[all]->engiopt==0.0.1) (0.3.0)\n", + "Requirement already satisfied: ceviche>=0.1.3 in ./.venv/lib/python3.12/site-packages (from engibench[all]->engiopt==0.0.1) (0.1.3)\n", + "Requirement already satisfied: zipp>=3.20 in ./.venv/lib/python3.12/site-packages (from importlib-metadata->diffusers>=0.33.1->engiopt==0.0.1) (3.21.0)\n", + "Requirement already satisfied: comm>=0.1.3 in ./.venv/lib/python3.12/site-packages (from ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (0.2.2)\n", + "Requirement already satisfied: ipython>=6.1.0 in ./.venv/lib/python3.12/site-packages (from ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (9.2.0)\n", + "Requirement already satisfied: traitlets>=4.3.1 in ./.venv/lib/python3.12/site-packages (from ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (5.14.3)\n", + "Requirement already satisfied: widgetsnbextension~=4.0.14 in ./.venv/lib/python3.12/site-packages (from ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (4.0.14)\n", + "Requirement already satisfied: jupyterlab_widgets~=3.0.14 in ./.venv/lib/python3.12/site-packages (from ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (3.0.14)\n", + "Requirement already satisfied: MarkupSafe>=2.0 in ./.venv/lib/python3.12/site-packages (from jinja2->ax-platform>=0.5->engiopt==0.0.1) (3.0.2)\n", + "Requirement already satisfied: typing-inspect in ./.venv/lib/python3.12/site-packages (from pyre-extensions->ax-platform>=0.5->engiopt==0.0.1) (0.9.0)\n", + "Requirement already satisfied: pyMKL>=0.0.3 in ./.venv/lib/python3.12/site-packages (from ceviche>=0.1.3->engibench[all]->engiopt==0.0.1) (0.0.3)\n", + "Requirement already satisfied: aiohappyeyeballs>=2.3.0 in ./.venv/lib/python3.12/site-packages (from aiohttp->datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (2.4.3)\n", + "Requirement already satisfied: aiosignal>=1.1.2 in ./.venv/lib/python3.12/site-packages (from aiohttp->datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (1.3.1)\n", + "Requirement already satisfied: attrs>=17.3.0 in ./.venv/lib/python3.12/site-packages (from aiohttp->datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (24.2.0)\n", + "Requirement already satisfied: frozenlist>=1.1.1 in ./.venv/lib/python3.12/site-packages (from aiohttp->datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (1.5.0)\n", + "Requirement already satisfied: multidict<7.0,>=4.5 in ./.venv/lib/python3.12/site-packages (from aiohttp->datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (6.1.0)\n", + "Requirement already satisfied: propcache>=0.2.0 in ./.venv/lib/python3.12/site-packages (from aiohttp->datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (0.2.0)\n", + "Requirement already satisfied: yarl<2.0,>=1.17.0 in ./.venv/lib/python3.12/site-packages (from aiohttp->datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (1.18.0)\n", + "Requirement already satisfied: smmap<6,>=3.0.1 in ./.venv/lib/python3.12/site-packages (from gitdb<5,>=4.0.1->gitpython!=3.1.29,>=1.0.0->wandb>=0.18.7->engiopt==0.0.1) (5.0.1)\n", + "Requirement already satisfied: decorator in ./.venv/lib/python3.12/site-packages (from ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (5.2.1)\n", + "Requirement already satisfied: ipython-pygments-lexers in ./.venv/lib/python3.12/site-packages (from ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (1.1.1)\n", + "Requirement already satisfied: jedi>=0.16 in ./.venv/lib/python3.12/site-packages (from ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (0.19.2)\n", + "Requirement already satisfied: matplotlib-inline in ./.venv/lib/python3.12/site-packages (from ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (0.1.7)\n", + "Requirement already satisfied: pexpect>4.3 in ./.venv/lib/python3.12/site-packages (from ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (4.9.0)\n", + "Requirement already satisfied: prompt_toolkit<3.1.0,>=3.0.41 in ./.venv/lib/python3.12/site-packages (from ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (3.0.51)\n", + "Requirement already satisfied: stack_data in ./.venv/lib/python3.12/site-packages (from ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (0.6.3)\n", + "Requirement already satisfied: mdurl~=0.1 in ./.venv/lib/python3.12/site-packages (from markdown-it-py>=2.2.0->rich>=11.1.0->tyro>=0.9.2->engiopt==0.0.1) (0.1.2)\n", + "Requirement already satisfied: opt-einsum>=2.3.2 in ./.venv/lib/python3.12/site-packages (from pyro-ppl>=1.8.4->botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (3.4.0)\n", + "Requirement already satisfied: pyro-api>=0.1.1 in ./.venv/lib/python3.12/site-packages (from pyro-ppl>=1.8.4->botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (0.1.2)\n", + "Requirement already satisfied: sqlitedict>=1.6 in ./.venv/lib/python3.12/site-packages (from pyoptsparse@ git+https://github.com/mdolab/pyoptsparse#egg=7e862db49bbc8bde2b49dee686cce68c8d2ab9f0->engibench[all]->engiopt==0.0.1) (2.1.0)\n", + "Requirement already satisfied: mdolab-baseclasses>=1.3.1 in ./.venv/lib/python3.12/site-packages (from pyoptsparse@ git+https://github.com/mdolab/pyoptsparse#egg=7e862db49bbc8bde2b49dee686cce68c8d2ab9f0->engibench[all]->engiopt==0.0.1) (1.8.0)\n", + "Requirement already satisfied: mypy-extensions>=0.3.0 in ./.venv/lib/python3.12/site-packages (from typing-inspect->pyre-extensions->ax-platform>=0.5->engiopt==0.0.1) (1.1.0)\n", + "Requirement already satisfied: parso<0.9.0,>=0.8.4 in ./.venv/lib/python3.12/site-packages (from jedi>=0.16->ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (0.8.4)\n", + "Requirement already satisfied: ptyprocess>=0.5 in ./.venv/lib/python3.12/site-packages (from pexpect>4.3->ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (0.7.0)\n", + "Requirement already satisfied: wcwidth in ./.venv/lib/python3.12/site-packages (from prompt_toolkit<3.1.0,>=3.0.41->ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (0.2.13)\n", + "Requirement already satisfied: future in ./.venv/lib/python3.12/site-packages (from pyMKL>=0.0.3->ceviche>=0.1.3->engibench[all]->engiopt==0.0.1) (1.0.0)\n", + "Requirement already satisfied: wadler-lindig>=0.1.3 in ./.venv/lib/python3.12/site-packages (from jaxtyping->gpytorch==1.14->botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (0.1.5)\n", + "Requirement already satisfied: executing>=1.2.0 in ./.venv/lib/python3.12/site-packages (from stack_data->ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (2.2.0)\n", + "Requirement already satisfied: asttokens>=2.1.0 in ./.venv/lib/python3.12/site-packages (from stack_data->ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (3.0.0)\n", + "Requirement already satisfied: pure-eval in ./.venv/lib/python3.12/site-packages (from stack_data->ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (0.2.3)\n", + "Building wheels for collected packages: engiopt\n", + " Building editable for engiopt (pyproject.toml) ... \u001b[?25ldone\n", + "\u001b[?25h Created wheel for engiopt: filename=engiopt-0.0.1-0.editable-py3-none-any.whl size=16674 sha256=d7d123eb09c72b58f08ef781c4a39dabdce219f28c2c741d1bf51cb9663f2b3c\n", + " Stored in directory: /private/var/folders/fw/69hmbxm946l1xycg8w29nw0800l9gd/T/pip-ephem-wheel-cache-6qjf1rk_/wheels/b6/44/39/3c3f0715f4343d6ef24ce1d8f1606494f927246a0d3d312dcc\n", + "Successfully built engiopt\n", + "Installing collected packages: engiopt\n", + " Attempting uninstall: engiopt\n", + " Found existing installation: engiopt 0.0.1\n", + " Uninstalling engiopt-0.0.1:\n", + " Successfully uninstalled engiopt-0.0.1\n", + "Successfully installed engiopt-0.0.1\n", + "\n", + "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m24.2\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m25.1.1\u001b[0m\n", + "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n" + ] + } + ], + "source": [ + "!pip install -e ." + ] + }, + { + "cell_type": "code", + "execution_count": 28, + "metadata": {}, + "outputs": [], + "source": [ + "# Some imports we will need\n", + "import os\n", + "\n", + "from diffusers import UNet2DConditionModel\n", + "import matplotlib.pyplot as plt\n", + "import numpy as np\n", + "import torch as th\n", + "\n", + "from engiopt import metrics\n", + "from engiopt.diffusion_2d_cond.diffusion_2d_cond import beta_schedule\n", + "from engiopt.diffusion_2d_cond.diffusion_2d_cond import DiffusionSampler\n", + "import wandb\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let's make sure we can interact with the problem in EngiBench." + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(
, )" + ] + }, + "execution_count": 8, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from engibench.problems.heatconduction2d import HeatConduction2D\n", + "\n", + "problem = HeatConduction2D()\n", + "design, _ = problem.random_design()\n", + "problem.render(design)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Training a model\n", + "\n", + "In EngiOpt, we have a collection of models that can be used to generate designs.\n", + "\n", + "Each model is defined under a dedicated folder in the `engiopt` package.\n", + "\n", + "For example, the `diffusion_2d_cond` folder contains the code for the Conditional Diffusion model.\n", + "\n", + "In each model folder, there are two scripts; one for training the model, and one for evaluating the model.\n", + "\n", + "To train a model, we can use the `diffusion_2d_cond.py` script. There is a command line interface that can be used to specify the model parameters.\n", + "\n", + "To see the available options, we can use the `-h` flag.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/gymnasium/spaces/box.py:235: UserWarning: \u001b[33mWARN: Box low's precision lowered by casting to float32, current low.dtype=float64\u001b[0m\n", + " gym.logger.warn(\n", + "/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/gymnasium/spaces/box.py:305: UserWarning: \u001b[33mWARN: Box high's precision lowered by casting to float32, current high.dtype=float64\u001b[0m\n", + " gym.logger.warn(\n", + "\u001b[1musage\u001b[0m: diffusion_2d_cond.py [-h] [OPTIONS]\n", + "\n", + "Command-line arguments.\n", + "\n", + "\u001b[2m╭─\u001b[0m\u001b[2m options \u001b[0m\u001b[2m─────────────────────────────────────────────────────────────────\u001b[0m\u001b[2m─╮\u001b[0m\n", + "\u001b[2m│\u001b[0m -h, --help \u001b[2mshow this help message and exit\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --problem-id \u001b[1mSTR\u001b[0m \u001b[2mProblem identifier.\u001b[0m \u001b[36m(default: beams2d)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --algo \u001b[1mSTR\u001b[0m \u001b[2mThe name of this algorithm.\u001b[0m \u001b[36m(default: \u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m \u001b[36mdiffusion_2d_cond)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --track, --no-track \u001b[2mTrack the experiment with wandb.\u001b[0m \u001b[36m(default: True)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --wandb-project \u001b[1mSTR\u001b[0m \u001b[2mWandb project name.\u001b[0m \u001b[36m(default: engiopt)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --wandb-entity \u001b[1m{None}|STR\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m \u001b[2mWandb entity name.\u001b[0m \u001b[36m(default: None)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --seed \u001b[1mINT\u001b[0m \u001b[2mRandom seed.\u001b[0m \u001b[36m(default: 1)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --save-model, --no-save-model \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m \u001b[2mSaves the model to disk.\u001b[0m \u001b[36m(default: False)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --n-epochs \u001b[1mINT\u001b[0m \u001b[2mnumber of epochs of training\u001b[0m \u001b[36m(default: 200)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --batch-size \u001b[1mINT\u001b[0m \u001b[2msize of the batches\u001b[0m \u001b[36m(default: 32)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --lr \u001b[1mFLOAT\u001b[0m \u001b[2mlearning rate\u001b[0m \u001b[36m(default: 0.0004)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --b1 \u001b[1mFLOAT\u001b[0m \u001b[2mdecay of first order momentum of gradient\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m \u001b[36m(default: 0.5)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --b2 \u001b[1mFLOAT\u001b[0m \u001b[2mdecay of first order momentum of gradient\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m \u001b[36m(default: 0.999)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --n-cpu \u001b[1mINT\u001b[0m \u001b[2mnumber of cpu threads to use during batch \u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m \u001b[2mgeneration\u001b[0m \u001b[36m(default: 8)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --latent-dim \u001b[1mINT\u001b[0m \u001b[2mdimensionality of the latent space\u001b[0m \u001b[36m(default: 100)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --sample-interval \u001b[1mINT\u001b[0m \u001b[2minterval between image samples\u001b[0m \u001b[36m(default: 400)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --num-timesteps \u001b[1mINT\u001b[0m \u001b[2mNumber of timesteps in the diffusion schedule\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m \u001b[36m(default: 250)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --layers-per-block \u001b[1mINT\u001b[0m \u001b[2mLayers per U-NET block\u001b[0m \u001b[36m(default: 2)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m --noise-schedule \u001b[1m{linear,cosine,exp}\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m \u001b[2mDiffusion schedule ('linear', 'cosine', 'exp')\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m│\u001b[0m \u001b[36m(default: linear)\u001b[0m \u001b[2m│\u001b[0m\n", + "\u001b[2m╰────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n", + "\u001b[0m" + ] + } + ], + "source": [ + "!python engiopt/diffusion_2d_cond/diffusion_2d_cond.py -h" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Now, let's say we want to train the model for the heat conduction problem, use weights and biases to track the training process, and save the resulting model in wandb. The resulting command would look like this:" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/gymnasium/spaces/box.py:235: UserWarning: \u001b[33mWARN: Box low's precision lowered by casting to float32, current low.dtype=float64\u001b[0m\n", + " gym.logger.warn(\n", + "/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/gymnasium/spaces/box.py:305: UserWarning: \u001b[33mWARN: Box high's precision lowered by casting to float32, current high.dtype=float64\u001b[0m\n", + " gym.logger.warn(\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mflorian-felten\u001b[0m (\u001b[33mengibench\u001b[0m) to \u001b[32mhttps://api.wandb.ai\u001b[0m. Use \u001b[1m`wandb login --relogin`\u001b[0m to force relogin\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Tracking run with wandb version 0.19.10\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run data is saved locally in \u001b[35m\u001b[1m/Users/ffelte/Documents/EngiLearn/wandb/run-20250709_094019-9lgcuj7t\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run \u001b[1m`wandb offline`\u001b[0m to turn off syncing.\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Syncing run \u001b[33mheatconduction2d__diffusion_2d_cond__1__1752046819\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at \u001b[34m\u001b[4mhttps://wandb.ai/engibench/engiopt\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run at \u001b[34m\u001b[4mhttps://wandb.ai/engibench/engiopt/runs/9lgcuj7t\u001b[0m\n", + "^C\n", + "Traceback (most recent call last):\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/huggingface_hub/utils/_http.py\", line 409, in hf_raise_for_status\n", + " response.raise_for_status()\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/requests/models.py\", line 1024, in raise_for_status\n", + " raise HTTPError(http_error_msg, response=self)\n", + "requests.exceptions.HTTPError: 404 Client Error: Not Found for url: https://huggingface.co/datasets/IDEALLab/heat_conduction_2d_v0/resolve/86f25d44b03df207b32d1a3c2909aedec86dd2fb/heat_conduction_2d_v0.py\n", + "\n", + "The above exception was the direct cause of the following exception:\n", + "\n", + "Traceback (most recent call last):\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/datasets/load.py\", line 1645, in dataset_module_factory\n", + " dataset_script_path = api.hf_hub_download(\n", + " ^^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/huggingface_hub/utils/_validators.py\", line 114, in _inner_fn\n", + " return fn(*args, **kwargs)\n", + " ^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/huggingface_hub/hf_api.py\", line 5475, in hf_hub_download\n", + " return hf_hub_download(\n", + " ^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/huggingface_hub/utils/_validators.py\", line 114, in _inner_fn\n", + " return fn(*args, **kwargs)\n", + " ^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/huggingface_hub/file_download.py\", line 961, in hf_hub_download\n", + " return _hf_hub_download_to_cache_dir(\n", + " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/huggingface_hub/file_download.py\", line 1024, in _hf_hub_download_to_cache_dir\n", + " (url_to_download, etag, commit_hash, expected_size, xet_file_data, head_call_error) = _get_metadata_or_catch_error(\n", + " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/huggingface_hub/file_download.py\", line 1484, in _get_metadata_or_catch_error\n", + " metadata = get_hf_file_metadata(\n", + " ^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/huggingface_hub/utils/_validators.py\", line 114, in _inner_fn\n", + " return fn(*args, **kwargs)\n", + " ^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/huggingface_hub/file_download.py\", line 1401, in get_hf_file_metadata\n", + " r = _request_wrapper(\n", + " ^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/huggingface_hub/file_download.py\", line 285, in _request_wrapper\n", + " response = _request_wrapper(\n", + " ^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/huggingface_hub/file_download.py\", line 309, in _request_wrapper\n", + " hf_raise_for_status(response)\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/huggingface_hub/utils/_http.py\", line 420, in hf_raise_for_status\n", + " raise _format(EntryNotFoundError, message, response) from e\n", + "huggingface_hub.errors.EntryNotFoundError: 404 Client Error. (Request ID: Root=1-686e1ce5-303454961774557100b1b837;74b32938-f841-4655-b879-df3a9f7c33e2)\n", + "\n", + "Entry Not Found for url: https://huggingface.co/datasets/IDEALLab/heat_conduction_2d_v0/resolve/86f25d44b03df207b32d1a3c2909aedec86dd2fb/heat_conduction_2d_v0.py.\n", + "\n", + "During handling of the above exception, another exception occurred:\n", + "\n", + "Traceback (most recent call last):\n", + " File \"/Users/ffelte/Documents/EngiLearn/engiopt/diffusion_2d_cond/diffusion_2d_cond.py\", line 287, in \n", + " training_ds = problem.dataset.with_format(\"torch\", device=device)[\"train\"]\n", + " ^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiBench/engibench/core.py\", line 110, in dataset\n", + " self._dataset = load_dataset(self.dataset_id)\n", + " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/datasets/load.py\", line 2132, in load_dataset\n", + " builder_instance = load_dataset_builder(\n", + " ^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/datasets/load.py\", line 1853, in load_dataset_builder\n", + " dataset_module = dataset_module_factory(\n", + " ^^^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/datasets/load.py\", line 1694, in dataset_module_factory\n", + " ).get_module()\n", + " ^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/datasets/load.py\", line 1041, in get_module\n", + " exported_dataset_infos = _dataset_viewer.get_exported_dataset_infos(\n", + " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/datasets/utils/_dataset_viewer.py\", line 71, in get_exported_dataset_infos\n", + " info_response = get_session().get(\n", + " ^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/requests/sessions.py\", line 602, in get\n", + " return self.request(\"GET\", url, **kwargs)\n", + " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/requests/sessions.py\", line 589, in request\n", + " resp = self.send(prep, **send_kwargs)\n", + " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/requests/sessions.py\", line 703, in send\n", + " r = adapter.send(request, **kwargs)\n", + " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/huggingface_hub/utils/_http.py\", line 96, in send\n", + " return super().send(request, *args, **kwargs)\n", + " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/requests/adapters.py\", line 667, in send\n", + " resp = conn.urlopen(\n", + " ^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/urllib3/connectionpool.py\", line 789, in urlopen\n", + " response = self._make_request(\n", + " ^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/urllib3/connectionpool.py\", line 536, in _make_request\n", + " response = conn.getresponse()\n", + " ^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/urllib3/connection.py\", line 507, in getresponse\n", + " httplib_response = super().getresponse()\n", + " ^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/opt/homebrew/Cellar/python@3.12/3.12.9/Frameworks/Python.framework/Versions/3.12/lib/python3.12/http/client.py\", line 1430, in getresponse\n", + " response.begin()\n", + " File \"/opt/homebrew/Cellar/python@3.12/3.12.9/Frameworks/Python.framework/Versions/3.12/lib/python3.12/http/client.py\", line 331, in begin\n", + " version, status, reason = self._read_status()\n", + " ^^^^^^^^^^^^^^^^^^^\n", + " File \"/opt/homebrew/Cellar/python@3.12/3.12.9/Frameworks/Python.framework/Versions/3.12/lib/python3.12/http/client.py\", line 292, in _read_status\n", + " line = str(self.fp.readline(_MAXLINE + 1), \"iso-8859-1\")\n", + " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/opt/homebrew/Cellar/python@3.12/3.12.9/Frameworks/Python.framework/Versions/3.12/lib/python3.12/socket.py\", line 720, in readinto\n", + " return self._sock.recv_into(b)\n", + " ^^^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/opt/homebrew/Cellar/python@3.12/3.12.9/Frameworks/Python.framework/Versions/3.12/lib/python3.12/ssl.py\", line 1251, in recv_into\n", + " return self.read(nbytes, buffer)\n", + " ^^^^^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/opt/homebrew/Cellar/python@3.12/3.12.9/Frameworks/Python.framework/Versions/3.12/lib/python3.12/ssl.py\", line 1103, in read\n", + " return self._sslobj.read(len, buffer)\n", + " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", + "KeyboardInterrupt\n" + ] + } + ], + "source": [ + "!python engiopt/diffusion_2d_cond/diffusion_2d_cond.py --problem-id \"heatconduction2d\" --track --wandb-entity None --save-model --n-epochs 200 --seed 1" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Restoring a model and using it to generate designs\n", + "\n", + "Now we can: \n", + "1. Download a model from wandb\n", + "2. Use the model to generate designs" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "\u001b[34m\u001b[1mwandb\u001b[0m: Downloading large artifact heatconduction2d_diffusion_2d_cond_model:seed_1, 200.90MB. 1 files... \n", + "\u001b[34m\u001b[1mwandb\u001b[0m: 1 of 1 files downloaded. \n", + "Done. 0:0:1.4\n", + "/var/folders/fw/69hmbxm946l1xycg8w29nw0800l9gd/T/ipykernel_88800/2524680856.py:13: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n", + " ckpt = th.load(ckpt_path, map_location=device)\n" + ] + } + ], + "source": [ + "## Downloading a model from wandb\n", + "problem_id = \"heatconduction2d\"\n", + "seed = 1\n", + "\n", + "# This is the name of the model in wandb, for the sake of the example we will use a model that is stored on the official engiopt wandb project. If you want to use a model that you trained yourself, you can change the artifact_path to the path of your model.\n", + "artifact_path = f\"engibench/engiopt/{problem_id}_diffusion_2d_cond_model:seed_{seed}\"\n", + "\n", + "api = wandb.Api()\n", + "artifact = api.artifact(artifact_path, type=\"model\")\n", + "artifact_dir = artifact.download()\n", + "ckpt_path = os.path.join(artifact_dir, \"model.pth\")\n", + "device = th.device(\"cuda\" if th.cuda.is_available() else \"cpu\")\n", + "ckpt = th.load(ckpt_path, map_location=device)\n", + "\n", + "# We will also need the \"run\" object to get the model configuration which was used to train the model.\n", + "run = artifact.logged_by()\n", + "\n", + "# This is the model definition that was used to train the model. You can find the definition in the engiopt/diffusion_2d_cond/diffusion_2d_cond.py file.\n", + "model = UNet2DConditionModel(\n", + " sample_size=problem.design_space.shape,\n", + " in_channels=1,\n", + " out_channels=1,\n", + " cross_attention_dim=64,\n", + " block_out_channels=(32, 64, 128, 256),\n", + " down_block_types=(\"CrossAttnDownBlock2D\", \"CrossAttnDownBlock2D\", \"CrossAttnDownBlock2D\", \"DownBlock2D\"),\n", + " up_block_types=(\"UpBlock2D\", \"CrossAttnUpBlock2D\", \"CrossAttnUpBlock2D\", \"CrossAttnUpBlock2D\"),\n", + " layers_per_block=run.config[\"layers_per_block\"],\n", + " transformer_layers_per_block=1,\n", + " encoder_hid_dim=len(problem.conditions),\n", + " only_cross_attention=True,\n", + ").to(device)\n", + "\n", + "model.load_state_dict(ckpt[\"model\"])\n", + "model.eval()\n", + "\n", + "# The noise schedule also must be restored\n", + "options = {\n", + " \"cosine\": run.config[\"noise_schedule\"] == \"cosine\",\n", + " \"exp_biasing\": run.config[\"noise_schedule\"] == \"exp\",\n", + " \"exp_bias_factor\": 1,\n", + "}\n", + "betas = beta_schedule(\n", + " t=run.config[\"num_timesteps\"],\n", + " start=1e-4,\n", + " end=0.02,\n", + " scale=1.0,\n", + " options=options,\n", + ")\n", + "ddm_sampler = DiffusionSampler(run.config[\"num_timesteps\"], betas)\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgAAAAGgCAYAAADcjN+JAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/GU6VOAAAACXBIWXMAAA9hAAAPYQGoP6dpAACfJUlEQVR4nO2deXgUVdbGTxIMCUunWRPBYACRTSSasAluEEVBBgQVFSVGBx0FFeO44Ci4x4UPUURRR9xRxhn3BYeJIorIEmRGEHEjgmiCiE0QkEBS3x9o6r1vpW4T0iF2cn7Pw/Pc06e6upZbxc157zk3xnEcRxRFURRFqVfE1vYBKIqiKIpy4NEBgKIoiqLUQ3QAoCiKoij1EB0AKIqiKEo9RAcAiqIoilIP0QGAoiiKotRDdACgKIqiKPUQHQAoiqIoSj1EBwCKoiiKUg/RAYCiKIqi1ENqbQAwc+ZMSUtLk4SEBOnTp48sXbq0tg5FURRFUeodMbWxFsDcuXNl7NixMmvWLOnTp49Mnz5dXnzxRVm7dq20bt067PfLy8vl+++/l6ZNm0pMTMwBOGJFURQlkjiOI9u2bZM2bdpIbGzN/C3666+/SmlpaUT2FR8fLwkJCRHZ1x8Gpxbo3bu3M378+Aq7rKzMadOmjZOXl7dP39+wYYMjIvpP/+k//af/ovzfhg0bauT/mZ07dzopKSkRO86UlBRn586dNXKstUUDOcCUlpZKQUGBTJo0qeKz2NhYycrKksWLF1f6nV27dsmuXbsqbOe3oEUHcTWMZrB9F/r+TWQHoH0p+f5L9nBo3/wnch5nmu//1W0/TZuuIxvjHMeT71L+nWfuA4NHoO3JfsttDn7A8Fz7sbnlZ9Dmazb1VPpgGrR7mK5xe0y7EbSn0242kX0ntD8n3yFk41UIbB1N3nPJ/gna2y0+EfOappPvU9N87hq3fY3pkmSyP4B2E+4R30O7udjB499Kvh/I7oM/Sr6GZMO+rh1reK59xNxyPrQ70F4eI7v5QLd99rum7xPaFp/bG8j3J1YFO3euaJYkrTVcI2hTvMN8vCeTfenfwLj2DfLa/uKjI25jHvAouG3c484h+5LGbnssddevaduO0H6au866E037+/cqmt92NV2P0lc3QzuNfJPouz+vcdvX0rafkb0b2rugXS4i60WkadOmUhOUlpZKUVGRbNiwTgKBQPgvWCgpKZHU1PZSWlpap6IAB3wAsHnzZikrK5PkZPNtmZycLJ9/zv8F7CUvL09uueUWz+exIhL3WxtPJJ624+6FXeGgSvaJ4CszwBsnmiY8w5798oVGP3cnz+8E8Id468ZkwxE38PV43HzNPMeAF5FUF94U98WP3U7LtrZrxPsKBPiI+TrgL5WTj68h2ryfRH+T1SfuPHjATRqRE3+TfQwe/y7y8XVo5NPm3xQxXs3UObiv4KnZuoaISABuZLjnKw7afLQBHr8E4sQP7juxFp/necMPAnz/bS972rPlueAjp14lAfguXzP+LvoDnj5Hx7TNbfJ94nuMPclzjeggynyOR8R7vNh7Kwv017SMGwgEqj0AqKsc8AHA/jBp0iTJzc2tsPeOxlJlh7gdCkeZ28TkMrKD0ObRKv9X8SMa/McXvbHwL9Yi2pT/AkBV6nvyya/8QQq0W7CTgIEVPW18THhuxbybFWS3hZM9Yofh2k1/1mEUhf+q47/G8O/rLeTjFwsef8BzNvw6w7+TNoodfAz2WHxidJiCX0xX269MOwUvahO+qYeFOSYE/ycMWnwiIi0t++FXM3yX/nfg/orPBd8nvhMt4Ge4t9r+sygjn7TiD9x7Ee5vRny+viVfIW/8JRocLeLri/2jp+lK+cgwy6A/8N3nP3U+h74UIh9/1/DzO4nvXBu3mUbvg+/pZfcNtD3Dnt2m+R9om7EYe9/Bc3H4N2qMPeJ9rvdnH3WPAz4AaNmypcTFxUlxsflfTnFxsaSkpFT6nYYNG0rDhvyCVxRFUZRw6ADAjwOeBhgfHy8ZGRmSn59f8Vl5ebnk5+dLv379DvThKIqiKHWaPRH6V/eoFQkgNzdXsrOzJTMzU3r37i3Tp0+X7du3S05OTpX201TcECKGljhEyRP7MHIfIh+HvoyYBIckKbaI0oMZJPfaqLd5Iv4eqRODrB3Jt5nsoNsMM1cFf/dH8m3/zrQbO3AGFMdvTBIAhgDfpv3yiBPjQBweZruTYXGoO2ixedtfLDZfNNoW4p2sklCUVFLwonbk+4a/wy8XtjkMjYQs2/K5WH6ns+kJ0pYYnmeJja9Dd9C0+Oj4GqHdhnwePQ7OJ4ZuaSw9Bvh88ePEz6KpPXQnJz+deA3pntL7IQASAIfJ3yEbpchvxI7xDHmmj1B/jYENOptn3maNuSnexw9pr2fSCeDERJYw+R5jLwxCu1xEfhalNqmVAcDo0aPlxx9/lMmTJ0tRUZGkp6fLvHnzPBMDFUVRFKV6lEn1/4L3zFCpE9TaJMAJEybIhAkTauvnFUVRlHqBzgHwQ9cCUBRFUZR6SFSkAfpxi7gS2D/h8/dpu6oUgmxLtlEPh3e0zDRXQduj6xMYUPKkNHnmAAShzSNRzgsHsY4OgjOc8Wc4+YlOTU7Agi4DTd8R/zLtJZb9MngdWLFOJTsmEy3W9fm6oCgcTgsPQpvv3BJfc5Xp8WTk90Uxt2+QvLYeYks940eWt61CSiMy3DS5ONV70ObHgPXs80HQtmn+jCfB1XOJ4HzooYnnqTAA/5XDc0t8fyOsTQlvlMSEU5pX0l54HgLPEUD4/WDMPKCiXJ5ZFz/DL/UxXaNpDgC+N1nX5/cB3hq+njxVCud2BKG9WypJyawRNALgR1QPABRFURTFjg4A/FAJQFEURVHqIRoBUBRFUeowZVL9WfyaBfCH49Subo3qYZBzPYTq2rK+hhompxpzWm1/FLgo799ZbtqooXGuNIO6HpfH9eb2osDJsxRYOQW921K/m2G5lUskn4Anl2X6TqNtcT4Ga4l8DKjOs+ZPsrSIMQeAVeqqdGVbXj3PF3jVsFZD3+J8bY+2jF8916bVh8hn0+75TvF52+YLMLCvgLmy1UBZaNgPQ5tL69IsCTkBhH4uwswhR1zLhkoReOfCbIRMdRLR+dpjP/OUE2Cs9TLYGYI2VSelLGZcaMyW9y9iP0aeA2AsZkS6vqcuCD6MdM160XumOfg30F5toeIg2eeRfYbPIfwiIi9b9hs5NA3QD5UAFEVRFKUeEtURAEVRFEWxo5MA/YjqAUDeGjdAh2FJDs1yKpr/oqLe0NxwCGd2oJA/l8/FtDD+TQ4kYqqMRwLgfD0jgJjBTgLikGbtXGn8nmmX+bRFaIE0EVPf6GW6DqeMvBYQheRQIl97DAHzOu3DOCZsSA/cdavSlW0PM5VSfd1MesMSqXz/PZFk4+RD5Gzg064q4SSBffWNNKwWXU0JoBOkjLGsw0eAZWJtZWFFzC4acxw5Od8Q+m8pyXzVCtAa+gHnE/ICZSjlUBpg0DQTIb6aRjF+TiG10ZzswWgcQc4C+iGMsX9gulaTJGBL0WQf/gqn/V1BdmJ7MEDBLNkjIh9bfjRi6ADAj6geACiKoiiKHR0A+KFzABRFURSlHqIRAEVRFKUOo1kAfkT1AGCRuCeAkiDr75xGgzok65ds4xKZ4VJ3bOUxWccbAO0TeBHETLKlANqcWHUo2UG3SfrgIbxbgM+FdV7B+Q88QYAyEctARmXNn69LF2hfwr/J4mIQDX6gbVq4LVVOxOwRNAeA6kpjfwjRXjxrWRoaqy21L1zJ3j0+7cpshK+JrSRyF9NFuZ1nwByAEO2F0wCxL9nmvnh+phs5SeeXN9wmP4ucdouaNYc5PVNsjIPitwdfQ5wjYP9PpRQuBO+Vj6nc4uPjTUwCI0hOrtkLuv8t1LV5yV98rG3zpETM55jTQs8mu8M6t50C7XDl0iOHSgB+qASgKIqiKPWQqI4AKIqiKIodjQD4oQMARVEUpQ6jAwA/onoAECeuVoUaO+uMLC2i/r6CfKyLFUGbNSubTsba57Fk/xWNS8nZhuyFoMgfN5Ocd5Ld020ONj18DFgldAv5WH79GPTsvu+Sk/K1bZVVbSVPPXrUT5Vt9Tvh6gDYdHMG7yz9KCX743Xi3GhPfzAmPITIifnlVdH12beVbLwOrB5zjjsWcKDenWOaJ0E+eZ+vTN+faK+F4g9n1Y/Em051KzxC/3y3+Ty5Qpbf5HknPXkD4znxLEpMYJUDEtXp5PCYuN+H09gRz/OE14nn4/zHNGf/4uvy1OiwzUNIIxvLdn9OPr5tWJcFe2Td/C81utA5AIqiKEodZk+E/lWdmTNnSlpamiQkJEifPn1k6dKlvts++eSTEhMTY/xLSLAuVFFtdACgKIqi1GF+TwOszr+qpwHOnTtXcnNzZcqUKbJixQrp2bOnDB48WDZt2uT7nUAgID/88EPFv2+/5RyLyBLVEsAcEQn81g7B50HarjEt2oVRvuGU7zaHNv03tDnaxhFqHE3xynaTyI4/B4yg5UdFzBjb91TP92zuIB3d5uHjDM+xbR4z7FQ4dw75c3jwWWj35RgfRZpt4U1+jELQpsqk0oyjsYYdLmyOJVs59B0kG8PfRaYrZJqYzsWjZ89qgEbuF/cWPL6qPIa8Hw7AYhyaJQD+nZD/z3SlAtVpbiA3RBKAZ/FKgFNw+/EGWO6ZO8B/TXMBhLM/ok1tZWz5KnhWmex/NBj8FxdfswegTX1ljGm2hoemD5UR57uGQg73K1YEDUiz/JTeHVgJmJ9pBn83SL6LyT4fNNdnSD/8p2kashk+/3X9r89p06bJuHHjJCdnr542a9YsefPNN2X27Nly/fXXV/qdmJgYSUlhoazmqOv3QFEURanXRE4CKCkpMf7t2rWr0l8sLS2VgoICycpyFzCJjY2VrKwsWbx4se+R/vLLL3LooYdKamqqDB8+XFavXl2tMw+HDgAURVGUOkzkBgCpqamSlJRU8S8vL6/SX9y8ebOUlZVJcrJZHiw5OVmKiooq/U7nzp1l9uzZ8uqrr8qzzz4r5eXlcswxx8h3331XrbO3EdUSgKIoiqLYiVwa4IYNGyQQCFR82rAh68v7T79+/aRfP1ckO+aYY6Rr167yyCOPyG233Rax30GiegDQ+CuRxr+JjI1RXArRhqxZfwL7oFKvI0lTw9Q5Lnl6H9lYjpS1zxRaMtcQw141XUUk8+PhD+SDOPtc+gCF0z6m61JzDsDom9w264OcFogpRMPWmT7WgD+FNmv+rMdiyiZfMzmR7MPQICEa5z6IiNkJWNel9C2DAtOkOSJ4j22loEUkTClgtG0lehmezzCbbLwOk8nXRPzha0LJco3cOQCceMjSPcL3+yTeAE+dc3Kpr0MlYFbfPf0M52NwGe7GA+gDuZA/AN4xzVlwkGtMl0wn+3G3eeUg0/Ul3Ub8FZ5D45lTAw9K6Wumayptiu8Ovka8X7xXR5PPc9+OcJvn08vjWHo/FEI7BO0d4pk28YcnEAgYAwA/WrZsKXFxcVJcbM6uKi4u3meN/6CDDpKjjjpKvvqK33WRQyUARVEUpQ5z4NMA4+PjJSMjQ/Lz8ys+Ky8vl/z8fOOvfBtlZWXy6aefysEHH1yl364KUR0BUBRFURQ7tbMaYG5urmRnZ0tmZqb07t1bpk+fLtu3b6/IChg7dqy0bdu2Yh7BrbfeKn379pXDDjtMQqGQ3HvvvfLtt9/Kn//852oeuz86AFAURVGUCDN69Gj58ccfZfLkyVJUVCTp6ekyb968iomB69evl9hYNwj/888/y7hx46SoqEiaNWsmGRkZ8tFHH0m3blzLNnJE9wCgTNyBGeaIs/DIYi36k0xXC0rgbwH6VheaS/ASCeWroO3RRfkDmGuwgHTmZ8WfWBJgT7iFTm4KzgFIM303mtrVuVNLKtprab+cZ401EFaSj7VE21i5M9lYBTnAmj+tUGsI8E0oSVyGWn41HKje037pnmO+OZ+nZw6AUQeAdbyMfTweEfOvF1K/8+kAG4Pdl68JZ8Cj7h8kH80faeVmlJfQlnwdbNNxuIp0L7jcpZ+YvidoW+yTnPfPNk4t6EE+LpEtggJ9iHw3mOZ4t/kxPXp9eaXuf8Lsg3+ZL4urjjc3XQvtb0wXVkAWEZFtMEeI5+7wdCfsgvxa5OWr06E9kXwpWfTBRdAeaLrS6GakYekKKFtSskO8awfXCPgfRXX2UXUmTJggEyZMqNS3YMECw77vvvvkvvt4ZlnNEt0DAEVRFEWxErksgLqGTgJUFEVRlHpIVEcAfuzsBkuD8DkHazg8GMBsKM7I4IwssEsp5M/7xUXxuGzwCSQBNAU7RNtyCt4hlt/0xFQnn+e2Y+aRc7xpTnKLWNxyq+kqpOPFRCk+N1t6Ea/MyMGwvhj2P4OcHAnHcqrDOBHMNkJnH++4ga/vJ7oOVQoEomRh5jCKGX7nTsiPJeaMUd7X/9GmmJP5z6fJSbloXtECoOMFGTLe8pPs304+XsVvuWV5yEKyKy+fshd+bNOg7QnAXsEf4L2gvjLDFDxuh+PlcPuWf5n2kMnwJJ9n+jr2Nu2zYI0YvmuFZFNiogHf0SC0Wfk4newTsObwWHLys5nRDoz+5KT+3Bb685HQLtklItOk5tEIgB9RPQBQFEVRFDu1kwUQDagEoCiKoij1EI0AKIqiKHUYlQD8iOoBwHhxS36i1miRFUVEpClIfk0pO4tDIigB/0g+TsHZ7dMW8ermtt9kjRX1zWP4y7zxw9C+7DJy3mma18GSr++ayUdpVI64B5wQL0BclbS/k6jCrFFzdBn5KD3SEHaHcde1lfdl+Lv+aXa8+G6VAoGGMM21oP1+vzLgeD82EzRXv21uialeaS9SMueZvF88QL4mh5omCMaZV5suTrPDZ4Z1+xDZWP3Xlk7IcDFW7meY8JhyDjkDOfQB5u9R/WHKycUuyul6nL47AMq3B3iSwrGmOQ7We9lB/f5R+moI2vzuYLsTtO/naSj8MsEVoOn2c3asfLjebW9cb/r4RYnnE4S2bQ3niKIDAD+iegCgKIqiKHZ0AOCHzgFQFEVRlHqIRgAURVGUOoxGAPyI6gHAD+LmnGMlW1t2s4i5VGg4UIfcFWY/uJwtZ3az3LXR4uPSuijzN+5KzhDZmGj9I6mUUyhRGROih0w0PDv/bW7ZyKctQunuYh5/C/IJL58Nwukic+VMz7KzQ45AizV125K/4ZR7fAxChofvuW1+iWfZVmMdWj5x/E1+ufBjCedGdR9oCoBRi+KGB8l5JudcXwttW20EEWl/eUUz5uoZhutuqkXwMbSpaoG8RTb3HYTvKPYlrkzLqf2tca7JybznINl4/e82PNuXmltu8WmLmMtgi4jMhfY4niDA+ju8PHiaDD9vIWjzu4PfSbitQ/OdYvgE8AXG3YF+qAy+y6WKafqQMVcKQ84HbAqApgH6ohKAoiiKotRDojoCoCiKoih29kgl8bn92EfdI6oHAI3EPQFMPSqtZFs/OLDDIV4MkXAQl8PbuMAXFd31ZLRh5V1OEeSuiiG0O9eYPo4k4jH14LS6bi+a9pmw4tsY05V4j2mfBCfA6U+UOGVcUz63QgqpPgVtLq3Ki2AOMXQVXk2PY5YoIITI15FseAz2hEsirQJGDJsftQZ+G1YClAKmi/85bYn9bPVC09d95cPmB+m40l0T0+e5nhCYvsj0xLUx7f5w0/uTBvANPQgrxR+W0XBV9Gxeyo4fBIyjdyKfpz/AuRaY958z2vDZ5PCprXL1ONazSLt5F67L32lTlkkwUt+cfPyewXcHF4JuQRIAhvXD9cgQtDnVk5UFvGYoZxy4oLoOAPxQCUBRFEVR6iFRHQFQFEVRFDsaAfBDBwCKoihKHUazAPyI6gHAIHG1qkL4vNi7qQGmynAqCutZ6OelbVl2RN2/NaXrdSLtHtR3j2bGmh/qkLxUKKdZpUL7VMphu/Am2jjrr2675ammb4wpUg6c6baTaIlckoQNfZDnAJxv2bYp+QaSLX9Cow85uSs/B23Wt7nOKe7GVMWaWhL/uO940pqMr35NTlxC1ZbCKGLMZ1hreljzxTLNL5Ov+430wRu4s7bk5BcmXEN+wJaTjZM5aD1g1hxtMy5Syc5uD0Y6OZPIxgtD/dVcXpls+kORewpW0+V3BS99jIWYh9ON4grehdDm9wGnAZ4G7b+Sjx/xD6DNJbx5Lo8N/vsZ51lxv7fNowpB26nC7ys1Q1QPABRFURTFzh6p/nQ3lQAURVEUJcrQAYAfOgBQFEVR6jA6APAjqgcA1z4tEvhdIMNkdNYkuUYniFTbSXArpE1xVzwHoB/ZzVD3D5q+GOp/6XAMLFGybo65vDw/gHVIXL6Wu/xA0o/TMNc/b7XpPIm+vMptZlA52jY01wCPiafOFJKNWiina48lWy5F4xByUjWCjVDzgAs4tBzNH0C7i+FJpeoECT7tSjFOnl8gqPPzHAXetqjSpoi376AOzd3eU6NVYGKHPBDmGMAOmZ5Nz5s2divWnbluAebOh1va1jhZfkjomIyJPgN4R/zaW+Q2qYhIzFDTnvSm236C9vIG2fgsriIf1wzArsJ6eweysR5Ci/am7yGaa/AIPJt0m4xy5JUd075Slfn1+Bs6B6D2ieoBgKIoiqLYKZPqz+LXLABFURRFiTI0DdCP6B4ADO8oEvgtADVqMDg4jYrSfn55vaLZmOJ43T8kG0ONnKfWhWxMReLSnxQnG13ots/4zvRtME15Fdrvk4/DvBhi4yjpB2Sn4XJlWev9dyRi1hylGr1NPzFtDI3zY8PhQpRVTiNfIucXNsDl7fgeTzZNrLV8hOmSy2lZNKOAcmfDE5dsSgCNIAWOV17zhKyNa2gTDPhc+OKDRBBmKUu83lzGdj3pBe0WQaJg/0dpaz5eKMxLx8D9bAW0uS+zZIHHy6fGKzHuhGu/g1IRfzJNORyfRU/WZ5B2DC8BrmtNfb0jZEveTnl0of+YNkoCfEerokjzM2SU/2Vd8jzTvAS6eh/SAO6gr2JP53vBP4NSE6ctcjpknE+7XER2ilKbRPcAQFEURVGs7BGRmAjso+6hAwBFURSlDqMDAD90MSBFURRFqYdEeQSgr7iJZDgHgOuC0uitCYjLl1Ni0uUsAmIJV07Xakw2CpO0342UXwQ5OHG0Dm4aCf1XglR7Gs0t4KVD/wFtnoZAEqUMXOe22/6DnCz6gd5ZRJp/iDa1TZcJkn06tC/kzL4z+Nu4xiulLc7+wjA3gqTdlvd7+b/oA+w7tHGcvxluKWmzRiqXn8Xf4TkADHyX7ksjKjGNhMj+L9ntngWjP2/NwJkfbXp4KgzarH1zGivb+3gEnhLInhS2NGgfzU56H+CD8azp8pQ9xqWPaU4QTxGy9ZUA2ThNgTV1PoSJ0B5Oy2uP5BrDT7rN9J6ma8b1pn0ftHlJ9WFk4xuW5yXxGxTvzW5qcxpzzaARAD+ifACgKIqiKDZ0AOCHSgCKoiiKUg/RCICiKIpShymT6kcAtA7AH5Akceu8Yt4ynxaHb9DPCmZLslGfJd1ZSEv+AdQ7FilDZGNCdJB8Q8iGGrkdPzJdeYtN+79wCFwjgKYaGMuVnsl1TA8zzVL4nftMl1AZAwPWOrPIvg6TmieQ85Tm9AHeN7r2U0wT50b0pAMcsYZq4nbFvtPR9JGwi68B1vw9OrQhcPP8kRC0bX1OxDhv0nh5FgrqzpxHzxWyh+Fa0g/zvIhRZMM1OthcCLvF6WannAhzVmi6iOcxwNT0QvKxPvxPaIcrl2vUfmjShpxvmiaUti4lTZ3L/S6D2sZc02AF2dgf+L+PzmTj1ee5Opyvj7/D1+hXqmNy7g1gUG3tFLLHw1rjXL+B35ItYApLBl0IhyYxoO6PNQJKxJyqUXNEIny/f/uYOXOm3HvvvVJUVCQ9e/aUGTNmSO/evcN+74UXXpBzzjlHhg8fLq+88sp+/fa+oBKAoiiKUofZE6F/VWPu3LmSm5srU6ZMkRUrVkjPnj1l8ODBsmnTJuv3CgsL5a9//asce+yxVf7NqhLxAUBeXp706tVLmjZtKq1bt5YRI0bI2rXmKjS//vqrjB8/Xlq0aCFNmjSRUaNGSXExz3VVFEVRlD8OJSUlxr9duzjO5jJt2jQZN26c5OTkSLdu3WTWrFnSqFEjmT17tu93ysrKZMyYMXLLLbdIhw6euFbEibgE8P7778v48eOlV69esmfPHrnhhhvk5JNPls8++0waN94bsLzqqqvkzTfflBdffFGSkpJkwoQJMnLkSFm0aFGYvTPbxQ0w4QiNg7FBsvG0eeBBSS2bnnHbD9OmlDL0M5Td5DKsnJ3T1KctIhLPkW8cpu02XYso1w/PnEPUnF6EgfAzS8hJ0gKGQrmkMH8VA+onkm9qMn1wNbT7kM+z4h/cqzUvGp5XKMyPQWlWY0Y8TR/kYSA1zfRR9DgBxrIc1mV7O4RGG3sC3P2hHS4NEPorDdltI3h+ClgS+hkuTLMSygkLWCQAOd50jTUlgMOh/x7OugM9GO/DMXBJ4UKycb1CymjzhNRNTYBecwvoSYD8SFow0xOOx/A7p8rRo2lcf37Gua/EwIqVJ1Gq50fUtzHKz1IIH/9OSO9NzCQnldpuBxe1Hb8s+P8iLJFMnTCGdCl8n8XDRYrbJt4y3TVC5CSA1NRU49MpU6bIzTff7Nm6tLRUCgoKZNKkSRWfxcbGSlZWlixevNiz/e/ceuut0rp1a7nooovkgw/4TRt5Ij4AmDdvnmE/+eST0rp1aykoKJDjjjtOtm7dKo8//rjMmTNHBg4cKCIiTzzxhHTt2lU+/vhj6du3b6QPSVEURam3RGIC3959bNiwQQIBd2ZTw4a81vheNm/eLGVlZZKcbP7Fk5ycLJ9/zoti7+XDDz+Uxx9/XFauXBmB4903anwS4Nate/9Ebd587zCwoKBAdu/eLVlZ7nSwLl26SLt27WTx4sWVDgB27dplhFpKSvhvTkVRFEWpWQKBgDEAiBTbtm2T888/Xx577DFp2ZInBdccNToAKC8vl4kTJ0r//v3liCP2xnqKiookPj5egsGgsW1ycrIUFVVeFyovL09uueWWmjxURVEUpU6yR0Scau6jalGEli1bSlxcnGduW3FxsaSkpHi2//rrr6WwsFCGDXNrLpaX7xVxGzRoIGvXrpWOHTt6vlddanQAMH78eFm1apV8+OGH4Te2MGnSJMnNza2wS0pKftNiksVNA0ThKVwaIN6AMClYQZgDQNMDXqRVZTFN6VvT5ek+uNwmzw9oSPob6nysM/IyqNjdWAPmY5gP7YF02rwYLA7NWB7k40+HtmfYdi3ZqCWuIl/H/5l2WwidvW26uIfhSq2eEr1c7jUPrxr1FdJjg9Bm/ZVBubuxpxAvr3WM8J2Dm2MpTRzumLjv4H1sxvVch9nmJRSapicXDdqUTsqd8CCYA8B9jq9CCNq85CzPCTDyTXfSUteUSov5se+Si+cl2NJAeQldtG2LQYuIrIa5ZN3TTN8Umqa0Em4k77cLZ5uOhPZo8rU71bSPw0kCPKuC11TGX+Yf5aMKQRtmKZTsEJELpeY58AOA+Ph4ycjIkPz8fBkxYoSI7P0PPT8/XyZM4HznvVHwTz81Z+nceOONsm3bNrn//vs9cw8iRY0NACZMmCBvvPGGLFy4UA45xJ3MlZKSIqWlpRIKhYwogN/ISGSvzuKntSiKoijKH43c3FzJzs6WzMxM6d27t0yfPl22b98uOTk5IiIyduxYadu2reTl5UlCQkJFlPx3fv//kT+PJBEfADiOI5dffrm8/PLLsmDBAmnfvr3hz8jIkIMOOkjy8/Nl1Ki9M43Xrl0r69evl379+kX6cBRFUZR6zYGPAIiIjB49Wn788UeZPHmyFBUVSXp6usybN69iYuD69eslNrZ2S/FEfAAwfvx4mTNnjrz66qvStGnTCl0/KSlJEhMTJSkpSS666CLJzc2V5s2bSyAQkMsvv1z69eunGQCKoihKhKmdAYDI3kh4ZSF/EZEFCxZYv/vkk0/u129WhYgPAB5+eG+y/AknnGB8/sQTT8gFF1wgIiL33XefxMbGyqhRo2TXrl0yePBgeeihh/bj1w4TkUa/tVE+8BRltcB61WDTjIds+TvNrPYzjzM3XQ7a3Demy1MXwHaErONWpeuhLhlOo8YKnny8YZe6BVi4wdT+FheQkw/qVmhzEjOvFXsdrKFMB8zXF/VunifhKQxgLN2cYboogTsIbdZ8+RqZhx8kL/aAcBoqiMCkt4eb54Hw3xrGreAO4AFfFTRPgnL9y55z26zVc1kALDnLefVMI2gfQ76YofQBlM/2CPtUz+NBuLz/NF3Wpa75/nP5DjwEXl2b5yy0QIPl9smmmY45+bzjAL2UjOIarOvzpDLbTAVbOfVw71v0N/BpK7VBjUgA4UhISJCZM2fKzJkzI/3ziqIoigKUSfUjALY/gaIXHYIpiqIodRgdAPgR5QOABuKeQhP63AaGszjdiW0IcPelwqC7zLXt7obo2+uUl/Yo7RUjrqFKj9EFQ7UcduSgna3EsE1a4O7NYfOQ38HJ3mRMpC/O++QDzDXN++GHOZB43Uv8QaHbppPh76IEwOdWSh/EG/2BantTTB3D6OGkGr5XJpuhzf2VLxpsS7KITQLg47OVo/bqJPlkY4ydSgHHmWWZX4c2l9JllYdXnbOBYfKR7KRDMvJwHzBdD1Fp3TnQZnWIUyfx/vNKjCQeyl+h3YJVHl5e7zxoX8l36mqy8eyD5OMegX073HvRIvN4vmuTsGz7xbcFC0Q1xR6p/rI3dXMAoKsBKoqiKEo9JMojAIqiKIpiQyMAfugAQFEURanD6ADAjygfACSIq5fiqbCOz5qqbb4Aa2i4bXfTFbPGtD94s6I57H9/NVzDzjI3nQNi6Kv0i6w7YvoTS4fdyD4a2mksQtMz8DPI3Vzel7XbN6C9knyeJCCQMDfNMl030aa4vGoa78eTrge6JOWM2U6V7358I/rA0FGp71AZ1hB/FeA0MHMlYVsfZFh/hTkA9okFBjwngV+BxmXglEvhtFwU2akXnmyaqTDhhftyiGxbApnteFO4MCj/0Dtuc+W/TRc/b9jNwl0znIfAS11fwtviBn8j56CT6ANcfpnT9fgdhRM2eDlzmuBgXHGe6LGZbPwur3PPdwqvRFXmAGDSMP+GcqCJ8gGAoiiKotgok+r/BV/dLII/JjoAUBRFUeowe0Qkppr7qJsDAM0CUBRFUZR6SJRHALAOAJ5KkLZjTdWvNGVl29qWRWVdbLjbPJJqk35uVgI4d859bnsq7SZENtYUzSJfL7KxTCjLzktMsxksg9qM1j3taK5MKWcsddvDTJcnt3sIfMDSLMv6KOVz3QJPmVMpcJtUupbVRMyBT+LddOIPfH5DxFgqVsRUSfnu8+UOGD9sW5Latp4u0dRqWuF7YWSbc0L+x3SB++J1IY16VDvDzGjvLr87fJ25KR8vVunlcs4MBnFL6YbH81LSy9zmU+SyLdXNgeIg2fj43Ua+xIvpg5uhffDl5ORiwF9Dm+ZfbKJlsXESA68yzSWd8YErETvYQbgz7yDbMsnGoWXS8b7i/d8Z5nAih0YA/IjyAYCiKIqi2NABgB8qASiKoihKPSTKIwA/iRubwlA9h1sZmz9cWVYkZPkuh3GvMM1zL4T207TtIrK52C5CIesFbvhVnqBNV5BtyX9ytpr2XGhzyiCfKUYleYTJVxOj/JxW5YmSrtvktilkHaJNUQLw3EGPtIC8b1ilFC/G3+GUMU/g3gib8lE0sPhCZEN4mO5LVWAJAAuxBni/3Ff6XgvGNHJS7PuqGyuaQ24wXW1IUcNj4kX7ODMxBG2WC9qat00WQd/mFQhtBWhZohhANi7Ml8gVeseTfTD26H+ZvjdnmDYqhHQhVtM1exvapNx5eg52Qe6vtpUjua9w9inul3sv7xefRUyNDbf6Y8Rwyqv/B3zdDABE+wBAURRFUSyUS/WzAOtmHSAdACiKoih1mDKxhzv2dR91EJ0DoCiKoij1kCiPAPwklZcC5tOqSioffzdUhePB3+HymJzqFYR2f/u2Pz/jtnld4YdN8xXQrN82XZ4UPDxz1kVZA0Q9ls+El0W1wRrradA+n8u7ppP9ntssoSrMtpKyvLiqp2avtHSbP39keDzViAFebpejhA5cxBhrein3Ocpx2wNpYCHTZTtvPj7WcRvbnJzbOR9yyE56k5w0WeMMaNMFTH/ZtC+G3+EKza+Rjbv6C/kOo9/BbDjbPRQxr0Mf8rHMHxgLxnnkDJKdDx32cdO1/nnTngntlbQbnu+A+rvt/lcHm45fmd8GPn/4ngk3UytiaATAlygfACiKoiiKBZ0D4ItKAIqiKIpSD9EIgKIoilJ3UQnAlygfALwm7ikMtW1I4GmzVh8iG7Ncq6JasTp3CNm4Ly5Wm2KamOM8xXRNIrkYl+3lfP3qgKGiFuRj+RjnD3CuL88XGI0Gl+jlyQZwHf5JLs4Zx+fVUwrYMwcA7hWVS6ZKwMZ++Z3A8ybQ9sxDMODHkCZD4Iqv9KMcmYzzaYt451/EYZcMkpNPHMXbPiTkByzLwbKoTte+O0xUyaOytl9RB4bqvp4yBZ+Tjdee71OQ7HOgnU2+tqfSB5nQ5vLJL5EN5T2eoXoSz9Km6LbVKagq2AfC/R+G23J/tYWKeb/cJ9H/q8/nNYpKAL6oBKAoiqIo9ZAojwAoiqIoioVyqX64oY5GAKJ7ALBgqRtTHoSlNofThrYwf7hEGpvftpIgywWbLftleeBC0xyR77bPMYOfvZ40N10M7RDtlfswhu45NM9h87bQPop8qWRjFJ1TEfmYPoR2R65ryqlosAraO2H2i+fWlnzShT+AlDyqG8uh5qq8B8wQvK2vcB+jvmOpmWorw8rSjEf5QMml0HT9vNi0m2GOHtfHvSBk2ngNWavhg0K1izSLQ/5j2igBhC3DDHA4mytM39AejHRyHko2rjrIeYpmBqlMhXw9VgdsKbks3QTJRimHL6dtdUiWqPga4juAUzL5mFD9YKmRJQz8HTzvA/Z/qs4B8EUlAEVRFEWph0R3BEBRFEVRbOgkQF90AKAoiqLUXVQC8CW6BwC3iitOTXaXIJWhmbQhq8ConLJ6aLODYQ7ItnQwX2qcl8C/yfsB8fuJyYZnZNBc8zd1utt+ivZCmUiGzteLfJz91B2q5UoaOSkH73wQCAfuMH3fmKaRtjiWUhpjVpr2XBiF83QBJghtzkSTs/gDyD8j0d+WDReOOEOg5XkoqPNz36C3DS5fTCJvK5pawlot4vkjBn5mNWn+b5imdIP7OOw9cl5ApbY3QvtV0/UTbYpaMi8H/CHZ2JVYu+Rzs2mbnhWVcZIAT/r4wDS3b/Z1eY73LWgXk4//P8HbytmwPOViILR5/k3blvQBCvv8miHxvhQmCYRoU+5XOM+Hr8MqsnFfOA9hj3jLHNcIOgDwRecAKIqiKEo9RAcAiqIoSt2lPEL/9oOZM2dKWlqaJCQkSJ8+fWTp0qW+27700kuSmZkpwWBQGjduLOnp6fLMM8/4bh8JdACgKIqi1F3KIvSvisydO1dyc3NlypQpsmLFCunZs6cMHjxYNm3aVOn2zZs3l7/97W+yePFi+d///ic5OTmSk5Mj77zDSc+RI8ZxHKfG9l5DlJSUSFJSkjwjro49sg1ssJEV7Glk2/L1WShDxTA5zLao8/J+WddH3Tfctvg7rCVT/dQvYIHdm2hTyqs2Tmcw+Y6xHBIL8KTz/rzQbY+iTVkfxKK33cjXimycw1BoOTze16ucAP/T0fQBCLtp6w3PSTRxAkvOcn55B7Ln4/UtmkpeLF3Ny1WzfaXbPOZ/hudu0u5xRgj3ziDZR0CbtVj+gwd16CnHkfP9NqZ9r5vtvfJa0/UAfRW7Eh+DbYlqLkfdmWx8X/M8Dj43/C7n1TN4TWl6i6dyta02QYDs06F9G/lizqEPcIoTPzQ9yMaHKJ5+dWeJaeOFoteKZ1LAp9DmiT1kl33ntnHOxzYROVJEtm7dKoEAX5Hq8/v/E1s/EQnYCiTsy762iSQdVbVj7dOnj/Tq1UsefPBBEREpLy+X1NRUufzyy+X666/fp30cffTRMnToULntNu4VkUEjAIqiKErdxZHqh/9/+zO5pKTE+LdrF81c/o3S0lIpKCiQrKysis9iY2MlKytLFi9eXOl3jEN2HMnPz5e1a9fKccfxiDty6ABAURRFqbtEUAJITU2VpKSkin95eXmV/uTmzZulrKxMkpPNqHFycrIUFXGsyGXr1q3SpEkTiY+Pl6FDh8qMGTPkpJNO2t8zD0tUpwHOEzcMux1qTJ4/jQrQ5t5D3wxZ9spJQhjI+458HHjuCO2gZT8i5qXn3B1b2WCmo2keDkHVuXNp21tNcxPUmOU6oby8GqxstvNp0/UwbYopZFzNl6U0HD+zPGArc8u+NLJHouFJ+yPJ5XU37L+IQv4/8VcBDhcHeQMjLk0rPBphfg7581XLcJvnmRLA8fTHBKo8rNTwueBXw62YZ2R6ck7bRipsCzH2cN0KQ8LhUvkw7M+3dDLZMaCUDaHLy1cXba66zMeP2FIuRcxryufG8hYGeGN6h9kxPmD8iHN0+jBoD6SQ/2mmKRnwH00Gi3dV6b/mf3Bx8JSnwVNeUrJbJGmeRBMbNmwwJICGDRtatq46TZs2lZUrV8ovv/wi+fn5kpubKx06dJATTjghor/zO1E9AFAURVEUKxGsAxAIBPZpDkDLli0lLi5OiovN0XJxcbGkpPBgyiU2NlYOO2zvqC09PV3WrFkjeXl5NTYAUAlAURRFqbvUQhpgfHy8ZGRkSH6+u5BbeXm55OfnS79+/fb90MvLfecZRAKNACiKoihKhMnNzZXs7GzJzMyU3r17y/Tp02X79u2Sk5MjIiJjx46Vtm3bVswjyMvLk8zMTOnYsaPs2rVL3nrrLXnmmWfk4YdZZI0cUT0AKBT3BDCUcT5p1JKbTx9gcdhwZXhR3/qX6VpHcw1QKD2SU81YpUTtntMAGVuJYf4u2oeR7xLa7Qy3Tal88rxpzoPyr9wdOc0KK4yyTspngtk5fCcsq+AKJZ7JGWRn4wYDyfmzqaNj7hydtrdsLMBLpvIys2LM3eGwH6Zzfk2+R8mGUtaXmf2q7xKzdu1Y6Pv8GPCcALy+/AcOVXc2tXC+iRvJhikBPN2Jl47F/XJfYRt785/JF8PVv+GH00ii5v6K0eFwuj4eL0eV+bJg/+C+vYHsU6DdlmrFHET2rz5tEe/cjU4wKWTwLNPXg9f1njTfbZ/J7xXOE8YkUu75PEkEj/JkaG+XvTO5aphaKgU8evRo+fHHH2Xy5MlSVFQk6enpMm/evIqJgevXr5fYWPd/ru3bt8tll10m3333nSQmJkqXLl3k2WefldGjR1fz4P2J6gGAoiiKolipxbUAJkyYIBMmTKjUt2DBAsO+/fbb5fbbb9+/H9pPdACgKIqi1F10OWBfdBKgoiiKotRDojoCsEVczc6o9OhJ3uY5AJBXLTzDkvM6g9AmXcxcidesc1pG64oeO8K0/wbt/pNoR+eSjb/LNQKCZKPgyaWLqYBqAIrZ7iDFfa2/SVnfHh0StU8+Oq5UiuohVxR9n2wsDcuVPblaqrFjTuZ+mWxYt5VrEfA8BNSIWdXP4WMw1iGmeg3GfSowXU/ymrRgX3C66XrKvPrn/+gq3MfTFBUuPopLuvI95L8MbPnw8hXZUCaW76FtP/ybvNTtsdBum0ROrsMMHeRUKt9hKznMVaNZzca+z8fLh4DHz32b5yHg88Xb2uYl8PXk5wJrVXiW3uX1wfFmFdGaz03JxjrYXLM5RDZeRJzIsZMPqIYol+pLAHU0AhDVAwBFURRFsaISgC8qASiKoihKPSSqIwBNxD0BI9jNdTY9sa450B5LPltKXtA0l5jmg5Az9jF9sy2FYy8FO20M1ZN+huwYrP3JoWTOwcLYKJ/LILK7uM0rbzBdv5qByByIH3MKE5dWxct/IvnO5/q5mL5Fke+RpM6gm9PzOGh+At4bXtmQctM+ht+xlf4VMUOqnIrYtj19YJw834sgtP/PdNEqjoUQwk57m/SLubQcHOT+tZtouq54zrSnQDvcinlGlJdjySvJXuY2eVE5jhbj9eSezWt6GuV/OVbPOZlwc4aQi7+K/ZVlhw/JxjRRlocuJHsgvJS+IC2Bi5Pj77AcY5VfCO6TV0D7cE77Y80CH2w+cX7NQJ7o6/Sc8vsAHzeMxNdceRuiFrMA/uhE9QBAURRFUazoAMAXlQAURVEUpR6iEQBFURSl7qKTAH2J6gFAurjZKIbOx/lYe6j06zawm3EBz8vIRgXrENNF+Tkok3E6GU0XMDTXs0ibPf+ftPEsWLfzgkvJySmDeEtZd+bb3R3aj5uu6xYZZuCsGyvaU4eam25cY9ptUY/lGr0sUkLK2E4SBTnqhjbL0P8guxPUnG3LAnfING3pcKy/4jpgaeTz5AU2QH2e7wUcxHzz7fI6pa3h9JE+dKLZP9LJ/RvaV5muvtQp/woC/XV0dDzHwrgOHDekdNj7IcORU9p4CgheIT6GgZyR2wvavJ5KL7JB2I9LM13936FtsTPRJJDmVEYY5wuwLM4FnQdCTt7hnU3ftQtNOwjtdy2HJ2L+X8RvL+6Ch+PSwvyfGOVoboQM4/+YLs/7C8+dUyVDZONrEl8NByyqrhKALyoBKIqiKEo9JKojAIqiKIpiRSMAvugAQFEURam7OFJ9Dd+JxIH88YjqAcBt4mqyiagJ8uqJLGhhKvVZ95m+QTwHAFW1PqbrNNPE0quLaS8hsldCmzW+taSF/xnmNKQtp8V4H+R1fHERWCr96zkKVLy5K/Q3zfawbOdndxuutvOoTGghtAOmSz4wTedNt013wlNvAJ9h1uo/JRtzkduuJCcJ+9/4uzxlWFGW5pxxjwBrzLFgQDkl0Zf1VqxxwNckmS79KX8C417a+E+medLnbvt56nN0m4zrO5rmKPA1w2vPef+sWWNp6IGcy88r0GJ55+PJR4+mJMJazCc0MX1XUy0FfD/QetDptClWPeY/ClkLN24WVXA+nB76a2EeDS+uy1OCcNbHDvJ9RPZwWEqY+zLfG5z3wctB87Z47uH+bw1CGx+RqtQ3qBYaAfBF5wAoiqIoSj0kqiMAiqIoimJF0wB9ieoBQOKtIom/xxSxrCXXH6XSqquXu+3uHOv8bDJ9gEU7DzVdl7U2zFMe3lTRXkUpVxzGw+RClgA4SykI7Utnmr7GCbQ231Qs6cvnwgHYELQpTOpZFRGLLc82XaeQDPHLRLf9gOniECueDlVL9qSi2aDqvsb1HriFnGTjFeSwJIfIUNHgTDQ5mj9ACSBEvkK3SXoGyxuYmcahWErelFZwETO45i2HyaESdE96ZpaZpnH0y8nH0VE8fn5v8vXFIt3bKZ7dmLUQPIhC8vEDduh8t51NvtbXmPaZ8Fyf+bThint9qWGPBp3qZJJfPKvtNYY2nzhJGK3hog1fZ/qo2LORdckVe0Nk423l+2T7Py3cc4CwtMCpnvjmwGfmV/G+62oElQB8UQlAURRFUeohUR0BUBRFURQrGgHwRQcAiqIoSt1F5wD4Et0DgP7iStco5E4xN3uEREuUFk+jMrYjHyKR+rJrwQjSAdxqmnf9paL5V0r7SSJR7TVocwobS9aosLOKP55Wko05HjTLYaywsbr8L7fp0K/GUB1ZGWk5Csq7bALzCXL+bPpIq231idvm1Umbko1Hz+lwrFliZt0Qys9izRKy4TwDfc5Mw4quh7ckJ68sbextNflg1kKa6eHrEA9tTs9aSTbOCehGJaYT+faDOMvXujHZ+Lt8jfh64nd5zgKDff9K8g2n6S2DyUa4P+B0gh40b0YmU37khWeCQc/0sM1ku2+PZntmGK5mVBLZyIdMJh/n78FayAH6zyaDLmILOCR+EnlOgGdeAsD3LQht7vec4shLISM8BwBtLKW807IP5cAQ3QMARVEURbGhEoAvNT4J8K677pKYmBiZOHFixWe//vqrjB8/Xlq0aCFNmjSRUaNGSXGxp4yGoiiKolSPcnEHAfv7r45KADU6AFi2bJk88sgjcuSRRxqfX3XVVfL666/Liy++KO+//758//33MnLkSJ+9KIqiKMp+Uh6hf3WQGpMAfvnlFxkzZow89thjcvvtt1d8vnXrVnn88cdlzpw5MnDgQBEReeKJJ6Rr167y8ccfS9++fff9R9qLK16+4X78OtXhnUtfQ7Wbta6RNAVALnsJDM6Np/qpQ93zlLdvNFzjSB/uAnomTVnwLKGKOe5c+JelxTP/DsYwzu0nEfgxuBKP0qadqTDvOWAPNesfiDxENhQzPZjqMC/MMszRsLTwAFoilfXjILRJqfVon6hZ82rADF4V1i/TyDZmO9D0Bml3jOWoaLIG1migFZ4vmWjaeAU5P58jkyuhzWn0J3BJ7KDb5HkoIbJRL04iXyeyUZfm+Szct/EeU1q9UQJZRASLYIfLPce5Gud+a/p6/N20pfRF2PhF0xe4nDaG+sQNaF3scTRfAJ/WPbRfXi/cNlmCTi4NbtZwKsvMczewzAnfC16ZG59M/nPsQbLhdeuZWcT/V2I9D5jyY51HoBwYaiwCMH78eBk6dKhkZZkv/IKCAtm9e7fxeZcuXaRdu3ayeDFX0N/Lrl27pKSkxPinKIqiKGGpbvg/EnMI/qDUSATghRdekBUrVsiyZfz3ikhRUZHEx8dLMBg0Pk9OTpaiIq7ntpe8vDy55ZZbauJQFUVRlLqMpgH6EvEBwIYNG+TKK6+U+fPnS0ICJ6nsH5MmTZLc3NwKu6SkRFJTU0U2ixt/gnghl5TlFCHEkyZTyB9gjO190/UvCmiOgqDaIEqjm26G1I8d47ZHUg4bh9swxMaHx6HvM41D/Bd5SYf40m1+TKmSa8k+HlLK0npuMp3jzzDtccPAuMj0NaFB4ftuUL3tZLqer5kmHu95lEbVyjSNO7VW7GCENUg+OjMZmAnGBN4TSURzIKjNGZnTr3fbzcxa1XHTbzPsP09025wJt5Fs7Os3kK8D1VZuDDYP1Tmsi1WFubLuFWTHwMalFHe+xzTlLWjzufGzibYthU3ElCWogLf3rzlMl3yDfJ3MVD85B+zeLA+cRTYIVQ0uMV3p9IClT3Pbi+j5Yg0L+uCx/zBdHaizHwFtVh06kz0c2o0pxfXPpG7gK4ulGk7J5RRT5Y9DxAcABQUFsmnTJjn6aLcwellZmSxcuFAefPBBeeedd6S0tFRCoZARBSguLpaUFM96qiIi0rBhQ2nYkPV3RVEURQmDpgH6EvEBwKBBg+TTT80pRTk5OdKlSxe57rrrJDU1VQ466CDJz8+XUaNGiYjI2rVrZf369dKvn2d5FUVRFEXZf3QA4EvEBwBNmzaVI444wviscePG0qJFi4rPL7roIsnNzZXmzZtLIBCQyy+/XPr161e1DABFURRFUfabWqkEeN9990lsbKyMGjVKdu3aJYMHD5aHHuJUsn1gs7j1JEFAZF2U9UycmcBaoic3BhXERaRRjzFNuQcOYhGl0Z15uGm/80VF80+0piun+oH07Tnen8h2QNeNWfmF6UynjSFnyLacroip1R5LS8cef7Fp9/jn664x+3XT2fZJ2jMIsLeOouMjVRh09L6kdfalxJC5v8Bu6Rf5mmF/SCXfeWQLlnhuS0V7F5qJbDuhf/BcjZPwpn5EPfRKc13hIa+6KuuPlCvHkjWWNQ6nqeO8Jn5GGKhU68l+jOlNH8Dhx9O8gxspJ7cNHMSztBtOGSzzaYt4s+jwHgcob7HMXOHXmDfB7w6exZQ23W236EfzAy4l+/xxYFxIe+pPNqzV3J8uWv+VtC08gDe+bHjaPmJueSb0lzM515NzJ7Hzk3DfhSZS3Q8TCr6g+Tg8BwDdmPq3XbzLWdcItTgJcObMmXLvvfdKUVGR9OzZU2bMmCG9e/MDs5fHHntMnn76aVn121ryGRkZcuedd/puHwkOyHLACxYskOnTp1fYCQkJMnPmTNmyZYts375dXnrpJV/9X1EURVH2m1qqBDh37lzJzc2VKVOmyIoVK6Rnz54yePBg2bRpU6XbL1iwQM455xx57733ZPHixZKamionn3yybNzIw9LIcUAGAIqiKIpSn5g2bZqMGzdOcnJypFu3bjJr1ixp1KiRzJ49u9Ltn3vuObnsssskPT1dunTpIn//+9+lvLxc8vPza+wYdQCgKIqi1F0iWAqYC9Lt2kXVYH+jtLRUCgoKjIJ3sbGxkpWV5VvwjtmxY4fs3r1bmjdvHn7j/SS6VwMsFJHE39ogcPJKm6xJBaDNmq8czR+AHkc5t7fTvT8ItMXrzHmQIp/TpYYc8nZUfrgznQBqoax98vKwqOUmcg3kdAoldXObLSz7ETHnIXBa8rtkn/pvtz2uOzlfuMC0T7kLjFzTd9ZfTbsQ2lzXlOYA4JyGcMvXos7Lt7/xYfRBL2jvJJX6LtPEuQdci2AbvANGPkbL044bZ9rT3TkA2VRW4TxKJ0cJmFfBDbc0L8LyMNKYM3K5EwahTenvPG/mQrhI3Uibf9g0BU+Va4Fyf8VX7HCS1Pl9gHMjuK/wHAB8TrrRe/xEsk+56THXmP6Y6RwxifZ8MrS5ckFbsqEEcTw9M5fTS+nyj8GgC2HMGBERwZrJ/Jt8hd1tDxcqErDTUqkVLn5JiVTyAq4BIpgFkJpqHvCUKVPk5ptv9my+efNmKSsrk+Rks1h7cnKyfP45X/fKue6666RNmzaearqRJLoHAIqiKIpiI4IDgA0bNkgg4P4JWVP1ae666y554YUXZMGCBRErqFcZOgBQFEVRlH0gEAgYAwA/WrZsKXFxcZ5l7m0F735n6tSpctddd8l//vMfz0q6kSa6BwAbxY3RQRyPQ50c8sUVAPuQz7P8mxS4TaoE/BFtiVHpIMV8L7mTAufXQnu46RpMkgD+bMjymyJmuc9eXBP5NgqqwuJ1x1LMN5PipJhSxhFfygo0jnEHRR2v5NTJf0NJ3Awqrdqe0l96QYyY8uoKfzFtPKaQ6fL0Bywj7Am2sfyGA35aXe8nut64Gp/5GjDTKkfeTM7RFC7GAyYJIK6naV8GneWNr0wfZ4EhPBmI/2DC715JUeYBa0z7WLBTWPu4g+wX3GZfklCOphUqr4E2L2zIQWd8B3D/5HNjSQDhvoLZcHw9PyD7NYion3W66TshK8/8YCbYh9OFkOPJTvBpi3iLIuP6lRzG32PZL/ts24ZMVyIfE3w3EWXIHSJyrtQ4tZAGGB8fLxkZGZKfny8jRozYu4vfJvRNmOCpIV7BPffcI3fccYe88847kpmZ6btdpIjuAYCiKIqi2KilSoC5ubmSnZ0tmZmZ0rt3b5k+fbps375dcnJyRERk7Nix0rZtW8nL2zv4u/vuu2Xy5MkyZ84cSUtLq1gcr0mTJtKkCS/tHhl0AKAoiqIoEWb06NHy448/yuTJk6WoqEjS09Nl3rx5FRMD169fL7Gxbuzt4YcfltLSUjnjDHMJMr+JhpFABwCKoihK3aUW1wKYMGGCb8h/wYIFhl1YWLh/P1INonsAsFYq8pV+Bl0yXFlTnIJxDDtZ8y1d6LYLTRenG2L5UZouIJc8QB/gD5PUM4xqAd8HP8SaP5d3nQPtXixSkqYq50CbZMdJV5s2atjzxQ7qpJwieAydQC+smHwblVJtTzcDddQvTVcaac1JIOyyIslzGLA8qac0NIv3+GW6yXzPQ9DmPonX6Ceq2duCJ5cUQptqp+6kNMCnoc3lnePJxrkPnO34HdlfQ5vLD7P2jVNYBpqVamXCEtM2lnx+xJxcFd/BVPavgukifD15OWNjP2TzHCG/UrUi3v7g6R8AP5s464dLazemSQy9cG4MnqiIyLmn0revFX/4ygQt21q0eo+P/6sIQfsQy2+ImD0R93uAVthxpPpzAJxIHMgfDy0EpCiKoij1kOiOACiKoiiKDV0O2BcdACiKoih1l1pcDfCPTnQPAD6SChEDJTXWeFnNwuKMKSyCfEs2itg0CuRyqZhPzEvObiItuTUKqa1MH9spcEy8RCrnMKO/kDptGidPo1xItSkCtMTv9TB/gHVSXuoWj4l1Ub43RoGB58h3EH0byyubE2U9O74U6uBSOrysIBt/hS/RsbymLpZ7LTRdIdp0X98ZHh2Oc+dB4H6XNH9eVgR1Z54fwlo4ZpffTZMA1tNFmwhtvn5cYBb9fPk20gd5/cBYTtn815l1INJi3ToQk0gGZ409BG2uNstzAPCacf/ke4ivAH7+aQVdo5guZ3T3akQf4MrSnuJy/FJaDe008iWTjQ85vwmpeEaV6gAELT4G99vYdyvlwBPdAwBFURRFsaESgC86AFAURVHqLjoA8CWqBwDbv3fTct6Bzzk9jzP7eqDB4Xde6g7j3bRUXLeFpo2ZaRxK5GwyQwKgcq7SyTTbQATQJjuw7Qktc1pgIbQ5fkk1kg8HSeDZOaavL0USMfTMx8sBSuOAOY+O0+FwZ+eR7zTTPBwyj2b8y/SdTV/FJCVOJ/uaLnBHrOGbZvra0HdtKTYYCPVUFufzhrTQZ8nFC4tid+W+YV1SJGSa7Shm/WeQHh4xXR7FwrJbj8SSAAc55ThyfkXLA16TU9FM6/aE4frrdPpuIf4I+TiXDzss6wP8XKRBm18s9NziapvSmXycfxyDJ9+fnPyCCEKbQ+qcBogV5Gxh/Mq+i9jC/OEkAHxB1ExFOys6B8AXTQNUFEVRlHpIVEcAFEVRFMWKSgC+6ABAURRFqbuUS/X/A6+jEkBUDwBGi3sCmF3E95p1aEN+Y9mLyvAaOh/p4iQ7GxmDnP42l+xkEE5bc81WmsSAbu6HfPhot6MKnaWU2rXNpy0icijNb4gBDbOUNH+WSbHMLcut3blqaKHb/Jnqz1IVWeOa/mmm6Tuc13w90W2mUZ5aPxKicRVfqjAsT5A96RO33Ziys1aZpuCqubY+GceyKPXBqbAjnqPAkjXul9VhXoUcM9HKNpu+ONKohxzltg/9xPRxaWC8FZwOy6mI2NVX033qPp02vh205qGU+DeUawyHoM2FjbnwcXef74l4rxqK+Xz1k8jGG0sXWFpatg23bG8Diy9k2ZaxpQGGA38n3PfwGPA3+fopB5qoHgAoiqIoihWdBOiLDgAURVGUuovOAfAlqgcAW8UNMWMImwdrnIKXC+02XMaMSIDIXas3TR/X58LAHYfUuVIZBiyb0jGEaFsMqYcLmmEI+ySKfHKgzrYKnqfKGUgWfG4cNsdnhSsXjqRjwjA03wquZIfHywHfHCpHOORzMCgWzueK6XIcfOW0O5R5mpLOw/0BrxOn22CYfAhHYgncb8i+qZFSOIB8lMVqVKeL4xvOOY2Xus3u1Am7U3VCozRkiHzcgTHCPtD/N/cShDYnlI4l25aaxppLA592ZbYNvpH43bQw26JdFQmA4XOzVQJkbNtW57ogu3zaSm0Q1QMARVEURbGiEoAvOgBQFEVR6i4qAfiihYAURVEUpR4S1RGAfuIunIVVbrmaL1MIbc7A40hPrMXHg0JU0FhSDaeb++2Hf5f3axvB2X5DxD6otc0J4BKzXHoZYcmXV5KzHQ9fb9sqg570PZi7cRBlYHFFZDzGcPeYdX6E7xvfK7/f5FK6fE9xv+FSXHFRuWnka3wofYB2N/KdQ/ZxuMFlpm9Ud9M2rlIh+bhHoL7dkXycM4ppduFWtrOVnLVp7LxftnFb3k9VtuXjw+O37YexJQJXti/E9voPN+/Ads3C7esAoxEAX6J6AKAoiqIoVnQOgC86AFAURVHqLloJ0BedA6AoiqIo9ZCojgBcI24ZWsz1v5C243xyTAvndGebmsVzC2x6LC/xymom/g5r6Jw7j/AKpLxQKMLVcXm+A8IlWhuRjasms+5MFYaNfP3qDJx5dNoQ2nzt+R5jCdpwur6tfC73D4TnSXxONv4un0sLaHcgH++3ENp8D/m+4fE25mqz/EOYd38q+TJa0wf4qniUfKzdD4b2EeTjiQj4JLCOHyQb9W2bJs12VeYLVEVTr0rpXN6W92sr72uD5xLYrgP7bDUDGP5ugsXH96Yq16kGKJPq/6mrcwAURVEUJcrQOQC+qASgKIqiKPWQqI4AtDpSJPBbvLQprtJG23HYFIuIziJfKtnxEHceQpUrWRLAUDIHPmeT3Riib7y63gjaFksBc7bW3zkeD/VdZy82XffQpphWx4VVjyd7NLTTG5q+kXRdUHrg4B+vw4b3ileO43TDoMXH30VJgFMw+bsYNmf56GKy4+F6l9KORtC2WCGXV0zEMrx8Dx3a7/9B+ynaDwdtsa98QemPh3P9ZJQE3iffq5tM+32wPXm2/zPNIKzjyLIDL6GZA+0m55NzMNltoc1lZENk40NlK/1bmY3wCn8o5oWTFn71aYf7zXD7xZA77ydo+R3e1hbyDydZ+P1GOLDg9wFaDVAlAF+iegCgKIqiKFZUAvBFJQBFURRFqYdoBEBRFEWpu6gE4Et0DwCOkIo8qEQQnmMpXGMra9uKfPFD6QOo05pC+W4shaJq1ot8jTPpAxDz41+ibUnyw+P3lJflEwCb5z7YStXy8eax9IlryZJezM8G2nx4j5CdAllhRVRnd4NpGglkIfLxvcAVad8gH18HVJbPIF98e/ogHXxUU7g550MCfI2MeSonmr4Ykl9zXnPbfC5fk43X7AnyXUL5pmlYT/ll0/c5La+M941Pk69nI5h70Ik2zvm3aXdH/13PmM5EWjsaZ3qspnkHH9GmOAnEVt9bxKxcXEo+lsKxs/D8hqPI7oqplNnk5NRJJEQ2zyjihxOpSjokg3fSlvbH21Zl6WDbnIQaQgcAvqgEoCiKoij1kOiOACiKoiiKDUeqP4nPicSB/PHQCICiKIpSdymL0L/9YObMmZKWliYJCQnSp08fWbp0qe+2q1evllGjRklaWprExMTI9OnT9+9Hq0B0RwAOEbc+LCZabzU343uHbl6KtRdLcyBip5KeyXo86rqeEr19yA66ze2k+XN2LKptnMPuqQ2c4jbpMniuAyp3nuMdSDbIjitpNF0s+04KXzSQN1OCtO1G2hYuRGu6EB0pLXwE/M6PtC3J28b15p9M4aIS/aBNkxRYCcVyCTwPxTgk/uuErn1zmAPApYm5/HAI2lwK2qMAw8mupPfSHbQpTndgzZ/PDRXqFPJ5SiufDO3Ew03fpvf8D+rvpusLmt+Apbi5xgWDdRZ4LglPCcDuwPNbuMjxseLWTujS717TOZY2HgLtdvxQDyIbJyrZioGL2Msc8+vftoQyYytdbLPjfNo1SJmIxERgH1Vk7ty5kpubK7NmzZI+ffrI9OnTZfDgwbJ27Vpp3ZpLbYvs2LFDOnToIGeeeaZcddVV1TzgfUMjAIqiKIqyD5SUlBj/du3iglQu06ZNk3HjxklOTo5069ZNZs2aJY0aNZLZs7ks3F569eol9957r5x99tnSsGHDSreJNDoAUBRFUeou5RH6JyKpqamSlJRU8S8vL6/SnywtLZWCggLJysqq+Cw2NlaysrJk8eLFlX6nNohuCSBNRBJ/a0NsMY5i3xxoCkGbK6D24mXlTnebRzxpuji8iSHAk3g5Pa4p+w+3actgEjGjT56gGcch+Xct30XbE5rlmsMQPeSKslxqF/fLI8yNFI5vi7HafqbPo8/gd+lH47gWMISEm24xXRzWRTfLAxmci9jJpy0iTZebti1907D5htPNiIH6yQNIhuJUzy+hzatK3kD2oRD25z7oqfYLsJQQJBu7Tg75mv0ffTD0OLf9y0LTR+/Wwgfc9n20G8rItF57DoSjhMXyG0d+bfvl6/I2tE+ld/44vue4PGQ73lN/svHNY1utUMRbRtiGbeVAtnGpSdvqin8AIigBbNiwQQIB96Xl95f65s2bpaysTJKTTQEqOTlZPv+chbvaI7oHAIqiKIpygAgEAsYAINrRAYCiKIpSd6mFtQBatmwpcXFxUlxsTpEuLi6WlBSOHdceOgdAURRFqbvUQhpgfHy8ZGRkSH5+fsVn5eXlkp+fL/36sdZZe0R3BOBgcfNyoCxnozXmZrZStR6tk/PAYF3ckeRaQfYxaFxNTl5nGH6YNfUQHxPgGbEFyYbBJU8PYDxlhRGeA7DMbbLuzMeE0xBIJvfMmzB0/zTydSEbdf7vyUf6Owr7toVMRUzdl+9pCc0nCeDcA1ra9pjnTRuT2LgPGolefKP4okEfvIwu/vEkv14EbZaZuZ+h7h8iH88XwUPirsEZrnhZWlxPztFkrwbdn3IPP6bribo/pzjyPBSkKu/ucIlp+IegLQ1YxLytnFUrw8k+E6/qNHJy6V98ilh/t5XsDafrI/xfA2+L+w23zDAeY9DyvbpFbm6uZGdnS2ZmpvTu3VumT58u27dvl5ycvTNjxo4dK23btq2YSFhaWiqfffZZRXvjxo2ycuVKadKkiRx2GC+kHhnq9h1QFEVR6jflUv1a/vshIYwePVp+/PFHmTx5shQVFUl6errMmzevYmLg+vXrJTbW/fPp+++/l6OOcheUmDp1qkydOlWOP/54WbBgQTVPoHJ0AKAoiqLUXcql+lkA+zmHYMKECTJhwoRKffyfelpamjjOga05rHMAFEVRFKUeEt0RgJbippzGux8Hw3wNB3OcPu4RF2FncReZrkmPm3YAl3W9gvbDwjkkaX9JLlbQ4i0+z5q5p7rNo8nFGmWCT1tEvPVTQfBsSi7W9VHNnES+uOn0weVH+v/ozvmm/Q60qfSrp/4BXCje1FaWl28T6+gBvN60wuuZLU37fVgWl1Y6lmPROIucvPorzh/pYbq60ySWJJizwOdSQjY+B1xiugXZmM/Ph5t4Mn0AcxY860zTLUXdfzbVOKApAEZdA87Xt634G07Xx+gw/0XEz0wQ2qzKnkR2NhYGoWWQ5fDp9AFeNH4aWX+HjuV5hfNdxzccvz04X9+2X37K8Q7Y5geIiBT6HI9/Fb2IEomlfOvocsDRPQBQFEVRFBs6APBFBwCKoihK3aUW5wD80YnuAUAz8cajpZJwNoEBKg4P/0RpXy2waiMt0BToTF/GPMGW7Uzfa+sN0wGpgcPM3NewdK0n3YnrvUK+1kCKZx5q6cRp/EGQbIgJczokp4H1x9DnTHIOPZM+wNDnO6aLT3aLxUexWgfONVx5V9uz7UlbxJB7W9IdXjF7090D3DZHvgceAsa4eLGy1u0BH1NJ2WdpU4yih0t/tKW02YK43F/bhegDlEkoPbPoZdO+HdqcpsjBbM9KmAAXZG3l0xbxrpKJzxdvy4uDQmVwOekoct5J9in4wriMnBx+x7O1hfxFRL6GNj0zJS+aNqTvet4VXGob/Vwvm4vfYdfnlyjLkvDMlMAzzJKUcuCJ7gGAoiiKothQCcAXHQAoiqIodReVAHzRNEBFURRFqYdEdwQgQVzBH0RM1m1ZO8S0INY6WSZrgZrv2aeazu6cRNjWba4msfMR08SVTlky42gT6rMeHZQFbtTf6XBPe9O0cVCb1pP2w+VocVvS9dM437AvLPHq0T55hgbsbAEp5XQJDf2d138lYRoPka8vD+Ytqwx7UjTT/wPGYyR+jjMXv228/ImK9oibaEfnoTHG9K18wjBLIR+Ol8HlrFXsDnylm5ONqX7ck7nS8qvQ5mk3Vy017Ri4F1+vM328JDFOEQg3ZwHhc+NnfjC0OW2xmGys/n08+dr1pg8wr3UEd/xrycaHikv2fmc5Kp4N8ZRp5oOQ/g/TxVMC8Lkoo+7Km74FbVpB2/OXIs5aCZGPX0n4s/jutRUijiiR+Ou9jkYAonsAoCiKoig2ykSkugX26ugAQCUARVEURamH1MgAYOPGjXLeeedJixYtJDExUXr06CHLl7vBPsdxZPLkyXLwwQdLYmKiZGVlyZdfcrBVURRFUapJeYT+1UEiLgH8/PPP0r9/fznxxBPl7bffllatWsmXX34pzZo1q9jmnnvukQceeECeeuopad++vdx0000yePBg+eyzzyQhIVwWPx397wI55ORzJVXOcUbNl3XHEP+GkV8+NMwBgcJ5q+l5inTSf1p+09bXPFeHTw61cFpy9LJVtC1KlFTWWALjTDsD5jdk8EGwcoqKLCcbTzbNW/7ntqearo0km6J++LnpMq6niKmict46z7HAUTBr31PIPh6k2tzLydnB1O5lEAj/b7HiineSTpQkX9T9eflq1luxu3Yg321k94VU9JV0CHT3DU34ffLlkB0Huv/D5OPllvH4uWQvd+020GbNn1dYx+VXAqTjd+ecdkz+55M5aRh9gPXAuRAIE4L21+R71DT/B8siv02bfkQ2XMT5NJWA60JgdQGe38Lp+yFoh5uPgb2Xnye+j7gtzkM5YHMAVALwJeIDgLvvvltSU1PliSfcl2H79u0r2o7jyPTp0+XGG2+U4cP3/g/19NNPS3Jysrzyyity9tlne/a5a9cu2bXLrRtdUqIlJBRFURSlOkRcAnjttdckMzNTzjzzTGndurUcddRR8thjj1X4161bJ0VFRZKVlVXxWVJSkvTp00cWL15c2S4lLy9PkpKSKv6lpqZWup2iKIqiGJRF6F8dJOIRgG+++UYefvhhyc3NlRtuuEGWLVsmV1xxhcTHx0t2drYUFe0NSiUnmyu/JScnV/iYSZMmSW5uboVdUlKydxDwlbgxT1h1zFhpTUS6kI3hY05p4rCpZKHBoW4K673oRiY2UnrOS/RNTPoJ17cwrMtldz0nh/k5fybf6WS3PAaMIeTMJBsDdknk41W9IGns5+tNF+VkvQVpdXyNOByP14nLuXIaG4Y7WTbhle6wJ9KCdMKqCd63Yjrtu0fwQdwPBksAeAaUHEebYgSYw7gcmcQQKy9I2ZfvP5RzTScNpTmdG6ZSchj/T2Tj9eYUTD5+DPNzWJ8D7AOhPZh8bflPGXxuOaw/muyYSy175r6OsLzF4ggko86j2rqsWWF6KXVmh+SZN6DNIX9OC8XrHe49g32JL6dtRUUuZM3llFFpxD/dfhWzUnGNoWmAvkR8AFBeXi6ZmZly5517C2MfddRRsmrVKpk1a5ZkZ2eH+XblNGzYUBo25GrfiqIoihKGcqn+HIDqfv8PSsQlgIMPPli6detmfNa1a1dZv37vYjgpKXvH+cXFZjmO4uLiCp+iKIqiKDVLxAcA/fv3l7VrzdDYF198IYceurdEXfv27SUlJUXy8/Mr/CUlJbJkyRLp14/n8iqKoihKNdA0QF8iLgFcddVVcswxx8idd94pZ511lixdulQeffRRefTRvWkvMTExMnHiRLn99tulU6dOFWmAbdq0kREjRlTtx74WkcTf2qe5H6eMNTc77WnTxhQXrnjbkUviJuKCpbyEZ6FpznWbnP7E2rJNj2PVEVaVles4N4qWKJZ0TFuihMiWh5q2cfvTyMeSC84B2Eg+qjG84Bm3TXl08xeaNur+rKiGyMa0JU5/Y/AycQncS8keBxf8KppcwBlZmFL4BvkSSKu9JRuyVZ7ixCpcaJYSokg4Rx2XS0Fzd8DuO4xvN4v1UIf3C9L8OUUM3398DFzBA/XicCliadCeQL7RfOMwsMhVeA8jG6X8w48jJ18IfK4XkY+f3AK3udpc4huffxExdX3q3F9TrV2o9hw21RO7KD+JfN8QWjHb81ygn/OsuDQwwn2Qb8UjsIEDnadERO6w7DdilEn1FwOqoxJAxAcAvXr1kpdfflkmTZokt956q7Rv316mT58uY8a49c6vvfZa2b59u1x88cUSCoVkwIABMm/evKrVAFAURVEUZb+pkbUATjvtNDnttNN8/TExMXLrrbfKrbfe6ruNoiiKolQbjQD4oosBKYqiKHWXctEBgA/RPQCYJa6oiAkEN5ubXUlibRwIWiyT8sqsIn2hvZl8pBe+5jY5V5o1NdRCG5FvINkz0GBRetAk+gDr9PKcBb7deBSsNPLypahhU7ax84xpoyhMid+c59ED2kHLL4qY+jsvmMrbYm5yN/JxeQQUzq+jOQqct46Z3vybXLHVmODwFM9waAltsyYG530j4Zb4NVYZvpqcXD/rHrfJlaC5fDLOFg6nJYegHa6kLD5/o9uQk4OI6OcCHtY1wD8m50rTXAhPJ91/zyQGfFiXk49y+z+AiRNcIYBXs8ZHhvV2zxLgFvhwsb9wifS/ko1VQT4gH69mjTU6+B5z7Yzz4ASwbElVln9WaoboHgAoiqIoig2VAHzRAYCiKIpSd9EBgC9RPQD46hM3yH04ppt91M7ccKaZrjMBlzpra27qiYsZ4VkKWv3yomG+BKEuLmPLoVsMUQ8g3yOcB5gP7QwO+Z8r/lhiySJipp8lk4+L7WI61L9MF1dwxhQtWvyvxzdkY9yc68ZyTlOh2xxGp8YhVQxZ87WP4Zg16BBci2oSlXTG0Dina7Fc8AUc4+Ge1eAwYc+s5xxzyHzDDsKKb/wbXN33BCy1zIU37zPNt9a47XdpU/4djLizRMVKwwPQ/g/5WGgybjHrDnPIxlg4yxlcIxv39TmV4V1C9nNu82d6cJtxPWIsVfKh6bqdcsUx7M+lqvnpwjcLh/xtZXnZxyl52NV7kO9YllzgJgdJsWIZDe8jn0uI7E+gjYsXHrDVABVfonoAoCiKoihWHKmzf8FXFx0AKIqiKHWWSCzmV0cXA9QBgKIoilJ30QGAP1E9AHhRXH138GL3877zqUTn2ST6hiDRhteGjbmdPsBUOpoDQLloKB9ylhIr7KjcZ/MkgA94gU2sZcyiJCtpeEs5N8qW2sdFRaeZ5gLIj2LBuJhsXFOZc494lWHUcnktU8qdQk2dU4j4AUWbE/BWU55Vd8zB+pvp60gTCLLhVjxC++UpC99C+3DPcsAXQ5sE7F5G4qcMAOGUL/UNrOPeC20W3KlULWatceoZa8vYk7hk7+FdTftOmFtAUz488ybwlo8m8TuVbLwVWVSh9wTOTcRpFTy35DnTfAgeXNbQx/F3oT8X0nwBTp3DbsX9lUsiY3/la8+vqEOgzfNbeL+Ytcgpzw4dfwy8h9Lpxl1M9wLnBHDKM08JwmPC7loq3sLLyoEl4osBKYqiKMofhdpcC2jmzJmSlpYmCQkJ0qdPH1m6dKl1+xdffFG6dOkiCQkJ0qNHD3nrrbf285f3DR0AKIqiKHWWsgj9qypz586V3NxcmTJliqxYsUJ69uwpgwcPlk2bNlW6/UcffSTnnHOOXHTRRfLJJ5/IiBEjZMSIEbJqFZdWihw6AFAURVGUCDNt2jQZN26c5OTkSLdu3WTWrFnSqFEjmT17dqXb33///XLKKafINddcI127dpXbbrtNjj76aHnwwQdr7Bijeg5Aobj59JhO3peX5TyJlLK/oMbOovTxZKN6963pomRplPI51Xww2S2mg3HlqeS9kmxU72yaP9usPLINtWrn5ZkuWsb3J4hccY0DHh2jztea6xwH6buws1dpU1bNUWtknZFDdDiy5eO9h+xp/3bbLUaTk+oYjIQf3v1v08flXg092fmf6YzB+0TqLJWjvupl8Wc62e2gCMNDpjq7YI25KdZOoMx4D0Foc064pJtmABLDh9PFZ0k9BG2urMvzBXC6CD+1ngcOOwDVaP6Ujuk9y27Oo4kdiXCQ/Jr5juyq/NWIUxh4WkcW2VdAO0g+rt+Aczu4vEhMb/oAX1Inmq6MHy02vRY9DyeeHHS0knKRZ/i7NUB1Qvi4DxGRkhKzqHvDhg2lYUNeOl2ktLRUCgoKZNIkt25LbGysZGVlyeLFiz3bi4gsXrxYcnNzjc8GDx4sr7zySrWO3YZGABRFUZQ6SyQlgNTUVElKSqr4l5dHfzj9xubNm6WsrEySk83p38nJyVJUxCOkvRQVFVVp+0gQ1REARVEURTlQbNiwQQKBQIVd2V//0USdGQAYWSocB/Okw2GMlWsB7yIbU+dCpotWK0ufDm2Ot11wOH3wELR5Y17Fz1Y0k317LL47TfMWN7b89c2mi8PkmNLEpUo5dQpTIFMohBpHNkYSuVguV4bFO8Oh2lZko9jB42deSRAr5N5+HTlfJ3ue2xx9uek6faZpx+NtjeHUzhC06X6PMoPAMTMhZs3x4RH0wRoI+z9huqiqsadkK8LpZHgqcS3JyamecOMGU7jdJuuEkyEwpN6LndwhYNXMZVSylxXVz6HNfZsXB20FIWvuR7ziJ0a+uXIxpwmjn0P+wzhOi7WYjzJdAdJnArhjPohOZMegFOkRWQj8S5XfM5xujIWQC91myW6RJFqqtQYol+rn8f8uAQQCAWMA4EfLli0lLi5OiovNxN3i4mJJ4Xrjv5GSklKl7SOBSgCKoihKnaU20gDj4+MlIyND8vPdhVzKy8slPz9f+vXrV+l3+vXrZ2wvIjJ//nzf7SNBnYkAKIqiKMofhdzcXMnOzpbMzEzp3bu3TJ8+XbZv3y45OTkiIjJ27Fhp27ZtxTyCK6+8Uo4//nj5v//7Pxk6dKi88MILsnz5cnn00Udr7Bh1AKAoiqLUWWqrFPDo0aPlxx9/lMmTJ0tRUZGkp6fLvHnzKib6rV+/XmJj3SD8McccI3PmzJEbb7xRbrjhBunUqZO88sorcsQRrLFFjhjHcaJunaSSkhJJSkqSm8QthZkO/iH/R1/InUof4AXlMRDr8ahn8aKeVI/UKMzZk3xdyMZtN5PPpvlXZYnfG0zX6PcM8ykQhZ+lvXAJV9RnWR9mbMuVshKOuimH2XhuAer8w8l3LdmF0L7Y4hMxpdFLyXfuYfQBrgCdTnlUL1KVL5w4cQPPAcFCwvx6+Zhs/AsgaLp+oPTC6+FwnjZd/Fjg3AhOEKVfkfOgfSOnj91P9lNuc/ss03UrbYpzAvjpYvA+vUS+FCpHvAhSHvm8eeloz5QhC/jUhvtPAecpvEC+eEqzMx4a7nOcmYxzAFpzvxpFNmr1PLeICwnju5CLl/PED3wP8Ts0aNmvO9OnpGS7JCUNl61bt+6Trl5Vfv9/Yo1451xUlW0i0lWkxo61ttAIgKIoilJniWQdgLqGTgJUFEVRlHqIRgAURVGUOosuB+xPVA8AThZX1UrHNTLP4y1puVVD+2JNnW1cU5WVUp6cgZm/4ZbiRQW2Knn/4bYFNfwMU/N/8F/mliDVhi2ta3sAWNcv82mLePOs0WbNP43sHGhnc+zqGNM8HJK7B9AUi0L6Kpan5bkQzWmaxylYJ+Ax0vzPzDFtWQ3tDPGH7yGLwFif+E3TRbn+G0H3f572who7Xnue18F1FYw5Fx3JybUJ4Mu2MtEi5pPJ804+JxtLCowjXyaVOcYcfdb8uQ8itmV6w32X+y+WJojnTC5eAhwvBOfnc40DfOB++IKcVJkuBG0uccKvM9tEH36FNjgJDJ7fxAsYQ21oYznznXIgUAnAH5UAFEVRFKUeEtURAEVRFEWxEclKgHWNqB4AHHmsSOD3M5gEjtbX0JacwoIhV06F4bgY+vlycegWw/Mc8rel3HAaILPd4jMrR8kstwbqIgr5cxnWELQ54sdHi0E9Dg9zhBLDpMvIxylXGDblSqWcvncu3kauBduBbNA0WALg4qNYcngt+SjCLimwAmD6ReScz1d4GrS5r+ArqTH5eFu4+j/TGnlvm+Zb0C6kvXDPRnjRRr6chs3lz9tRWlQLN7kzQDHG0fQmxfvPqVojyP4vtFeS70uy8Vw58s2Jvnhu/BywjTIKrzLJoKSxjBaA60l2PKYx8sPH//vg64BKa3v+p8MHjvSLUrLxOeDzTuFX6LHz3fZB800fX5hCaOOL5AAJ6zoHwB+VABRFURSlHhLVEQBFURRFsaGTAP3RAYCiKIpSZ1EJwJ/oHgDcKZAHeA44BoX5YhDarL+zAIdaPqe3sM6P6S7h1onG74brXqgJrzJdP1PaDwjcnLLEKVj4q3zWvBjoMGj3YD2QxXsQPwdvNV2kYBvqN61kKmeQbej+XGmZJxeACMyjd74OqOvykrT/JfthaF/7H9PX8RkSP89Ho7PpM36VFVcuwwoTGr4lV6Fp4uVm7dtGkGyeA4DXpTFvzGmLx8Miv5SSm8KdEieUUAXkTpRlifMz+D7ZZsnw/AaePjIHOwSt6evQI4794WHTJe+TjXMALiMfJwmnQhojvzm2kI3zG/hyck+yldq2vXV4nk82vSaHwHSXz2keAqfS4jP/63du25ZSqRwYonsAoCiKoigWNALgjw4AFEVRlDqLzgHwRwcAiqIoSp1FIwD+RPcA4MiOIoHfVa+h4GAVjTOg8bTD1QFAbHn/7OdLa1vGl5U7Vm9xv0HT1exI057kLg97AlXoPIE0a6MQACdhs2R9LLRZgKX9LvvEbbN+ybof5oHzHIB4vrw414DnAKwgG24j69mnWY6Jy8+ytox65nLydXyVPjgf91aVUsDcJ4Nuk689bcrlaG2gNs7TOHqQbfRQvqByqGmmF7rtP1EP+M40jZq9NL2Fy1P7Hk8Y+Gp6wEd+iOmKIbE+HfrZLR+avttpt1iI+0fy8blhueJw5Yht/xnxX6qY5x1uvoBtThCXaX4CdP8Q+TaQ7bfccl39qzqaiO4BgKIoiqJYcKT6gw0nEgfyB0QHAIqiKEqdRSUAf6J8ANBeKg968u3iECuG4/kSsI3BMJYH2MbQPZd3DRuItBwDHj8nEF1rmv3h3PpvpG35ePE6LCHfStMsgDE01cf96d+m/Qi0OfTJBKHN6VmerEuMU7NewDFK2LYLLVfYhZbF6wG37XLaDR8/hjP56npS9Iwt+J7ivWCtg/sK3H9eepG6P4aWw720UPWhxRTlRLIDGOU/nvfEHwTdZmPqLLQ0XwksWXgr7YVTRlH94KeA+w6WBuaQNHeV2aBSXPgyOenxkivgGI42XZMfMG18A7xlujxhcdsKmja5I1xqn60P2H6H3xRc3ddWBtm231a0nU3mUWqeKB8AKIqiKIo/GgHwRwcAiqIoSp1F0wD90cWAFEVRFKUeEuURgLbiiqId4XM+rSDZtlQ/pirzBWy/wUEkXpQU4TkLX0Obc824ACn+LpVoZa3ZgVqrH9CmvGYu6PwrqT4ul0TF2QS8WimPODE1iTMRPTl4+OUjqaDv96RK4kHwjqnSbnNIWwwX6kO/Z36Dpx4t1kxmtRNVbO4rlrRArtFKf5rgVQiX9oU295Rm7ekDnCRwJE/A6EP2P9wm9aOdj5v2HdC2LVctYh4v/+KDvekDmEAwmjJwOX0TZyl8RhftljtMuzEuAZ1j+lofYtp33+W2j6ZsyLl0DDhngf/a5FuO3Zn7a4hsLA3NjwG/gXBRZ+7KPG8C/Xy83M8w3RRLjJeKJ/OzRlAJwJ8oHwAoiqIoij86APBHJQBFURRFqYdoBEBRFEWps+gkQH+ifABwhIgk/tZGvZU1VFupVfbZygaHAzX2qnyPf3OaaT4JycmP0qYkaO4EDTMxiTKpO9F3Mf7ziemaS1rou9Bea7o86nZVZlhgSj7PJbiCliDtbuiolHt+/POmjQfMQiMlhqOb5yww+CLwLGfqeUtgSeog+Wx1ABhQa5uTK2SaePxcNZixlg3mUgTG2sw3kJMU400fuW0Su+8yTZkPbS4bzWAJBM98ERaeR7rNIU+bLlt3YO2b8/WvhjkMh3NcmGsGQBcdTXMJTn3NtLGEwJemy1P1eiC0uYQzL1+Ns4e4MgnXfsBSDyHyUYkDY2oHz4Xh5wLtr6DNb96aolyqH8LXAYCiKIqiRBkaAfBH5wAoiqIoSj0kyiMAP4kbq8R4cUvaziYJhFuJzUZ1yvsWwCGcbbpONs1XYFmxf9JeeHE1pM1W0z6N5AIMJT5iujwpWcXQDhdaDkKbV5XjECumrXHpV168sHshWpS4lniqaXd4222zZkExy2XQ5nOzhQ6D/IFnOI03gIUR7Dshi49+KbGd6Wq63jAPKpZ9Bv+q4VQzTxnmERiAp9X/hDoadNKXSMZ51zStpaI5lIz3gtPSdi427cSxbjubyg+vJLkLw9l0JvKR+DPpSdNO4y//H7RfNVMnAyvN3n5jHhhUN3gTpTHivYoh9SiTtsU7xZIPl1OOgddmM/KdQ/cR+w6/K7hENt43lFsO1F/VmgXgT5QPABRFURTFHx0A+KMSgKIoiqLUQzQCoCiKotRZdBKgP1E+AMAEj6qU5UWN1bb0LhPuctnmFqw2zR/Oc9v9TNcUWlYWJUFOlbIt/1lIPpJC5UHLfkNkY5YV/yZri5ht+BJplKxn3gdtTmHyLDlqCMa0I0k3zVYwB4APmHL98HfCPehY1rQzO1k3l7aWPVXl0cO+REVwaQ4A9mzbMrIi5rl0YefR/MFkaPMcG+rbMHmDC1dzyigeI6epBckuhDZP66BMP7kEhX3Kqr3vQtMuX+e2wy3bi3MC7iHfRbSUcAaK4ffSDJfjxpn2XFTkZxqu1kvp6cQHheZ8BCjFNYDHwDmu/FzgNI8S03U4bToW5gRwZirPm8DHFvvnbvG+o2qCP7oEsGXLFrn88svl9ddfl9jYWBk1apTcf//90qSJf3rwo48+KnPmzJEVK1bItm3b5Oeff5ZgMFjl31YJQFEURVFqiTFjxsjq1atl/vz58sYbb8jChQvl4osvtn5nx44dcsopp8gNN3BNjqoR5REARVEURfEnkhGAkhIzNNKwYUNp2LCh9wv7yJo1a2TevHmybNkyyczcu1TSjBkzZMiQITJ16lRp06ZNpd+bOHGiiIgsWLBgv39bRCMAiqIoSh3GEXcewP7+c37bV2pqqiQlJVX8y8vD/M2qs3jxYgkGgxX/+YuIZGVlSWxsrCxZssTyzcgQ5RGAw8RVMlFdYn3YVu433CVAf1VKDJNK6Zxn2hPd5kuk+S8zTUO6Y72d9TfUVFlvZZ0f5UMeITciGxVKPgbO5Ubd1KFb0fos086DhO75VKfAU2rXSFZnwd1Sk4EO0CGhH2sThBsR469wGVZvfVpMDOd+ZuuDtnkoGaYZfNEw8b7xPbXNb2Ctu4WnTjDWXeBrTS8qkLt5HgcfA/bf0eTLtthfkY9z0XPedNvxM8hJJXuvvtRth2hT1rPxOvEK2ny9r4LVtrtwRPfqx0x7XGswJpu+3jTnojcm5dPLQyhhXxb5+34moT8EbZ42RVMYOsI8jyvp9l9JtYzL4B2Apb+3iXcZ6j86GzZskEDAXTi5On/9i4gUFRVJ69atjc8aNGggzZs3l6IifoNHHo0AKIqiKHWWsgj9ExEJBALGP78BwPXXXy8xMTHWf59//nmNnfO+EuURAEVRFEXxpzbSAK+++mq54IILrNt06NBBUlJSZNOmTcbne/bskS1btkhKCtdqjDxRPgBoI27y0CHwOYfBmKqsQ1WVbTHn5nHTxTl4EM4+kVyUyWOEUZPIxwv8YcHR8WH2i/A4ltPCJkGbVyc73/I715Bv8j9MO3Cz2z6Jlyfj2sCGipJGTopZon5AEgBLIeU+7SrjkQBQ/+B+hI8ep/vY+twhpkm/ia+McGmAKC0Vki/No7/gMYbsOwY3h8VZPMB+dh13QrN6rqRACV9e0a+QbFwbMpvrZ//VNNtByuClb5o+vgwoz9lSBEVMqSyHFMGMibTxP+E/gePJealpSjN8qgaTk5MpMeeRCnE3o7vRDN8u9Dx1pT45CqWHQvpN82Tj4H3cGnwJJeUiSbZi5pGhNtIAW7VqJa1aeQpse+jXr5+EQiEpKCiQjIy98t67774r5eXl0qdPnzDfrj4qASiKoihKLdC1a1c55ZRTZNy4cbJ06VJZtGiRTJgwQc4+++yKDICNGzdKly5dZOlSd1JJUVGRrFy5Ur76au9smE8//VRWrlwpW7aEW1TbRAcAiqIoSp0lknMAaoLnnntOunTpIoMGDZIhQ4bIgAED5NFHH63w7969W9auXSs7drgxu1mzZslRRx0l48btLSh13HHHyVFHHSWvvfZalX474gOAsrIyuemmm6R9+/aSmJgoHTt2lNtuu00cx6nYxnEcmTx5shx88MGSmJgoWVlZ8uWXX1r2qiiKoihVp7opgJGYQ2CjefPmMmfOHNm2bZts3bpVZs+ebVQBTEtLE8dx5IQTTqj47OabbxbHcTz/ws07YCI+B+Duu++Whx9+WJ566inp3r27LF++XHJyciQpKUmuuOIKERG555575IEHHpCnnnpK2rdvLzfddJMMHjxYPvvsM0lIqMoSuyXi6qU2TTVEdlXKBtv4iexb3eYXC00Xl+GEVKRmw03XlXNpWyz9yelZPDTFCQMkr/FStzj6Y7WKDkkG9vb5DREJ0u/gUO5t0+XRVP92s9tufRE5/0L2CExyC5KTZtSiNk4/Wp3RPF5+j0zu2XEDnzbbnLZq65/Ut2m95SPgDwB+kngpZtSwPWWXQ/wB6sO8Z1oeGK5942Jfl4hQYWOWs1lkB/hS80q8mKKX/Rw5WVO/ym32pXrUOywSNafr8uFiaiK/Vc6l90HTf7vtI/5t+hK57vEV97rtTveavgAXjhkK7UHks/VJ9gXJxrkolJrqAc8e5w7sEJGxotQeER8AfPTRRzJ8+HAZOnRvx0tLS5Pnn3++Qr9wHEemT58uN954owwfvve/maefflqSk5PllVdekbPPPtuzz127dsmuXbsqbK7GpCiKoiiV8UdfC6A2ibgEcMwxx0h+fr588cUXIiLy3//+Vz788EM59dRTRURk3bp1UlRUJFlZWRXfSUpKkj59+sjixYsr3WdeXp5RfSk11VOCRVEURVE8/L5kXHX+6WqA+8j1118vJSUl0qVLF4mLi5OysjK54447ZMyYMSIiFdWNkpOTje8lJyf7Vj6aNGmS5ObmVtglJSU6CFAURVGUahDxAcA//vEPee6552TOnDnSvXt3WblypUycOFHatGkj2dlc3HPf8F9wYau4yjaKdUHazjavgPVX3hb9tOzpZirvezW0X7L8pIhIL2ifSj7Oh0+DdiH5PjTNuXAZNpou2UU2qrrDyDeB5xocD+03TBeHkXC0zLrz+2Tj3INrqXRCfBptzEvh7isk1nOiDN5x27mw7amW6/kzwfZ44a9WpdYE6e10SY6FNpdzDpGNYU3PHABPbBDvpC2bX4ylhA+lmr1cx8Io/3s8OR81TZTNucYB9zM8n+2k6zfmidLnQsGBHLP4xIDbzE1RfGRdfwXZeLzvke8bsvGR57k6/fnvIlz7eL7pKl1j3sn4hlByuBeVHz6H9nsZFl6gcsSepa1xLgr3c+4fVZlXFXlqoxBQtBDxAcA111wj119/fYWW36NHD/n2228lLy9PsrOzK6obFRcXy8EHH1zxveLiYklPT4/04SiKoij1GJ0D4E/E5wDs2LFDYmPN3cbFxUl5+d4xVPv27SUlJUXy8/Mr/CUlJbJkyRLp169fpA9HURRFUZRKiHgEYNiwYXLHHXdIu3btpHv37vLJJ5/ItGnT5MIL95akjImJkYkTJ8rtt98unTp1qkgDbNOmjYwYMaKKv/atuOGm7vC5reyqiBmisoX8RUQKoJ1rum42zZUQmqPopWfFvEMhJngexQe51G4MRNvK6PBm07aY8cRpSRyyxlkUp5NPziAbF9+j+GsabYpTOXnkbLP5GrX1TAnBe0WhcA5wY0VUOl6+DrgnTu3i0B8eb/jAJspWVVkNkPukRSKwTIUJVwoYCfEHnhQ8PEY+PlqtDuLZWVT6mXebgvIX31KqXc2Shg3sOoXk6/4EfXBuZ7c92ZQA4unZ7AWSG19flpa+hjaH9VkCwD55GvmEUilx8cUHKf/xY9p0B+h+nUkuzCa7y5dw7vddRnu6guy+0OaVOW2SFm5blZTv/UclAH8iPgCYMWOG3HTTTXLZZZfJpk2bpE2bNnLJJZfI5MmupnTttdfK9u3b5eKLL5ZQKCQDBgyQefPmVbEGgKIoiqLYUQnAn4gPAJo2bSrTp0+X6dOn+24TExMjt956q9x6662+2yiKoihKddEBgD+6FoCiKIqi1EOifDngZBFJrORz1qDY/tWnLcJLWRoq+9dUgbC5aXaANo+sqFCtrIT2f8jHxTzjQPfndCfWHVFHZ607SDZmXaWTjOtJTcTLROLn0aZpZAmyjM8rXA+3+DwHbGjNLBfRt7HmLNU5bkM5b1nQfsd0ea4v/iXgKQVcpeE0Pnq2/hlm2/R4w2wR66rNCSRcsmaNtudcgvwBzmfgUtvEWW5zxFTy8UQP7GeUp+pQudwB0Cb52jO3AG1OPe3OS3P//LLbbnaX6TvjesNsCzL5r9Q5eB7Ns9Dm54D/osQCuc+Try8tiz0FbtZbtC1fh6Y+7cqOwbjnpXRy8Xea9i88q8EC9sPAkbYjqBF0DoA/UT4AUBRFURR/fq8EWN191EVUAlAURVGUeohGABRFUZQ6i04C9CfKBwCfi8jvGmh/+HwzbcfL9rpFiOSLh03X06YpqBd2Ix8tX4orcU6jJUevo6+iTs6yKB8tarXhOiKGdILkyyL7KjQ4738A2SikkkTN2iIeA+vOrFHiNIo4rhPrqbWLPxwiH325qb8rkeJeLSC+19h0eeYA2I7Am3iPcxZsS6+Gq1thqxmQbpqtllY0g5w/TmBf8oQC+cSNMtg8W4M6xMFQWPqq100fdwB8hmgCRoy5XIhkw/m8bLqEqv0aR7SEfOfQdWmGv3v2IaaTlw6GbVvRWtdcyRi7L88t4ZlGOAeDtz2BJmjg+4LnBAXJxtWiebXt7mPogz9Dm+uI30Wa/6twDHQ9vzRNo5zDDvkftA8MOgfAH5UAFEVRFKUeEuURAEVRFEXxRyUAf6J7ALDhETfU2w5L9lJN0fWUWHO52yyglcFooTsjNNeDtj2DMobirnHb8bSy3X1TTPszWLVvFf0mh5swshwvdjCk04l8vL5XADWBkeRsT7/0HIQASbN4l75qC+1xOtTN0L6aypp2FwaDrEHyUY1klAAohFpCF/gDaHNIlcFz43Ku3i8Hoc2pfZhKF670Lz6mXKqa9A34yaYUmvUoKpaj89woI8g+lHx8/OluczRJAHyNUE1gLYzeurb1CG2EyOb+2QxrV59NJx4/ybRPz6toBiidsOd3pg0Fhj3ywM1k40qCfIm+Jhu7L6kkchLZKPOlTCQnLWZq1BGmd9szy00bJRi+bazy4OOHQsKBCqurBOCPSgCKoiiKUg+J7giAoiiKolhQCcAfHQAoiqIodRYdAPgT3QOA58UVAw/6yP089JG5HQn7b4Ho/qzp8pTsRf2Ky26SuilX3uu2e51Mzh6m2Yr0QoTT6lBL5AQs1tswhZC3bcbrqeK6oydRXWPWmiG3ZxmliJE8aGiYrDHxg4TXmzX17h5hOuh/fKz0okhMoi9VAjbSxDyr4BJ4/B6Z3FNPd1+xLZ8qYk8ZTDNNmBLA+jBfezxcmn5RyQQHXGj6BsvxiRgLWrc9n3yUsLfJTQvziMkEzkrhS839DDVbvqeeZaeNNaA5CY8SeLPdOQA8eSeG8t8S4Ye704slmTOVgXB6M87l4Mzku/lCYGof1xinDOidMG/pZtp0Ptn4CrD1KxHzfHBbRw4MjlRfwz9Qx3qg0TkAiqIoilIPie4IgKIoiqJYUAnAHx0AKIqiKHUWHQD4E90DgB4i8ruujcJ5IW1HdTe7gXbHS9nyjcYylqwdsvadC+2j/236WBf7xPKbNF1AZkA7pSc5KUn4dZDGPXp2B7KN8r+jTN/qx0wbdNKXaDesqSNct4DnNwSh7ZH8eVqCIezzkrQWHZ0ucJDcNj2Tc+fxCFrw73hOwPOBz55C5OPHMtwcAQDETs/yygTO1eDKrxuphkDbfFgKe5CtpoGIeWW4ZkBHOgiYA+ApP2yCemVVXsghsnk14HR8P+whtbsBLYOLdQH+lGf6ePoAPOSrSfPnUsC2nmLTaT3POL88cO7BXNP1EE38gOq+nikgXDsBj4nfbXwuWMcEn6e6mlsfTUT3AEBRFEVRLGghIH90AKAoiqLUWVQC8Ce6BwCnHiIS+D0YhaFGOq0LCgwzbZGbJvjXW81Nf6bQPS6C90/6eV5lzCYXcFgMQ2oc4mtIdgqm7x1FTlrFbxiG+XhZruFkt8WlziitjsKZP61x27zyWkj8aUX26WRjZhJHLyWVP+jIHwAUJsdYI+kOHmUB4HAmL/CHKsqp/OV02zHZyv3aVv/j7/K2Qd9N+fLxSochaHNW6qtkX4b9YVCIvHQMxjFyMiKdGx4UX2w64AQIo3sWXqwCnu/iA8ix+e58vaFAdRq5KG5e+h+3fZ99U+MO831jGaoQ2p+RL5seTkwh/oq2XUm2LQvTlmbJPZufeRSE8FW2W0T+I0ptEt0DAEVRFEWxoBKAP1oHQFEURamzlEXoX02xZcsWGTNmjAQCAQkGg3LRRRfJL79woTNz+8svv1w6d+4siYmJ0q5dO7niiitk61ZPOa+w6ABAURRFUWqJMWPGyOrVq2X+/PnyxhtvyMKFC+Xiiy/23f7777+X77//XqZOnSqrVq2SJ598UubNmycXXXRRlX87uiWA075zz6AT1LWkFVI9uX649O1DpqvZ+6Y9Aur/jnjZ9J1EcSGU3FnF5RGkbeTF+uC7MGFg4HvkPIbsmy07msC/hPMmaLHgp03zEctubaNjWphZrutMH+CF4JxBvm/WRWDpimN+FImoK0zTSIdjvZX14i7Q9sxZOI4/sD1e+5oiyNtySiB1dpDcg7Sl7dxKycfTR4y8VU8ZZj5PTAsMkzKIgjF3pJBp4hwbPl5biDZAdh/ewHYMnnOFOQA8mYQeDFwmm6cW8POfBu2/kY/TObGCL8/VoNeXMU+JT43vDE654KrBVUmHzST7WGjjHIBt4p3SVBOUS/X/gq8pCWDNmjUyb948WbZsmWRm7r1yM2bMkCFDhsjUqVOlTRu+EyJHHHGE/Otf/6qwO3bsKHfccYecd955smfPHmnQYN//W9cIgKIoilJnKY/QPxGRkpIS49+uXbuqdWyLFy+WYDBY8Z+/iEhWVpbExsbKkiU8zdyfrVu3SiAQqNJ//iI6AFAURVHqMJGcA5CamipJSUkV//LyqBhUFSkqKpLWrVsbnzVo0ECaN28uRUWe5cYqZfPmzXLbbbdZZQM/olsCUBRFUZQDxIYNGyQQcEWlhg05aXsv119/vdx9993Wfa1Zs8bq3xdKSkpk6NCh0q1bN7n55pur/P2oHgCM/cDVpmJBG+dys13Ixuz3xlyplEVfSLR9nYSg7aZp6EwsZ7NciLpkIfmoCqv8HdqNvjV9fUmrF5wHkkO+xEvpA9CTvzaV8fWUT4wScDidHK8Dlwn+msTQjigYDqCND+fFThHWwikUhxeREqC5ngPmP4erA2DcY55r4qlbEIQ2K64NfNrhtuXzJhuOgedf8HMRgjafJ5d+NQ+JZxsfYtkzz2ego4oBjfNXs7dsop/BgKjn+AgMbfJ5e24TF0gw4DLHQDPun2ZWPh4+HwOvzD0M2iccQU66hNdCDXJa+NyzRDW/oxC+5zjl5mryZXCsGF9oPAmA5/ngpAsopFGyQ8z3VQ1RJtUPdf/+3AcCAWMA4MfVV18tF1xwgXWbDh06SEpKimzatMn4fM+ePbJlyxZJSbEX8962bZuccsop0rRpU3n55ZfloIP4zRyeqB4AKIqiKIqN2qgD0KpVK2nViksieenXr5+EQiEpKCiQjIwMERF59913pby8XPr08UxXraCkpEQGDx4sDRs2lNdee00SEmwTpP3ROQCKoiiKUgt07dpVTjnlFBk3bpwsXbpUFi1aJBMmTJCzzz67IgNg48aN0qVLF1m6dKmI7P3P/+STT5bt27fL448/LiUlJVJUVCRFRUVSVla1fIeojgCsFjeMhRFKvgQfko3B7iveNH19SdK5HSLLvNgXh7cxpMZhxrPIvgyiNWdS3HmZaRrH+wT5GtPJ9cDY4gUsPGSQDaNGyo3bYJpGxI/HpZwWiClkfI24JOqdEM4MsGQh/cnGJdU4NEvx+J/cZikdBK8GZ0vIs4bGOeLGqy1a0xbx0bOF/Nnm86YwIRxDd+rLLUglwcvC0gfb5iFyUVnuV7Zz4+PH3mTm2fLigLgnDqHbgp/hJCvjZD3L69nuDdXW7mVKAJ1BNxtIezmM7BH9/Hdr5BOK+ah67hOBf7mG+2sPr1NGS3IOJvt4aPMLIY1sfAzwR0ok6iSAmuC5556TCRMmyKBBgyQ2NlZGjRolDzzwQIV/9+7dsnbtWtmxY+/bZ8WKFRUZAocdZvakdevWSVpa2j7/dlQPABRFURTFxh+9FHDz5s1lzpw5vv60tDRxHKfCPuGEEwy7OqgEoCiKoij1EI0AKIqiKHWWP3IlwNomqgcAfktS8s3mVBhMJ/oL+ZJJJ8VlUkO0LWuLadCmVXrlMl4VFbTaPotNF88BQBWS9esPyO5hyJBp5O1ONuTK0UXrRFuOhTbPb+DqxFOhzelaXGIWdd6AZ51engOAd5nT4Qi4OXz/WebFfmQreSpCSzVz+ljLwy0HtH+zdPcSsvgoBQ9vHP1kU+rbqIVzKNAjhRsf5JNzkOWYODGNNXXIeetszgHgOdQ4xeJY8vFyxiXQ5vsfIrvFRjC4g/bnb+NN72m6aA5LOuTopfNEGdb5sYYLFYDbREuUPwvtEO2GexmmH/KzyHcCXx1TN5u+K54z7XjcGa/pyxN/8KWFPxqZKHZYykQkJgL7qIuoBKAoiqIo9ZCojgAoiqIoio0/+iTA2kQHAIqiKEqdRSUAf6J6AHCieEvuiniXCrVJUj+Rj21bni1rlKdBm4vuehKB4Yd4egDreCi/sm7nKYlqDFVZGeU9w+2nVSdb0zqdrfGHaTeZn5g2ukP0i3w9DdnfU6OVS8yi7h+m68JBVNZH/OAccS6na9xGnijBmrDxbRJVjXx4Phee34B+Xp6WbgYK5VR/NpYr+AL8F06INzC0ca6IMZtsPleE6wDAvJSTTE8LmhvTHx7cLnQuLEPjHIAfycdL6J4Dh5vIWr215kHQdPWl0sA5oKpzx+J1cHF+xvOm6wHTlFXQ5v+YePFYfA/R5fQsHYzzcV4iH79Tb5jvtj+gLjmXtsWaIvgshqthECl0AOCPzgFQFEVRlHpIVEcAFEVRFMWGzgHwJ6oHANMGigR+PwOMJ1HM36EV6E6EdiHtsyphKc5aw3K/cVnk5JAf5M5xYN4mAXAk0bPchKEnhLu9kNLUl1wsWWA2F8UO/0GbYiiRryen2RlRas/iVxyAR0Jk01WDWGNjqhubEG4pOYBDqsZl8SxWyBIAhvI59I2+cCmCVVgNEK8hvbVsKxtyiJOlppWwr/SFtOPjQrQ1ng/LAVxjNug2eRU8XlUODuqD5abLtuodKx/0VTkd2om02qZdzmCoXu4ZIAHwBQ2RDat6FtBKnFzKHOUNlrd4Yb7zoUueTymOV9JzgDIKyyb8jL8PYX9OGeXv4jHio3igwuoqAfijEoCiKIqi1EOiOgKgKIqiKDYcqX4I/wDVLDrg6ABAURRFqbNEInxfVyWA6B4ApIkrMKGeRXMAHqavYeVPa1qamIk+hWEOxxhlsnzdi2wQ3MKVn7XpNB6VPMHXqMSG2x9/ounqQAV+QZdcSdmFnE6EeizrzmzH4AXnNV6t3ZOVXdozHEQZaZ18vfHh5ivE0q3h95Qu5rRF32+Kmc5nW5BYxJw/EGbbAExMaGguT8svMbxi/Bywpo6PVDrNqZHjbGmMQfJxGmOLSpsi4hGTS0G8N4sGe5cOxmeRr9hGso2e46mBTIK8ZEKb1lvmZZHbweyCjXTElOqHJ/Q4uXhpbs9yxoCnd+DJnmq6Jv3LtPGe8zMdIhsvE6cI8rwknF6Ej/guEaEMYuUAE90DAEVRFEWxoBEAf3QAoCiKotRZyqX6WQB1NQ1QswAURVEUpR4S1RGAj2a7meyF8DnrV1wCEyVhqpYq55B9CbTPJx8vdPoutC9kZx+yoewt5/Lawk3sY+3T1E1tpXRFTH0203RtMecAlMCchQdpL6xRop4cbn6DAd8Mj16McAFlSuAG0fqftCUXSEZYz7QtodrYc7yc649ny+psA5/tRLznjbno/Mjyd493mynmHABZZ5p8rgj/xWP0O0+xDNu5MRZffMC0vywxTJTNeRoCHwEeL58LF/fF/V5Cl8zTr4zlgGl9Zc98B5gT8DXNAXjLNOfBg7yS9mI7N+6C1lvTw3Sl0IM7eanbHkO7KbT8DvdAruj9UPvKNy4pE3nwC6lxVALwJ6oHAIqiKIpiQwcA/ugAQFEURamz6BwAf6J6ADBZ3BPAUC2v6Ge7eZyycgnZLQa47c5Uk5ND30ugfSFHDhOpqGxTN0bN+7ElenFI+gOyRxopkLaEIREzZJ1kupaZJqZSLjFdnpQxDvsjnpE0nqzncG2CQZh0OOgEfI1KyC73aVf2K8a5BflHbeF5W5jctvof2yHy8Xehfi7FYuNJC7OVvWafoWgFLV8UEfvqhTaZxMRZY9oYNedys7a/0Lg/8v3HtQ3PIG2hxZ6l5gcNsJQxX3u+x13cJq0y+DXt9gloF9Ne+Nywj/J98kzqwo25UjWlSzeGYwqXvou7ZSmJ5ZnRID3hW5AFFOXAE9UDAEVRFEWxEYm/3jUCoCiKoihRhg4A/NE0QEVRFEWph0R1BKBI3BEM6mThlqBFPYuXe23Rnj6AWruNaQ4A62KGpMZCGC+DGnS3LqQtWee3lTXlrKXt8OXGsoi8rLeyhglQBdSPoM2ph3y8B/m0RbyVVn+GLzdL5IRImzbO59LYNGEOBkmd1tE86618brivNE8pYE5NbODTFrHPYQhXkBjhI4Z+FrTv1TZXg4vcGn2dT9Na7jdcn/M/t1Vk47Qa1p3p7huHyPebM3RxPsHn5Osfog9a4vHyufHsI7jnlHv4Km2J58rPiO2vNH4Hee4pdo9jyEeTeTB9mudJ2J62EPl4fga+NnG+0IFaYKcsAr9VVyMAUT0AUBRFURQbOgDwRyUARVEURamHaARAURRFqbPoJEB/onoA0EhcDcymkvKcANSvOvHGaWTDJIEguaxlbVlE5UsNYh0fu6fCLMAhG9YLUcPM2EhKaVveG+dk++/Ys0oqwLoj2qxR8n5wqkTfEjreAM2bMAiRTVnFoH3yMfA1RJvPhbVmTOdO8xyTTbu3LMVcJc3ftvQu2dSReOlo7A58X6gor7nyMQvu+33ejHm/y0iJxvvEx8dFr++ENp9bHtnYc3i+iGc14EFBMLjGAT/0UHSaOhYvZo3nRhU5PP0Xb6ttiWcRkZ1wcon8YkkxTZwbwbeYNhUoj4Irm4uIp+SB0Rtwls+B+k9VJQB/VAJQFEVRlHpIlQcACxculGHDhkmbNm0kJiZGXnnlFcPvOI5MnjxZDj74YElMTJSsrCz58ssvjW22bNkiY8aMkUAgIMFgUC666CL55Rfbwi+KoiiKUnXKZW8UoDr/6moEoMoSwPbt26Vnz55y4YUXysiRIz3+e+65Rx544AF56qmnpH379nLTTTfJ4MGD5bPPPpOEhL3hwDFjxsgPP/wg8+fPl927d0tOTo5cfPHFMmfOnCodSwBOACNsHEDlsFgHaPMifZ54YdBtcgiVs6GMsBlntEkL3/2SxyMBNOJd+e/VlClC5GzLV+YwaPc3XZ1NsymUkWVJha8vXgcOqXPmXAiNT8nZnxO2bGl1FDg92m3yffOkTlqOj0Ofxr3xROo5HI+D2s3ks0kANnGJU834N0Nus4PpocXgBIflXH6WS2TzdTDh9fWwX/Haiyw74UU071RPCiajXMfls1nKywA1oZQuPV8HTPX0pNEtJ3sQFg4eRE4uig3nRottdqMtMc2W90K3USZA+w3ycQoevs4SOX+XbnILf5fHvg7afcl3A9kboY09u1y8vbkmiMRaAAcqZfFAU+UIwKmnniq33367nH766R6f4zgyffp0ufHGG2X48OFy5JFHytNPPy3ff/99RaRgzZo1Mm/ePPn73/8uffr0kQEDBsiMGTPkhRdekO+/9yhwIiKya9cuKSkpMf4piqIoSjiq+9f/7/9qiv2JiF9yySXSsWNHSUxMlFatWsnw4cPl88+5ikV4IjoHYN26dVJUVCRZWVkVnyUlJUmfPn1k8eK9f0IuXrxYgsGgZGa6Q+KsrCyJjY2VJUt4mZm95OXlSVJSUsW/1FRecVpRFEVRoo8xY8bI6tWrZf78+fLGG2/IwoUL5eKLL7Z+JyMjQ5544glZs2aNvPPOO+I4jpx88slSVla1oUpEswCKivaGbJOTzeB4cnJyha+oqEhat25tHkSDBtK8efOKbZhJkyZJbm5uhb1161Zp166dMSrza1cGhrC50lsJfxmieLwth8LRLuGYUQmFandW+hMi4g3q2s6HjwFDfiU8iPR+AG0KPNJB4O+E08PQz8fO54a/WsKxT8/xYndl4YG+DDeLVx2zra7GPr6+eEQlns5jO14+Cjx+vir8q/hd/lFLZb2dpoe/iRkO/Iuc/WC/T/wB9kK+Jvxg4HfNc+Et8Zj4rPl4S8r9fXwn0O+5pfxwluBFZb2Qv73T12W7i+F6A14x27mImEeYEOZw0QzXHyxvDs/x+622+XvbcWo2wF4mkZMAOPrcsGFDadjQk/K1z/weEV+2bFnFH8UzZsyQIUOGyNSpU6VNG65VuxccIKSlpcntt98uPXv2lMLCQunYseO+H4BTDUTEefnllyvsRYsWOSLifP/998Z2Z555pnPWWWc5juM4d9xxh3P44Yd79tWqVSvnoYce2qff3bBhgyN774n+03/6T//pvyj+t2HDhv3/T8jCzp07nZSUlIgdZ5MmTTyfTZkypVrH+PjjjzvBYND4bPfu3U5cXJzz0ksv7dM+fvnlF2fixIlO+/btnV27dlXp9yMaAUhJ2TtVqLi4WA4++OCKz4uLiyU9Pb1im02bNhnf27Nnj2zZsqXi++Fo06aNbNiwQRzHkXbt2smGDRskEODMYEVk74g1NTVVr5EFvUbh0Wu0b+h1Cs/v12j9+vUSExPj+1dudUlISJB169ZJaSnHL/YPx3EkJsaMJVTnr3+R/YuI/85DDz0k1157rWzfvl06d+4s8+fPl/h4z+xzKxEdALRv315SUlIkPz+/4j/8kpISWbJkiVx66aUiItKvXz8JhUJSUFAgGRkZIiLy7rvvSnl5ufTp45mTXymxsbFyyCGHVIRjAoGAPmxh0GsUHr1G4dFrtG/odQpPUlJSjV+jhISEiuyzA8n1118vd999t3WbNWvWVOs3xowZIyeddJL88MMPMnXqVDnrrLNk0aJFVTrfKg8AfvnlF/nqKzftZ926dbJy5Upp3ry5tGvXTiZOnCi33367dOrUqSINsE2bNjJixAgREenatauccsopMm7cOJk1a5bs3r1bJkyYIGeffXaNjQQVRVEU5UBx9dVXywUXXGDdpkOHDtWKiP8+Kb5Tp07St29fadasmbz88styzjnn7PNxVnkAsHz5cjnxxBMr7N8n52VnZ8uTTz5ZEZK4+OKLJRQKyYABA2TevHnGqOS5556TCRMmyKBBgyQ2NlZGjRolDzzwQFUPRVEURVH+cLRq1UpateLqCV4iEREX2StPOI4ju3bxtNDwX4xafv31V2fKlCnOr7/+WtuH8odFr1F49BqFR6/RvqHXKTx6jUxOOeUU56ijjnKWLFnifPjhh06nTp2cc845p8L/3XffOZ07d3aWLFniOI7jfP31186dd97pLF++3Pn222+dRYsWOcOGDXOaN2/uFBcXV+m3YxynhnMwFEVRFEWplC1btsiECRPk9ddfNyLiTZrsrZpZWFgo7du3l/fee09OOOEE+f777+XPf/6zFBQUyM8//yzJycly3HHHyeTJk6Vz585hfs1EBwCKoiiKUg/R1QAVRVEUpR6iAwBFURRFqYfoAEBRFEVR6iE6AFAURVGUekjUDgBmzpwpaWlpkpCQIH369JGlS5fW9iHVGnl5edKrVy9p2rSptG7dWkaMGCFr1641tvn1119l/Pjx0qJFC2nSpImMGjVKiot5Ffj6w1133SUxMTEyceLEis/0Gu1l48aNct5550mLFi0kMTFRevToIcuXL6/wO44jkydPloMPPlgSExMlKytLvvzyy1o84gNLWVmZ3HTTTdK+fXtJTEyUjh07ym233WYsalPfrtHChQtl2LBh0qZNG4mJialY/v139uV67M+yuEo1iUgi4wHmhRdecOLj453Zs2c7q1evdsaNG+cEg8Eq50DWFQYPHuw88cQTzqpVq5yVK1c6Q4YMcdq1a+f88ssvFdv85S9/cVJTU538/Hxn+fLlTt++fZ1jjjmmFo+69li6dKmTlpbmHHnkkc6VV15Z8bleI8fZsmWLc+ihhzoXXHCBs2TJEuebb75x3nnnHeerr76q2Oauu+5ykpKSnFdeecX573//6/zpT39y2rdv7+zcubMWj/zAcccddzgtWrRw3njjDWfdunXOiy++6DRp0sS5//77K7apb9forbfecv72t785L730kiNiLhLnOPt2PU455RSnZ8+ezscff+x88MEHzmGHHWbkwyuRJyoHAL1793bGjx9fYZeVlTlt2rRx8vLyavGo/jhs2rTJERHn/fffdxzHcUKhkHPQQQc5L774YsU2a9ascUTEWbx4cW0dZq2wbds2p1OnTs78+fOd448/vmIAoNdoL9ddd50zYMAAX395ebmTkpLi3HvvvRWfhUIhp2HDhs7zzz9/IA6x1hk6dKhz4YUXGp+NHDnSGTNmjOM4eo14ALAv1+Ozzz5zRMRZtmxZxTZvv/22ExMT42zcuPGAHXt9I+okgNLSUikoKJCsrKyKz2JjYyUrK0sWL15ci0f2x2Hr1q0iItK8eXMRESkoKJDdu3cb16xLly7Srl27enfNxo8fL0OHDjWuhYheo9957bXXJDMzU84880xp3bq1HHXUUfLYY49V+NetWydFRUXGdUpKSpI+ffrUm+t0zDHHSH5+vnzxxRciIvLf//5XPvzwQzn11FNFRK8Rsy/XY/HixRIMBiUzM7Nim6ysLImNjZUlS5Yc8GOuL0R0NcADwebNm6WsrEySk5ONz5OTk+Xzzz+vpaP641BeXi4TJ06U/v37yxFHHCEie5ecjI+Pl2AwaGybnJwcdsnJusQLL7wgK1askGXLlnl8eo328s0338jDDz8subm5csMNN8iyZcvkiiuukPj4eMnOzq64FpU9f/XlOl1//fVSUlIiXbp0kbi4OCkrK5M77rhDxowZIyKi14jYl+tRnWVxlf0n6gYAip3x48fLqlWr5MMPP6ztQ/lDsWHDBrnyyitl/vz5tbI8aLRQXl4umZmZcuedd4qIyFFHHSWrVq2SWbNmSXZ2di0f3R+Df/zjH/Lcc8/JnDlzpHv37rJy5UqZOHGitGnTRq+RElVEnQTQsmVLiYuL88zOLi4uDrt8Yl1nwoQJ8sYbb8h7770nhxxySMXnKSkpUlpaKqFQyNi+Pl2zgoIC2bRpkxx99NHSoEEDadCggbz//vvywAMPSIMGDSQ5ObneXyMRkYMPPli6detmfNa1a1dZv369iEjFtajPz98111wj119/vZx99tnSo0cPOf/88+Wqq66SvLw8EdFrxOzL9ajOsrjK/hN1A4D4+HjJyMiQ/Pz8is/Ky8slPz9f+vXrV4tHVns4jiMTJkyQl19+Wd59911p37694c/IyJCDDjrIuGZr166V9evX15trNmjQIPn0009l5cqVFf8yMzNlzJgxFe36fo1ERPr37+9JIf3iiy/k0EMPFRGR9u3bS0pKinGdSkpKZMmSJfXmOu3YsUNiY81XZ1xcnJSXl4uIXiNmX64HLov7O/uzLK5SRWp7FuL+8MILLzgNGzZ0nnzySeezzz5zLr74YicYDDpFRUW1fWi1wqWXXuokJSU5CxYscH744YeKfzt27KjY5i9/+YvTrl07591333WWL1/u9OvXz+nXr18tHnXtg1kAjqPXyHH2pkg2aNDAueOOO5wvv/zSee6555xGjRo5zz77bMU2d911lxMMBp1XX33V+d///ucMHz68Tqe4MdnZ2U7btm0r0gBfeuklp2XLls61115bsU19u0bbtm1zPvnkE+eTTz5xRMSZNm2a88knnzjffvut4zj7dj3CLYurRJ6oHAA4juPMmDHDadeunRMfH+/07t3b+fjjj2v7kGoNEan03xNPPFGxzc6dO53LLrvMadasmdOoUSPn9NNPd3744YfaO+g/ADwA0Gu0l9dff9054ogjnIYNGzpdunRxHn30UcNfXl7u3HTTTU5ycrLTsGFDZ9CgQc7atWtr6WgPPCUlJc6VV17ptGvXzklISHA6dOjg/O1vf3N27dpVsU19u0bvvfdepe+g7Oxsx3H27Xr89NNPzjnnnOM0adLECQQCTk5OjrNt27ZaOJv6gy4HrCiKoij1kKibA6AoiqIoSvXRAYCiKIqi1EN0AKAoiqIo9RAdACiKoihKPUQHAIqiKIpSD9EBgKIoiqLUQ3QAoCiKoij1EB0AKIqiKEo9RAcAiqIoilIP0QGAoiiKotRDdACgKIqiKPWQ/we5a+m/uPZY7AAAAABJRU5ErkJggg==", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# Now we can use the model to generate designs\n", + "n_samples = 3\n", + "\n", + "conditions = [\n", + " {\"volume\": 0.5, \"length\": 0.5},\n", + " {\"volume\": 0.4, \"length\": 0.7},\n", + " {\"volume\": 0.2, \"length\": 0.1},\n", + "]\n", + "\n", + "conditions_tensor = th.tensor([list(c.values()) for c in conditions], device=device)\n", + "conditions_tensor = conditions_tensor.unsqueeze(1) # Add channel dim\n", + "\n", + "design_shape = problem.design_space.shape\n", + "# Denoise the designs in batch\n", + "gen_designs = th.randn((n_samples, 1, *design_shape), device=device)\n", + "for i in reversed(range(run.config[\"num_timesteps\"])):\n", + " t = th.full((n_samples,), i, device=device, dtype=th.long)\n", + " gen_designs = ddm_sampler.sample_timestep(model, gen_designs, t, conditions_tensor)\n", + "\n", + "gen_designs = gen_designs.cpu().numpy()\n", + "\n", + "# Render the designs\n", + "for i in range(n_samples):\n", + " problem.render(gen_designs[i][0]) # 0 removes the channel dim" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Evaluating the model" + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "WARNING: The requested image's platform (linux/amd64) does not match the detected host platform (linux/arm64/v8) and no specific platform was requested\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.931e-17 (tol = 1.000e-07) r (rel) = 4.977e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "\n", + "List of options:\n", + "\n", + " Name Value # times used\n", + " acceptable_tol = 0.001 1\n", + " file_print_level = 5 1\n", + " hessian_approximation = limited-memory 7\n", + " max_iter = 100 1\n", + " output_file = /home/fenics/shared/templates/RES_OPT/solution_V=0.5_w=0.5.txt 1\n", + " print_level = 6 2\n", + "\n", + "******************************************************************************\n", + "This program contains Ipopt, a library for large-scale nonlinear optimization.\n", + " Ipopt is released as open source code under the Eclipse Public License (EPL).\n", + " For more information visit http://projects.coin-or.org/Ipopt\n", + "******************************************************************************\n", + "\n", + "This is Ipopt version 3.12.9, running with linear solver mumps.\n", + "NOTE: Other linear solvers might be more efficient (see Ipopt documentation).\n", + "\n", + "Number of nonzeros in equality constraint Jacobian...: 0\n", + "Number of nonzeros in inequality constraint Jacobian.: 10201\n", + "Number of nonzeros in Lagrangian Hessian.............: 0\n", + "\n", + "Hessian approximation will be done in the space of all 10201 x variables.\n", + "\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Scaling parameter for objective function = 1.000000e+00\n", + "objective scaling factor = 1\n", + "No x scaling provided\n", + "No c scaling provided\n", + "No d scaling provided\n", + "Moved initial values of x sufficiently inside the bounds.\n", + "Initial values of s sufficiently inside the bounds.\n", + "MUMPS used permuting_scaling 5 and pivot_order 5.\n", + " scaling will be 77.\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Least square estimates max(y_c) = 0.000000e+00, max(y_d) = 9.999282e-01\n", + "Total number of variables............................: 10201\n", + " variables with only lower bounds: 0\n", + " variables with lower and upper bounds: 10201\n", + " variables with only upper bounds: 0\n", + "Total number of equality constraints.................: 0\n", + "Total number of inequality constraints...............: 1\n", + " inequality constraints with only lower bounds: 1\n", + " inequality constraints with lower and upper bounds: 0\n", + " inequality constraints with only upper bounds: 0\n", + "\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.947e-17 (tol = 1.000e-07) r (rel) = 3.985e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 0:\n", + "**************************************************\n", + "\n", + "Limited-Memory approximation started; store data at current iterate.\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 0:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 0 4.6981653e-04 0.00e+00 1.19e-02 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 0 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 1.0000000000000000e+00\n", + "Current fraction-to-the-boundary parameter tau = 0.0000000000000000e+00\n", + "\n", + "||curr_x||_inf = 5.7071268558502197e-01\n", + "||curr_s||_inf = 3.3710614818985063e-01\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 9.9992815818945313e-01\n", + "||curr_z_L||_inf = 1.0000000000000000e+00\n", + "||curr_z_U||_inf = 1.0000000000000000e+00\n", + "||curr_v_L||_inf = 1.0000000000000000e+00\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 0:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 4.6981652935133546e-04 4.6981652935133546e-04\n", + "Dual infeasibility......: 1.1890392544966710e-02 1.1890392544966710e-02\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 9.9000001999999998e-01 9.9000001999999998e-01\n", + "Overall NLP error.......: 9.9000001999999998e-01 9.9000001999999998e-01\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 0:\n", + "**************************************************\n", + "\n", + "Setting mu_max to 4.999920e+02.\n", + "Staying in free mu mode.\n", + "The current filter has 1 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 1.302505e-02\n", + "Barrier parameter mu computed by oracle is 6.512420e-03\n", + "Barrier parameter mu after safeguards is 6.512420e-03\n", + "Barrier Parameter: 6.512420e-03\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 0:\n", + "**************************************************\n", + "\n", + "residual_ratio = 2.090756e-15\n", + "Factorization successful.\n", + "residual_ratio = 1.123732e-16\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 0:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 0 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 0.000000E+00\n", + "Starting checks for alpha (primal) = 1.00e+00\n", + "trial_max is initialized to 1.000000e+04\n", + "trial_min is initialized to 1.000000e-04\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.995e-17 (tol = 1.000e-07) r (rel) = 4.033e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 1.9915250594961395e+02 (reference 2.1626223139200656e+02):\n", + " New values of constraint violation = 2.7755575615628914e-16 (reference 0.0000000000000000e+00):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 1:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 6.437118e-03\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 1.695689e-03 snrm = 5.050718e-01 ynrm = 2.704053e-02\n", + " Perform the update.\n", + "sigma (for B0) is 6.647218e-03\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 1:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 1 4.6869866e-04 0.00e+00 6.30e-03 -2.2 6.44e-03 - 9.98e-01 1.00e+00f 1 sigma=1.30e-02 qf=13y \n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 1 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 6.5124202235485916e-03\n", + "Current fraction-to-the-boundary parameter tau = 9.8999999999999999e-01\n", + "\n", + "||curr_x||_inf = 5.6997274809964926e-01\n", + "||curr_s||_inf = 3.3285290627583736e-01\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 3.8559795683230300e-02\n", + "||curr_z_L||_inf = 2.1724872561545538e-02\n", + "||curr_z_U||_inf = 1.5901675300216689e-02\n", + "||curr_v_L||_inf = 3.4344526508572670e-02\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 1:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 4.6869866234459505e-04 4.6869866234459505e-04\n", + "Dual infeasibility......: 6.3037649451827341e-03 6.3037649451827341e-03\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 1.5280891981050969e-02 1.5280891981050969e-02\n", + "Overall NLP error.......: 1.5280891981050969e-02 1.5280891981050969e-02\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 1:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 2 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 1.000000e-06\n", + "Barrier parameter mu computed by oracle is 7.753027e-09\n", + "Barrier parameter mu after safeguards is 7.753027e-09\n", + "Barrier Parameter: 7.753027e-09\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 1:\n", + "**************************************************\n", + "\n", + "residual_ratio = 1.284014e-14\n", + "Factorization successful.\n", + "residual_ratio = 9.665972e-17\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 1:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 1 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.710247E-18\n", + "Starting checks for alpha (primal) = 1.00e+00\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.950e-17 (tol = 1.000e-07) r (rel) = 3.987e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 7.0570783360603562e-04 (reference 7.0578887277047671e-04):\n", + " New values of constraint violation = 5.5511151231257827e-17 (reference 2.7755575615628914e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 2:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 1.656226e-04\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 2.210383e-11 snrm = 1.074683e-03 ynrm = 2.503084e-08\n", + " Perform the update.\n", + "sigma (for B0) is 1.913845e-05\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 2:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 2 4.6861860e-04 0.00e+00 6.49e-05 -8.1 1.66e-04 - 9.90e-01 1.00e+00f 1 Nhj sigma=1.00e-06 qf=13y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 2 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 7.7530267954892570e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.8999999999999999e-01\n", + "\n", + "||curr_x||_inf = 5.6998421098192886e-01\n", + "||curr_s||_inf = 3.3284765987999715e-01\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 3.9757439530749804e-04\n", + "||curr_z_L||_inf = 2.2367460987769358e-04\n", + "||curr_z_U||_inf = 1.6845304188516097e-04\n", + "||curr_v_L||_inf = 3.5414383457792881e-04\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 2:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 4.6861859532962020e-04 4.6861859532962020e-04\n", + "Dual infeasibility......: 6.4897480859802347e-05 6.4897480859802347e-05\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 1.5732488550385427e-04 1.5732488550385427e-04\n", + "Overall NLP error.......: 1.5732488550385427e-04 1.5732488550385427e-04\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 2:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 2 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 1.302505e-02\n", + "Barrier parameter mu computed by oracle is 1.039736e-06\n", + "Barrier parameter mu after safeguards is 1.039736e-06\n", + "Barrier Parameter: 1.039736e-06\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 2:\n", + "**************************************************\n", + "\n", + "residual_ratio = 2.930508e-14\n", + "Factorization successful.\n", + "residual_ratio = 3.808328e-17\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 2:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 2 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.784396E-23\n", + "Starting checks for alpha (primal) = 7.27e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.355e-17 (tol = 1.000e-07) r (rel) = 4.396e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 7.265680e-01:\n", + " New values of barrier function = 3.1239024315965386e-02 (reference 3.2263967256597775e-02):\n", + " New values of constraint violation = 1.1102230246251565e-16 (reference 5.5511151231257827e-17):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 3:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 1.643432e-02\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 1.141847e-07 snrm = 2.866860e-01 ynrm = 1.844972e-06\n", + " Perform the update.\n", + "sigma (for B0) is 1.389296e-06\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 3:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 3 4.6176366e-04 0.00e+00 2.04e-04 -6.0 2.26e-02 - 9.83e-01 7.27e-01f 1 sigma=1.30e-02 qf=13y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 3 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 1.0397358898902715e-06\n", + "Current fraction-to-the-boundary parameter tau = 9.9984267511449609e-01\n", + "\n", + "||curr_x||_inf = 5.7023191573913001e-01\n", + "||curr_s||_inf = 3.3014142602921281e-01\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 1.4169907374203415e-05\n", + "||curr_z_L||_inf = 2.0401200255619860e-04\n", + "||curr_z_U||_inf = 1.1195518519408702e-05\n", + "||curr_v_L||_inf = 1.2976521882577077e-05\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 3:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 4.6176366280509224e-04 4.6176366280509224e-04\n", + "Dual infeasibility......: 2.0387405287655886e-04 2.0387405287655886e-04\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 4.8643644598696965e-06 4.8643644598696965e-06\n", + "Overall NLP error.......: 2.0387405287655886e-04 2.0387405287655886e-04\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 3:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 3 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 2.126374e-01\n", + "Barrier parameter mu computed by oracle is 4.858674e-07\n", + "Barrier parameter mu after safeguards is 4.858674e-07\n", + "Barrier Parameter: 4.858674e-07\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 3:\n", + "**************************************************\n", + "\n", + "residual_ratio = 9.907082e-16\n", + "Factorization successful.\n", + "residual_ratio = 1.204137e-16\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 3:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 3 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 4.968812E-24\n", + "Starting checks for alpha (primal) = 4.52e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.004e-17 (tol = 1.000e-07) r (rel) = 4.042e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 4.517635e-01:\n", + " New values of barrier function = 1.2061175807744772e-02 (reference 1.4843941377593832e-02):\n", + " New values of constraint violation = 5.5511151231257827e-17 (reference 1.1102230246251565e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 4:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 1.040402e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 1.621492e-05 snrm = 3.276010e+00 ynrm = 1.644879e-05\n", + " Perform the update.\n", + "sigma (for B0) is 1.510860e-06\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 4:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 4 3.7713666e-04 0.00e+00 5.29e-04 -6.3 2.30e-01 - 9.92e-01 4.52e-01f 1 sigma=2.13e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 4 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 4.8586735880636006e-07\n", + "Current fraction-to-the-boundary parameter tau = 9.9979612594712342e-01\n", + "\n", + "||curr_x||_inf = 6.4920348925804061e-01\n", + "||curr_s||_inf = 2.9883849060579321e-01\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 5.4574287619116684e-06\n", + "||curr_z_L||_inf = 5.2967094494315995e-04\n", + "||curr_z_U||_inf = 5.9782717018244279e-06\n", + "||curr_v_L||_inf = 4.2695877359171917e-06\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 4:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 3.7713665675027903e-04 3.7713665675027903e-04\n", + "Dual infeasibility......: 5.2918727256519977e-04 5.2918727256519977e-04\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 2.1836733900361981e-06 2.1836733900361981e-06\n", + "Overall NLP error.......: 5.2918727256519977e-04 5.2918727256519977e-04\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 4:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 2 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 8.431257e-02\n", + "Barrier parameter mu computed by oracle is 4.044106e-08\n", + "Barrier parameter mu after safeguards is 4.044106e-08\n", + "Barrier Parameter: 4.044106e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 4:\n", + "**************************************************\n", + "\n", + "residual_ratio = 1.872524e-15\n", + "Factorization successful.\n", + "residual_ratio = 3.467637e-17\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 4:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 4 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 7.737005E-23\n", + "Starting checks for alpha (primal) = 1.00e+00\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.397e-17 (tol = 1.000e-07) r (rel) = 4.439e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 1.0817665871449902e-03 (reference 1.3496550354855446e-03):\n", + " New values of constraint violation = 0.0000000000000000e+00 (reference 5.5511151231257827e-17):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 5:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 3.996796e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 1.016755e-04 snrm = 5.762415e+00 ynrm = 2.988430e-05\n", + " Perform the update.\n", + "sigma (for B0) is 3.062016e-06\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 5:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 5 2.1261573e-04 0.00e+00 1.85e-04 -7.4 4.00e-01 - 6.17e-01 1.00e+00f 1 sigma=8.43e-02 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 5 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 4.0441060660647348e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9947081272743477e-01\n", + "\n", + "||curr_x||_inf = 9.8126103107849083e-01\n", + "||curr_s||_inf = 2.5108965233471248e-01\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 2.5236602439941591e-06\n", + "||curr_z_L||_inf = 1.8555237791197571e-04\n", + "||curr_z_U||_inf = 5.8122187124815824e-06\n", + "||curr_v_L||_inf = 2.1383340779829057e-06\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 5:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 2.1261573362645707e-04 2.1261573362645707e-04\n", + "Dual infeasibility......: 1.8534233532916164e-04 1.8534233532916164e-04\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 8.4195637365473728e-07 8.4195637365473728e-07\n", + "Overall NLP error.......: 1.8534233532916164e-04 1.8534233532916164e-04\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 5:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 1 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 5.306571e-01\n", + "Barrier parameter mu computed by oracle is 1.098352e-07\n", + "Barrier parameter mu after safeguards is 1.098352e-07\n", + "Barrier Parameter: 1.098352e-07\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 5:\n", + "**************************************************\n", + "\n", + "residual_ratio = 5.194011e-15\n", + "Factorization successful.\n", + "residual_ratio = 6.412359e-17\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 5:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 5 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 0.000000E+00\n", + "Starting checks for alpha (primal) = 3.48e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.198e-17 (tol = 1.000e-07) r (rel) = 4.237e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 3.476103e-01:\n", + " New values of barrier function = 2.2646585305238442e-03 (reference 2.5731707289686824e-03):\n", + " New values of constraint violation = 1.3877787807814457e-16 (reference 0.0000000000000000e+00):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 6:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 3.008583e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 1.149335e-05 snrm = 5.471624e+00 ynrm = 1.165050e-05\n", + " Perform the update.\n", + "sigma (for B0) is 3.838965e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 6:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 6 1.6502648e-04 0.00e+00 2.40e-04 -7.0 8.66e-01 - 7.84e-01 3.48e-01f 1 sigma=5.31e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 6 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 1.0983518841748851e-07\n", + "Current fraction-to-the-boundary parameter tau = 9.9981465766467081e-01\n", + "\n", + "||curr_x||_inf = 9.9994528190258514e-01\n", + "||curr_s||_inf = 2.0296687018492257e-01\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 2.2864271431690781e-06\n", + "||curr_z_L||_inf = 2.3964832790028859e-04\n", + "||curr_z_U||_inf = 8.6986550843502061e-06\n", + "||curr_v_L||_inf = 1.7290386471529564e-06\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 6:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.6502648484603295e-04 1.6502648484603295e-04\n", + "Dual infeasibility......: 2.3951685525649055e-04 2.3951685525649055e-04\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 3.9210848213316732e-07 3.9210848213316732e-07\n", + "Overall NLP error.......: 2.3951685525649055e-04 2.3951685525649055e-04\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 6:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 2 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 6.069074e-01\n", + "Barrier parameter mu computed by oracle is 8.231647e-08\n", + "Barrier parameter mu after safeguards is 8.231647e-08\n", + "Barrier Parameter: 8.231647e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 6:\n", + "**************************************************\n", + "\n", + "residual_ratio = 8.194075e-15\n", + "Factorization successful.\n", + "residual_ratio = 4.881576e-16\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 6:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 6 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.202324E-22\n", + "Starting checks for alpha (primal) = 6.32e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.274e-17 (tol = 1.000e-07) r (rel) = 4.315e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 6.316483e-01:\n", + " New values of barrier function = 1.4999285968201378e-03 (reference 1.7386050838862769e-03):\n", + " New values of constraint violation = 3.3306690738754696e-16 (reference 1.3877787807814457e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 7:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 5.026466e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 4.164968e-05 snrm = 1.010274e+01 ynrm = 1.104624e-05\n", + " Perform the update.\n", + "sigma (for B0) is 4.080689e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 7:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 7 9.0938693e-05 0.00e+00 1.74e-04 -7.1 7.96e-01 - 5.11e-01 6.32e-01f 1 sigma=6.07e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 7 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 8.2316471722319226e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9976048314474353e-01\n", + "\n", + "||curr_x||_inf = 9.9987958881490280e-01\n", + "||curr_s||_inf = 1.1226715633901446e-01\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 2.0399870080882624e-06\n", + "||curr_z_L||_inf = 1.7431917112882426e-04\n", + "||curr_z_U||_inf = 3.2767280368029262e-06\n", + "||curr_v_L||_inf = 1.6778132125913295e-06\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 7:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 9.0938693341593464e-05 9.0938693341593464e-05\n", + "Dual infeasibility......: 1.7421276895974660e-04 1.7421276895974660e-04\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 2.1454017604689819e-07 2.1454017604689819e-07\n", + "Overall NLP error.......: 1.7421276895974660e-04 1.7421276895974660e-04\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 7:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 3 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 5.908181e-01\n", + "Barrier parameter mu computed by oracle is 6.304055e-08\n", + "Barrier parameter mu after safeguards is 6.304055e-08\n", + "Barrier Parameter: 6.304055e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 7:\n", + "**************************************************\n", + "\n", + "residual_ratio = 1.185736e-15\n", + "Factorization successful.\n", + "residual_ratio = 2.195808e-17\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 7:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 7 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 7.005455E-22\n", + "Starting checks for alpha (primal) = 7.79e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.696e-17 (tol = 1.000e-07) r (rel) = 3.731e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 7.794678e-01:\n", + " New values of barrier function = 1.0775074420967959e-03 (reference 1.1699875213672679e-03):\n", + " New values of constraint violation = 7.4940054162198066e-16 (reference 3.3306690738754696e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 8:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 4.924977e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 6.898312e-06 snrm = 8.168003e+00 ynrm = 4.454266e-06\n", + " Perform the update.\n", + "sigma (for B0) is 1.033978e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 8:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 8 6.7419287e-05 0.00e+00 1.15e-05 -7.2 6.32e-01 - 9.34e-01 7.79e-01f 1 sigma=5.91e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 8 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 6.3040545656067363e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9982578723104021e-01\n", + "\n", + "||curr_x||_inf = 9.9996010556813475e-01\n", + "||curr_s||_inf = 4.2454333745357481e-02\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 1.9099643631276152e-06\n", + "||curr_z_L||_inf = 1.1534024292732947e-05\n", + "||curr_z_U||_inf = 1.4932187423352213e-06\n", + "||curr_v_L||_inf = 1.8854399454963210e-06\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 8:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 6.7419287418862974e-05 6.7419287418862974e-05\n", + "Dual infeasibility......: 1.1468057740637961e-05 1.1468057740637961e-05\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 1.3163180480721070e-07 1.3163180480721070e-07\n", + "Overall NLP error.......: 1.1468057740637961e-05 1.1468057740637961e-05\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 8:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 4 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 7.141901e-01\n", + "Barrier parameter mu computed by oracle is 4.502846e-08\n", + "Barrier parameter mu after safeguards is 4.502846e-08\n", + "Barrier Parameter: 4.502846e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 8:\n", + "**************************************************\n", + "\n", + "residual_ratio = 9.426621e-16\n", + "Factorization successful.\n", + "residual_ratio = 1.216338e-16\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 8:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 8 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.768294E-21\n", + "Starting checks for alpha (primal) = 2.48e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.553e-17 (tol = 1.000e-07) r (rel) = 3.587e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 2.479107e-01:\n", + " New values of barrier function = 7.6363642718371054e-04 (reference 7.8890281077480544e-04):\n", + " New values of constraint violation = 2.2204460492503131e-16 (reference 7.4940054162198066e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 9:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 1.131385e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 2.270529e-07 snrm = 2.688633e+00 ynrm = 9.796588e-07\n", + " Perform the update.\n", + "sigma (for B0) is 3.140973e-08\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 9:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 9 6.1616508e-05 0.00e+00 3.02e-07 -7.3 4.56e-01 - 9.86e-01 2.48e-01f 1 sigma=7.14e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 9 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 4.5028460915594864e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9998853194225934e-01\n", + "\n", + "||curr_x||_inf = 9.9999880488680837e-01\n", + "||curr_s||_inf = 2.0657262339258076e-02\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 4.9292979228614336e-06\n", + "||curr_z_L||_inf = 2.2492120829200955e-07\n", + "||curr_z_U||_inf = 1.2912012036860158e-06\n", + "||curr_v_L||_inf = 4.9229087298590936e-06\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 9:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 6.1616507536971964e-05 6.1616507536971964e-05\n", + "Dual infeasibility......: 3.0169602048343037e-07 3.0169602048343037e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 1.0169386633401037e-07 1.0169386633401037e-07\n", + "Overall NLP error.......: 3.0169602048343037e-07 3.0169602048343037e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 9:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 3 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 2.599197e-01\n", + "Barrier parameter mu computed by oracle is 1.166838e-08\n", + "Barrier parameter mu after safeguards is 1.166838e-08\n", + "Barrier Parameter: 1.166838e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 9:\n", + "**************************************************\n", + "\n", + "residual_ratio = 5.519903e-16\n", + "Factorization successful.\n", + "residual_ratio = 2.529956e-16\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 9:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 9 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 6.165784E-22\n", + "Starting checks for alpha (primal) = 3.45e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.258e-17 (tol = 1.000e-07) r (rel) = 3.288e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 3.446822e-01:\n", + " New values of barrier function = 2.3283950449996471e-04 (reference 2.4353334985162379e-04):\n", + " New values of constraint violation = 5.2041704279304213e-17 (reference 2.2204460492503131e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 10:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 1.557527e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 6.923097e-07 snrm = 2.573810e+00 ynrm = 1.188048e-06\n", + " Perform the update.\n", + "sigma (for B0) is 1.045075e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 10:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 10 5.4256341e-05 0.00e+00 7.69e-07 -7.9 4.52e-01 - 8.27e-01 3.45e-01f 1 sigma=2.60e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 10 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 1.1668380333389575e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999969830397950e-01\n", + "\n", + "||curr_x||_inf = 9.9507586558926175e-01\n", + "||curr_s||_inf = 3.7677831445093179e-09\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 1.3138968831959693e-05\n", + "||curr_z_L||_inf = 5.2087917090374842e-08\n", + "||curr_z_U||_inf = 5.6912596151833623e-07\n", + "||curr_v_L||_inf = 1.3126783644354312e-05\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 10:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 5.4256340702876417e-05 5.4256340702876417e-05\n", + "Dual infeasibility......: 7.6857144345135908e-07 7.6857144345135908e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 6.2549075446186007e-08 6.2549075446186007e-08\n", + "Overall NLP error.......: 7.6857144345135908e-07 7.6857144345135908e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 10:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 2 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 5.017880e-01\n", + "Barrier parameter mu computed by oracle is 8.973299e-09\n", + "Barrier parameter mu after safeguards is 8.973299e-09\n", + "Barrier Parameter: 8.973299e-09\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 10:\n", + "**************************************************\n", + "\n", + "residual_ratio = 8.618274e-16\n", + "Factorization successful.\n", + "residual_ratio = 2.205774e-17\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 10:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 10 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 2.610811E-23\n", + "Starting checks for alpha (primal) = 1.01e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.357e-17 (tol = 1.000e-07) r (rel) = 3.388e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.013248e-01:\n", + " New values of barrier function = 1.9043355780963686e-04 (reference 1.9159161092197172e-04):\n", + " New values of constraint violation = 1.4220666744862998e-16 (reference 5.2041704279304213e-17):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 11:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 5.578051e-02\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 4.354765e-08 snrm = 6.053372e-01 ynrm = 2.589328e-07\n", + " Perform the update.\n", + "sigma (for B0) is 1.188420e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 11:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 11 5.3159759e-05 0.00e+00 4.51e-07 -8.0 5.51e-01 - 9.14e-01 1.01e-01f 1 sigma=5.02e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 11 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 8.9732992295176388e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999923142855651e-01\n", + "\n", + "||curr_x||_inf = 9.9999996712865646e-01\n", + "||curr_s||_inf = 6.9247794269786765e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 2.4378993444041145e-04\n", + "||curr_z_L||_inf = 4.6761919868473101e-08\n", + "||curr_z_U||_inf = 4.2123404786458668e-07\n", + "||curr_v_L||_inf = 2.4377915681495684e-04\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 11:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 5.3159758958582040e-05 5.3159758958582040e-05\n", + "Dual infeasibility......: 4.5071671022282290e-07 4.5071671022282290e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 5.8124118448248091e-08 5.8124118448248091e-08\n", + "Overall NLP error.......: 4.5071671022282290e-07 4.5071671022282290e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 11:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 3 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 5.701295e-01\n", + "Barrier parameter mu computed by oracle is 5.727484e-09\n", + "Barrier parameter mu after safeguards is 5.727484e-09\n", + "Barrier Parameter: 5.727484e-09\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 11:\n", + "**************************************************\n", + "\n", + "residual_ratio = 3.176722e-15\n", + "Factorization successful.\n", + "residual_ratio = 6.999270e-18\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 11:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 11 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.561954E-23\n", + "Starting checks for alpha (primal) = 1.00e+00\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.560e-17 (tol = 1.000e-07) r (rel) = 3.594e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 1.3102032591752507e-04 (reference 1.4077898616129297e-04):\n", + " New values of constraint violation = 1.0331091451071250e-16 (reference 1.4220666744862998e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 12:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 3.078676e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 1.102898e-05 snrm = 9.416229e+00 ynrm = 2.622160e-06\n", + " Perform the update.\n", + "sigma (for B0) is 1.243888e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 12:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 12 3.7956764e-05 0.00e+00 7.32e-07 -8.2 3.08e-01 - 6.77e-01 1.00e+00f 1 sigma=5.70e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 12 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 5.7274844182180619e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999954928328982e-01\n", + "\n", + "||curr_x||_inf = 9.9437953357957953e-01\n", + "||curr_s||_inf = 3.1351610257587199e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 2.2504329294282900e-04\n", + "||curr_z_L||_inf = 3.1309790137290428e-08\n", + "||curr_z_U||_inf = 3.0561553138126096e-07\n", + "||curr_v_L||_inf = 2.2504186731575406e-04\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 12:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 3.7956764385632902e-05 3.7956764385632902e-05\n", + "Dual infeasibility......: 7.3247186218490388e-07 7.3247186218490388e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 2.4231447904316846e-08 2.4231447904316846e-08\n", + "Overall NLP error.......: 7.3247186218490388e-07 7.3247186218490388e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 12:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 3 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 9.138677e-01\n", + "Barrier parameter mu computed by oracle is 6.293033e-09\n", + "Barrier parameter mu after safeguards is 6.293033e-09\n", + "Barrier Parameter: 6.293033e-09\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 12:\n", + "**************************************************\n", + "\n", + "residual_ratio = 1.605133e-16\n", + "Factorization successful.\n", + "residual_ratio = 1.423416e-17\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 12:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 12 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.556573E-20\n", + "Starting checks for alpha (primal) = 3.14e-02\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.464e-17 (tol = 1.000e-07) r (rel) = 3.497e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 3.135683e-02:\n", + " New values of barrier function = 1.4019061379382336e-04 (reference 1.4020969715885140e-04):\n", + " New values of constraint violation = 3.8678912503420371e-16 (reference 1.0331091451071250e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 13:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 2.907881e-02\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 2.740668e-09 snrm = 1.005609e-01 ynrm = 7.583800e-08\n", + " Perform the update.\n", + "sigma (for B0) is 2.710180e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 13:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 13 3.7840076e-05 0.00e+00 3.54e-07 -8.2 9.27e-01 - 9.52e-01 3.14e-02f 1 sigma=9.14e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 13 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 6.2930331650267817e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999926752813784e-01\n", + "\n", + "||curr_x||_inf = 9.9999998870057361e-01\n", + "||curr_s||_inf = 3.1223051299043446e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 2.2983971872904288e-04\n", + "||curr_z_L||_inf = 3.3881184349002631e-08\n", + "||curr_z_U||_inf = 3.9700808446844428e-07\n", + "||curr_v_L||_inf = 2.2983794081283382e-04\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 13:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 3.7840076303503619e-05 3.7840076303503619e-05\n", + "Dual infeasibility......: 3.5395331122452642e-07 3.5395331122452642e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 3.3989971175832244e-08 3.3989971175832244e-08\n", + "Overall NLP error.......: 3.5395331122452642e-07 3.5395331122452642e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 13:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 4 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 6.966207e-02\n", + "Barrier parameter mu computed by oracle is 4.420892e-10\n", + "Barrier parameter mu after safeguards is 4.420892e-10\n", + "Barrier Parameter: 4.420892e-10\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 13:\n", + "**************************************************\n", + "\n", + "residual_ratio = 9.454483e-16\n", + "Factorization successful.\n", + "residual_ratio = 3.860887e-17\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 13:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 13 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 3.650622E-21\n", + "Starting checks for alpha (primal) = 9.18e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.495e-17 (tol = 1.000e-07) r (rel) = 3.528e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 9.183855e-01:\n", + " New values of barrier function = 3.8793514754253226e-05 (reference 4.5030260798684637e-05):\n", + " New values of constraint violation = 4.3090598905899669e-16 (reference 3.8678912503420371e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 14:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 2.188907e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 2.672397e-06 snrm = 5.620350e+00 ynrm = 1.137752e-06\n", + " Perform the update.\n", + "sigma (for B0) is 8.460077e-08\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 14:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 14 3.1076341e-05 0.00e+00 3.56e-07 -9.4 2.38e-01 - 7.07e-01 9.18e-01f 1 sigma=6.97e-02 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 14 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 4.4208922200447680e-10\n", + "Current fraction-to-the-boundary parameter tau = 9.9999964604668878e-01\n", + "\n", + "||curr_x||_inf = 9.9999999092766290e-01\n", + "||curr_s||_inf = 1.2808792683518393e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 1.8163747627221880e-04\n", + "||curr_z_L||_inf = 1.9104449987264878e-08\n", + "||curr_z_U||_inf = 5.4610873970380496e-07\n", + "||curr_v_L||_inf = 1.8163689478518615e-04\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 14:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 3.1076340664289982e-05 3.1076340664289982e-05\n", + "Dual infeasibility......: 3.5598833436017428e-07 3.5598833436017428e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 1.0664751264724913e-08 1.0664751264724913e-08\n", + "Overall NLP error.......: 3.5598833436017428e-07 3.5598833436017428e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 14:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 5 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 1.000000e+00\n", + "Barrier parameter mu computed by oracle is 2.164006e-09\n", + "Barrier parameter mu after safeguards is 2.164006e-09\n", + "Barrier Parameter: 2.164006e-09\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 14:\n", + "**************************************************\n", + "\n", + "residual_ratio = 1.661119e-15\n", + "Factorization successful.\n", + "residual_ratio = 1.552665e-17\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 14:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 14 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.267328E-22\n", + "Starting checks for alpha (primal) = 3.30e-02\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.445e-17 (tol = 1.000e-07) r (rel) = 3.477e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 3.295347e-02:\n", + " New values of barrier function = 6.8743544301378399e-05 (reference 6.8851540453917292e-05):\n", + " New values of constraint violation = 3.2368008826607481e-16 (reference 4.3090598905899669e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 15:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 4.733475e-02\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 1.582016e-09 snrm = 1.971014e-01 ynrm = 5.017688e-08\n", + " Perform the update.\n", + "sigma (for B0) is 4.072221e-08\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 15:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 15 3.0906703e-05 0.00e+00 3.20e-07 -8.7 1.44e+00 - 6.69e-01 3.30e-02f 1 sigma=1.00e+00 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 15 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 2.1640056957869612e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999964401166563e-01\n", + "\n", + "||curr_x||_inf = 9.9999999314937627e-01\n", + "||curr_s||_inf = 1.2827375571853822e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 1.6771838170555599e-04\n", + "||curr_z_L||_inf = 2.0090482403092866e-08\n", + "||curr_z_U||_inf = 4.0346990677727674e-07\n", + "||curr_v_L||_inf = 1.6771662064446498e-04\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 15:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 3.0906703339077136e-05 3.0906703339077136e-05\n", + "Dual infeasibility......: 3.1980787284086894e-07 3.1980787284086894e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 1.2265452277247657e-08 1.2265452277247657e-08\n", + "Overall NLP error.......: 3.1980787284086894e-07 3.1980787284086894e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 15:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 4 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 1.000000e+00\n", + "Barrier parameter mu computed by oracle is 2.241880e-09\n", + "Barrier parameter mu after safeguards is 2.241880e-09\n", + "Barrier Parameter: 2.241880e-09\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 15:\n", + "**************************************************\n", + "\n", + "residual_ratio = 1.458171e-15\n", + "Factorization successful.\n", + "residual_ratio = 1.042701e-16\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 15:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 15 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.658530E-23\n", + "Starting checks for alpha (primal) = 5.84e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.651e-17 (tol = 1.000e-07) r (rel) = 3.685e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 5.837277e-01:\n", + " New values of barrier function = 6.8497431208030537e-05 (reference 7.0105147191734408e-05):\n", + " New values of constraint violation = 4.4002853390092550e-16 (reference 3.2368008826607481e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 16:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 16 2.8155857e-05 0.00e+00 2.59e-07 -8.6 5.78e-01 - 8.73e-01 5.84e-01f 1 sigma=1.00e+00 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 16 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 2.2418799668728096e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999968019212715e-01\n", + "\n", + "||curr_x||_inf = 9.9999999423380315e-01\n", + "||curr_s||_inf = 1.3912584531265601e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 1.5255675550238714e-04\n", + "||curr_z_L||_inf = 1.8194101808679699e-08\n", + "||curr_z_U||_inf = 5.6874469415335646e-07\n", + "||curr_v_L||_inf = 1.5255679148921584e-04\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 16:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 2.8155857302499072e-05 2.8155857302499072e-05\n", + "Dual infeasibility......: 2.5902092059556247e-07 2.5902092059556247e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 2.4338671833101106e-08 2.4338671833101106e-08\n", + "Overall NLP error.......: 2.5902092059556247e-07 2.5902092059556247e-07\n", + "\n", + "\n", + "\n", + "\n", + "Number of Iterations....: 16\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 2.8155857302499072e-05 2.8155857302499072e-05\n", + "Dual infeasibility......: 2.5902092059556247e-07 2.5902092059556247e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 2.4338671833101106e-08 2.4338671833101106e-08\n", + "Overall NLP error.......: 2.5902092059556247e-07 2.5902092059556247e-07\n", + "\n", + "\n", + "Number of objective function evaluations = 17\n", + "Number of objective gradient evaluations = 17\n", + "Number of equality constraint evaluations = 0\n", + "Number of inequality constraint evaluations = 17\n", + "Number of equality constraint Jacobian evaluations = 0\n", + "Number of inequality constraint Jacobian evaluations = 17\n", + "Number of Lagrangian Hessian evaluations = 0\n", + "Total CPU secs in IPOPT (w/o function evaluations) = 2.123\n", + "Total CPU secs in NLP function evaluations = 1.853\n", + "\n", + "EXIT: Solved To Acceptable Level.\n", + "v=0.5\n", + "w=0.5\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "rm: cannot remove '/home/fenics/shared/templates/RES_OPT/TEMP*': No such file or directory\n", + "WARNING: The requested image's platform (linux/amd64) does not match the detected host platform (linux/arm64/v8) and no specific platform was requested\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 1.748e-16 (tol = 1.000e-07) r (rel) = 1.764e-12 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "\n", + "List of options:\n", + "\n", + " Name Value # times used\n", + " acceptable_tol = 0.001 1\n", + " file_print_level = 5 1\n", + " hessian_approximation = limited-memory 7\n", + " max_iter = 100 1\n", + " output_file = /home/fenics/shared/templates/RES_OPT/solution_V=0.4_w=0.7.txt 1\n", + " print_level = 6 2\n", + "\n", + "******************************************************************************\n", + "This program contains Ipopt, a library for large-scale nonlinear optimization.\n", + " Ipopt is released as open source code under the Eclipse Public License (EPL).\n", + " For more information visit http://projects.coin-or.org/Ipopt\n", + "******************************************************************************\n", + "\n", + "This is Ipopt version 3.12.9, running with linear solver mumps.\n", + "NOTE: Other linear solvers might be more efficient (see Ipopt documentation).\n", + "\n", + "Number of nonzeros in equality constraint Jacobian...: 0\n", + "Number of nonzeros in inequality constraint Jacobian.: 10201\n", + "Number of nonzeros in Lagrangian Hessian.............: 0\n", + "\n", + "Hessian approximation will be done in the space of all 10201 x variables.\n", + "\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Scaling parameter for objective function = 1.000000e+00\n", + "objective scaling factor = 1\n", + "No x scaling provided\n", + "No c scaling provided\n", + "No d scaling provided\n", + "Moved initial values of x sufficiently inside the bounds.\n", + "Initial values of s sufficiently inside the bounds.\n", + "MUMPS used permuting_scaling 5 and pivot_order 5.\n", + " scaling will be 77.\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Least square estimates max(y_c) = 0.000000e+00, max(y_d) = 1.002046e+00\n", + "Total number of variables............................: 10201\n", + " variables with only lower bounds: 0\n", + " variables with lower and upper bounds: 10201\n", + " variables with only upper bounds: 0\n", + "Total number of equality constraints.................: 0\n", + "Total number of inequality constraints...............: 1\n", + " inequality constraints with only lower bounds: 1\n", + " inequality constraints with lower and upper bounds: 0\n", + " inequality constraints with only upper bounds: 0\n", + "\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.589e-17 (tol = 1.000e-07) r (rel) = 4.631e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 0:\n", + "**************************************************\n", + "\n", + "Limited-Memory approximation started; store data at current iterate.\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 0:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 0 4.8704052e-04 0.00e+00 2.22e+00 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 0 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 1.0000000000000000e+00\n", + "Current fraction-to-the-boundary parameter tau = 0.0000000000000000e+00\n", + "\n", + "||curr_x||_inf = 6.2118917703628540e-01\n", + "||curr_s||_inf = 2.4599847527097249e-01\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 1.0020464465022398e+00\n", + "||curr_z_L||_inf = 1.0000000000000000e+00\n", + "||curr_z_U||_inf = 1.0000000000000000e+00\n", + "||curr_v_L||_inf = 1.0000000000000000e+00\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 0:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 4.8704052360352306e-04 4.8704052360352306e-04\n", + "Dual infeasibility......: 2.2174831958962931e+00 2.2174831958962931e+00\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 9.9000001999999998e-01 9.9000001999999998e-01\n", + "Overall NLP error.......: 2.2174831958962931e+00 2.2174831958962931e+00\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 0:\n", + "**************************************************\n", + "\n", + "Setting mu_max to 4.999876e+02.\n", + "Staying in free mu mode.\n", + "The current filter has 1 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 3.360003e-01\n", + "Barrier parameter mu computed by oracle is 1.679960e-01\n", + "Barrier parameter mu after safeguards is 1.679960e-01\n", + "Barrier Parameter: 1.679960e-01\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 0:\n", + "**************************************************\n", + "\n", + "residual_ratio = 2.086259e-14\n", + "Factorization successful.\n", + "residual_ratio = 9.074391e-17\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 0:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 0 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 0.000000E+00\n", + "Starting checks for alpha (primal) = 1.00e+00\n", + "trial_max is initialized to 1.000000e+04\n", + "trial_min is initialized to 1.000000e-04\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.670e-17 (tol = 1.000e-07) r (rel) = 4.713e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 3.0260028813960107e+03 (reference 5.8473694132959017e+03):\n", + " New values of constraint violation = 3.3306690738754696e-16 (reference 0.0000000000000000e+00):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 1:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 1.847599e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 3.600166e+00 snrm = 1.343508e+01 ynrm = 3.386928e+00\n", + " Perform the update.\n", + "sigma (for B0) is 1.994537e-02\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 1:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 1 4.6302980e-04 0.00e+00 7.52e-01 -0.8 1.85e-01 - 3.70e-01 1.00e+00f 1 sigma=3.36e-01 qf=12y \n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 1 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 1.6799595239964976e-01\n", + "Current fraction-to-the-boundary parameter tau = 9.8999999999999999e-01\n", + "\n", + "||curr_x||_inf = 5.8821626494252255e-01\n", + "||curr_s||_inf = 1.3258421981061613e-01\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 1.0943399887739254e+00\n", + "||curr_z_L||_inf = 8.1412312750053450e-01\n", + "||curr_z_U||_inf = 7.6194608145019460e-01\n", + "||curr_v_L||_inf = 1.0532476431928279e+00\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 1:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 4.6302980268635125e-04 4.6302980268635125e-04\n", + "Dual infeasibility......: 7.5201101712860219e-01 7.5201101712860219e-01\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 6.2335954780289182e-01 6.2335954780289182e-01\n", + "Overall NLP error.......: 7.5201101712860219e-01 7.5201101712860219e-01\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 1:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 2 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 1.000000e-06\n", + "Barrier parameter mu computed by oracle is 3.803892e-07\n", + "Barrier parameter mu after safeguards is 3.803892e-07\n", + "Barrier Parameter: 3.803892e-07\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 1:\n", + "**************************************************\n", + "\n", + "residual_ratio = 1.803925e-16\n", + "Factorization successful.\n", + "residual_ratio = 1.202617e-16\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 1:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 1 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 2.776572E-17\n", + "Starting checks for alpha (primal) = 1.00e+00\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.416e-17 (tol = 1.000e-07) r (rel) = 4.457e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 7.3147281184980477e-03 (reference 7.3147341159054812e-03):\n", + " New values of constraint violation = 3.3306690738754696e-16 (reference 3.3306690738754696e-16):\n", + "Checking sufficient reduction...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 2:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 4.170181e-06\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 1.487384e-14 snrm = 3.732204e-05 ynrm = 6.853994e-10\n", + " Perform the update.\n", + "sigma (for B0) is 1.067806e-05\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 2:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 2 4.6302759e-04 0.00e+00 7.52e-03 -6.4 4.17e-06 - 9.90e-01 1.00e+00h 1 Nhj sigma=1.00e-06 qf=13y \n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 2 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 3.8038918431582506e-07\n", + "Current fraction-to-the-boundary parameter tau = 9.8999999999999999e-01\n", + "\n", + "||curr_x||_inf = 5.8821650715826079e-01\n", + "||curr_s||_inf = 1.3258387518455172e-01\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 1.0956029356307446e-02\n", + "||curr_z_L||_inf = 8.1472283403452028e-03\n", + "||curr_z_U||_inf = 7.6257446059045941e-03\n", + "||curr_v_L||_inf = 1.0544835756068016e-02\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 2:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 4.6302759046657865e-04 4.6302759046657865e-04\n", + "Dual infeasibility......: 7.5227639510524755e-03 7.5227639510524755e-03\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 6.2382984541305289e-03 6.2382984541305289e-03\n", + "Overall NLP error.......: 7.5227639510524755e-03 7.5227639510524755e-03\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 2:\n", + "**************************************************\n", + "\n", + "Switching to fixed mu mode with mu = 3.0453769082853552e-03 and tau = 9.9695462309171468e-01.\n", + "Barrier Parameter: 3.045377e-03\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 2:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 1.109790e-12\n", + "Factorization successful.\n", + "residual_ratio = 9.304854e-16\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 2:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 2 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 7.063344E-27\n", + "Starting checks for alpha (primal) = 7.46e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.392e-17 (tol = 1.000e-07) r (rel) = 4.432e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 7.463167e-01:\n", + " New values of barrier function = 4.4874493915599857e+01 (reference 5.4854836997459245e+01):\n", + " New values of constraint violation = 9.1593399531575415e-16 (reference 3.3306690738754696e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 3:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 3.450118e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = -1.481102e-04 snrm = 1.559662e+01 ynrm = 3.828571e-05\n", + " Skip the update.\n", + "Number of successive iterations with skipping: 1\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 3:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 3 3.1715370e-04 0.00e+00 7.23e-03 -2.5 4.62e-01 - 1.00e+00 7.46e-01f 1 Fy Ws\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 3 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 3.0453769082853552e-03\n", + "Current fraction-to-the-boundary parameter tau = 9.9695462309171468e-01\n", + "\n", + "||curr_x||_inf = 5.3977095189069146e-01\n", + "||curr_s||_inf = 4.0375790235178122e-04\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 3.7060376851556567e-02\n", + "||curr_z_L||_inf = 1.5394438715324700e-02\n", + "||curr_z_U||_inf = 8.1587009855454244e-03\n", + "||curr_v_L||_inf = 3.7055577302987197e-02\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 3:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 3.1715369999061113e-04 3.1715369999061113e-04\n", + "Dual infeasibility......: 7.2320863508398957e-03 7.2320863508398957e-03\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 8.3094709931381145e-03 8.3094709931381145e-03\n", + "Overall NLP error.......: 8.3094709931381145e-03 8.3094709931381145e-03\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 3:\n", + "**************************************************\n", + "\n", + "Switching back to free mu mode.\n", + "The current filter has 3 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 1.000000e-06\n", + "Barrier parameter mu computed by oracle is 2.996411e-09\n", + "Barrier parameter mu after safeguards is 2.996411e-09\n", + "Barrier Parameter: 2.996411e-09\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 3:\n", + "**************************************************\n", + "\n", + "residual_ratio = 1.949754e-16\n", + "Factorization successful.\n", + "residual_ratio = 3.899507e-17\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 3:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 3 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 4.558255E-18\n", + "Starting checks for alpha (primal) = 1.00e+00\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.283e-17 (tol = 1.000e-07) r (rel) = 4.323e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 3.6120591857759140e-04 (reference 3.6130635951824426e-04):\n", + " New values of constraint violation = 1.9130747333506726e-16 (reference 9.1593399531575415e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 4:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 4.081172e-04\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 5.237898e-11 snrm = 1.938419e-03 ynrm = 3.488610e-08\n", + " Perform the update.\n", + "sigma (for B0) is 1.393997e-05\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 4:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 4 3.1705327e-04 0.00e+00 6.54e-05 -8.5 4.08e-04 - 9.91e-01 1.00e+00f 1 sigma=1.00e-06 qf=13y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 4 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 2.9964112874601644e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9169052900686183e-01\n", + "\n", + "||curr_x||_inf = 5.3977579284508148e-01\n", + "||curr_s||_inf = 3.9420500698697812e-04\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 1.2124954685401744e-03\n", + "||curr_z_L||_inf = 1.3960018092507714e-04\n", + "||curr_z_U||_inf = 7.4145990765397962e-05\n", + "||curr_v_L||_inf = 1.2124520516770196e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 4:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 3.1705326771341679e-04 3.1705326771341679e-04\n", + "Dual infeasibility......: 6.5387572558797468e-05 6.5387572558797468e-05\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 7.5352799736152159e-05 7.5352799736152159e-05\n", + "Overall NLP error.......: 7.5352799736152159e-05 7.5352799736152159e-05\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 4:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 2 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 6.014650e-02\n", + "Barrier parameter mu computed by oracle is 1.636026e-06\n", + "Barrier parameter mu after safeguards is 1.636026e-06\n", + "Barrier Parameter: 1.636026e-06\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 4:\n", + "**************************************************\n", + "\n", + "residual_ratio = 4.714859e-15\n", + "Factorization successful.\n", + "residual_ratio = 3.258087e-16\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 4:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 4 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.766649E-21\n", + "Starting checks for alpha (primal) = 1.61e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.186e-17 (tol = 1.000e-07) r (rel) = 4.225e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.607539e-01:\n", + " New values of barrier function = 2.4429408667047930e-02 (reference 2.4424185715987642e-02):\n", + " New values of constraint violation = 1.6019087098473328e-16 (reference 1.9130747333506726e-16):\n", + "Checking Armijo Condition...\n", + "Failed...\n", + "Starting checks for alpha (primal) = 8.04e-02\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.186e-17 (tol = 1.000e-07) r (rel) = 4.225e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 8.037697e-02:\n", + " New values of barrier function = 2.4420157533596547e-02 (reference 2.4424185715987642e-02):\n", + " New values of constraint violation = 2.0350474777552918e-16 (reference 1.9130747333506726e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 5:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 2.668800e-03\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 7.955388e-11 snrm = 3.118260e-02 ynrm = 2.347997e-07\n", + " Perform the update.\n", + "sigma (for B0) is 8.181570e-08\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 5:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 5 3.1663452e-04 0.00e+00 3.68e-06 -5.8 3.32e-02 - 9.99e-01 8.04e-02f 2 sigma=6.01e-02 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 5 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 1.6360259772062144e-06\n", + "Current fraction-to-the-boundary parameter tau = 9.9992464720026386e-01\n", + "\n", + "||curr_x||_inf = 5.3877389584409507e-01\n", + "||curr_s||_inf = 1.9711235609572630e-04\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 1.1678020670160476e-02\n", + "||curr_z_L||_inf = 6.4301367485069725e-06\n", + "||curr_z_U||_inf = 7.8028122330760423e-06\n", + "||curr_v_L||_inf = 1.1678015555874600e-02\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 5:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 3.1663452449653268e-04 3.1663452449653268e-04\n", + "Dual infeasibility......: 3.6751614129219837e-06 3.6751614129219837e-06\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 3.6711435487567861e-06 3.6711435487567861e-06\n", + "Overall NLP error.......: 3.6751614129219837e-06 3.6751614129219837e-06\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 5:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 3 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 3.149303e-01\n", + "Barrier parameter mu computed by oracle is 5.259011e-07\n", + "Barrier parameter mu after safeguards is 5.259011e-07\n", + "Barrier Parameter: 5.259011e-07\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 5:\n", + "**************************************************\n", + "\n", + "residual_ratio = 1.553300e-11\n", + "Factorization successful.\n", + "residual_ratio = 2.322437e-16\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 5:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 5 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.994040E-21\n", + "Starting checks for alpha (primal) = 1.00e+00\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.224e-17 (tol = 1.000e-07) r (rel) = 3.254e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 8.0257073397595642e-03 (reference 8.0647200524303787e-03):\n", + " New values of constraint violation = 1.2293497383270013e-16 (reference 2.0350474777552918e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 6:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 4.056222e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 1.034491e-05 snrm = 1.877164e+00 ynrm = 1.846312e-05\n", + " Perform the update.\n", + "sigma (for B0) is 2.935772e-06\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 6:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 6 2.8884142e-04 0.00e+00 1.73e-06 -6.3 4.06e-01 - 9.96e-01 1.00e+00f 1 sigma=3.15e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 6 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 5.2590109721842701e-07\n", + "Current fraction-to-the-boundary parameter tau = 9.9999632483858703e-01\n", + "\n", + "||curr_x||_inf = 9.2410030777432806e-01\n", + "||curr_s||_inf = 1.1569533729284346e-04\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 7.5086990403476815e-03\n", + "||curr_z_L||_inf = 1.5505001896117894e-06\n", + "||curr_z_U||_inf = 7.3897621872819613e-06\n", + "||curr_v_L||_inf = 7.5086985370896699e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 6:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 2.8884142183483510e-04 2.8884142183483510e-04\n", + "Dual infeasibility......: 1.7308522604732651e-06 1.7308522604732651e-06\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 1.1092600152643912e-06 1.1092600152643912e-06\n", + "Overall NLP error.......: 1.7308522604732651e-06 1.7308522604732651e-06\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 6:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 2 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 3.911818e-01\n", + "Barrier parameter mu computed by oracle is 2.075177e-07\n", + "Barrier parameter mu after safeguards is 2.075177e-07\n", + "Barrier Parameter: 2.075177e-07\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 6:\n", + "**************************************************\n", + "\n", + "residual_ratio = 2.660681e-11\n", + "Factorization successful.\n", + "residual_ratio = 1.203473e-15\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 6:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 6 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.082418E-21\n", + "Starting checks for alpha (primal) = 1.00e+00\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 2.701e-17 (tol = 1.000e-07) r (rel) = 2.726e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 3.3095464179937330e-03 (reference 3.3417658050835710e-03):\n", + " New values of constraint violation = 1.1714804473705875e-16 (reference 1.2293497383270013e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 7:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 2.113738e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 2.950344e-05 snrm = 3.095500e+00 ynrm = 1.474585e-05\n", + " Perform the update.\n", + "sigma (for B0) is 3.079009e-06\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 7:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 7 2.2593574e-04 0.00e+00 1.68e-06 -6.7 2.11e-01 - 9.96e-01 1.00e+00f 1 sigma=3.91e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 7 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 2.0751765636208095e-07\n", + "Current fraction-to-the-boundary parameter tau = 9.9999826914773948e-01\n", + "\n", + "||curr_x||_inf = 9.5005893537771646e-01\n", + "||curr_s||_inf = 7.2930630089901880e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 4.5812734757940694e-03\n", + "||curr_z_L||_inf = 9.4171124529123179e-07\n", + "||curr_z_U||_inf = 5.2706162071997211e-06\n", + "||curr_v_L||_inf = 4.5813200352051407e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 7:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 2.2593574370060362e-04 2.2593574370060362e-04\n", + "Dual infeasibility......: 1.6818748693090317e-06 1.6818748693090317e-06\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 3.6654665539814339e-07 3.6654665539814339e-07\n", + "Overall NLP error.......: 1.6818748693090317e-06 1.6818748693090317e-06\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 7:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 2 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 4.383185e-01\n", + "Barrier parameter mu computed by oracle is 9.181682e-08\n", + "Barrier parameter mu after safeguards is 9.181682e-08\n", + "Barrier Parameter: 9.181682e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 7:\n", + "**************************************************\n", + "\n", + "residual_ratio = 2.266956e-10\n", + "Factorization successful.\n", + "residual_ratio = 4.493891e-16\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 7:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 7 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.466112E-21\n", + "Starting checks for alpha (primal) = 8.43e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.039e-17 (tol = 1.000e-07) r (rel) = 3.067e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 8.431664e-01:\n", + " New values of barrier function = 1.5661255970296102e-03 (reference 1.5902886476522439e-03):\n", + " New values of constraint violation = 2.1648806879104310e-16 (reference 1.1714804473705875e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 8:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 2.569186e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 9.665151e-06 snrm = 2.720363e+00 ynrm = 8.300661e-06\n", + " Perform the update.\n", + "sigma (for B0) is 1.306036e-06\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 8:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 8 1.8201417e-04 0.00e+00 3.04e-06 -7.0 3.05e-01 - 9.93e-01 8.43e-01f 1 sigma=4.38e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 8 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 9.1816817031789006e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999831812513074e-01\n", + "\n", + "||curr_x||_inf = 9.9999975562499555e-01\n", + "||curr_s||_inf = 5.3730159520642967e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 2.7025554934629560e-03\n", + "||curr_z_L||_inf = 9.9375752145242135e-07\n", + "||curr_z_U||_inf = 5.9307961397274790e-06\n", + "||curr_v_L||_inf = 2.7025701585926547e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 8:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.8201416983233039e-04 1.8201416983233039e-04\n", + "Dual infeasibility......: 3.0360774061120197e-06 3.0360774061120197e-06\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 1.4790706677325193e-07 1.4790706677325193e-07\n", + "Overall NLP error.......: 3.0360774061120197e-06 3.0360774061120197e-06\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 8:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 3 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 5.015646e-01\n", + "Barrier parameter mu computed by oracle is 4.710586e-08\n", + "Barrier parameter mu after safeguards is 4.710586e-08\n", + "Barrier Parameter: 4.710586e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 8:\n", + "**************************************************\n", + "\n", + "residual_ratio = 8.822559e-16\n", + "Factorization successful.\n", + "residual_ratio = 1.485100e-17\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 8:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 8 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 4.205836E-23\n", + "Starting checks for alpha (primal) = 3.87e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.545e-17 (tol = 1.000e-07) r (rel) = 3.578e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 3.871788e-01:\n", + " New values of barrier function = 8.7178547165690443e-04 (reference 8.9212103618765590e-04):\n", + " New values of constraint violation = 4.1046538997585591e-16 (reference 2.1648806879104310e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 9:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 2.293077e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 4.780517e-06 snrm = 2.350285e+00 ynrm = 4.877235e-06\n", + " Perform the update.\n", + "sigma (for B0) is 8.654337e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 9:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 9 1.5157112e-04 0.00e+00 1.69e-06 -7.3 5.92e-01 - 7.61e-01 3.87e-01f 1 sigma=5.02e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 9 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 4.7105855020060422e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999696392259385e-01\n", + "\n", + "||curr_x||_inf = 9.9474954502232282e-01\n", + "||curr_s||_inf = 4.7358888572468822e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 1.9423310763908390e-03\n", + "||curr_z_L||_inf = 1.4602105058149585e-06\n", + "||curr_z_U||_inf = 2.1675375657906144e-06\n", + "||curr_v_L||_inf = 1.9423358598698007e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 9:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.5157111954755266e-04 1.5157111954755266e-04\n", + "Dual infeasibility......: 1.6867106147978903e-06 1.6867106147978903e-06\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 1.0660132784860406e-07 1.0660132784860406e-07\n", + "Overall NLP error.......: 1.6867106147978903e-06 1.6867106147978903e-06\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 9:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 4 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 6.119179e-01\n", + "Barrier parameter mu computed by oracle is 3.638545e-08\n", + "Barrier parameter mu after safeguards is 3.638545e-08\n", + "Barrier Parameter: 3.638545e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 9:\n", + "**************************************************\n", + "\n", + "residual_ratio = 3.686412e-16\n", + "Factorization successful.\n", + "residual_ratio = 3.100413e-18\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 9:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 9 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.294812E-22\n", + "Starting checks for alpha (primal) = 5.55e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.134e-17 (tol = 1.000e-07) r (rel) = 4.172e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 5.552074e-01:\n", + " New values of barrier function = 6.8955753836830690e-04 (reference 7.0787819500982025e-04):\n", + " New values of constraint violation = 2.0771280745748855e-16 (reference 4.1046538997585591e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 10:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 4.031716e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 7.275344e-06 snrm = 3.216928e+00 ynrm = 4.980256e-06\n", + " Perform the update.\n", + "sigma (for B0) is 7.030250e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 10:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 10 1.2046287e-04 0.00e+00 2.91e-06 -7.4 7.26e-01 - 9.12e-01 5.55e-01f 1 sigma=6.12e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 10 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 3.6385446036460799e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999831328938515e-01\n", + "\n", + "||curr_x||_inf = 9.9348495109050994e-01\n", + "||curr_s||_inf = 3.9850597270977183e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 1.3773791778169796e-03\n", + "||curr_z_L||_inf = 2.0646667776405924e-06\n", + "||curr_z_U||_inf = 3.5229788276053282e-06\n", + "||curr_v_L||_inf = 1.3773814927461134e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 10:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.2046287079371508e-04 1.2046287079371508e-04\n", + "Dual infeasibility......: 2.9058395063124926e-06 2.9058395063124926e-06\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 6.6000103088276991e-08 6.6000103088276991e-08\n", + "Overall NLP error.......: 2.9058395063124926e-06 2.9058395063124926e-06\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 10:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 3 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 5.096845e-01\n", + "Barrier parameter mu computed by oracle is 2.007689e-08\n", + "Barrier parameter mu after safeguards is 2.007689e-08\n", + "Barrier Parameter: 2.007689e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 10:\n", + "**************************************************\n", + "\n", + "residual_ratio = 3.144128e-16\n", + "Factorization successful.\n", + "residual_ratio = 2.221641e-18\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 10:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 10 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 7.677864E-23\n", + "Starting checks for alpha (primal) = 2.89e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.506e-17 (tol = 1.000e-07) r (rel) = 4.547e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 2.891424e-01:\n", + " New values of barrier function = 4.2720975148139069e-04 (reference 4.3447995549844142e-04):\n", + " New values of constraint violation = 2.7057620467091370e-17 (reference 2.0771280745748855e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 11:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 1.889501e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 1.031849e-06 snrm = 1.455994e+00 ynrm = 1.976844e-06\n", + " Perform the update.\n", + "sigma (for B0) is 4.867402e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 11:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 11 1.0972743e-04 0.00e+00 1.53e-06 -7.7 6.53e-01 - 9.96e-01 2.89e-01f 1 sigma=5.10e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 11 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 2.0076891141407482e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999709416049365e-01\n", + "\n", + "||curr_x||_inf = 9.9999980896654761e-01\n", + "||curr_s||_inf = 3.6187122950754467e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 9.4313123085838183e-04\n", + "||curr_z_L||_inf = 1.6008460191352999e-06\n", + "||curr_z_U||_inf = 3.2752470267311920e-06\n", + "||curr_v_L||_inf = 9.4311930186213508e-04\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 11:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.0972742596878089e-04 1.0972742596878089e-04\n", + "Dual infeasibility......: 1.5293804154460305e-06 1.5293804154460305e-06\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 5.7418659357479277e-08 5.7418659357479277e-08\n", + "Overall NLP error.......: 1.5293804154460305e-06 1.5293804154460305e-06\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 11:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 2 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 9.900000e-01\n", + "Barrier parameter mu computed by oracle is 2.112851e-08\n", + "Barrier parameter mu after safeguards is 2.112851e-08\n", + "Barrier Parameter: 2.112851e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 11:\n", + "**************************************************\n", + "\n", + "residual_ratio = 2.759637e-16\n", + "Factorization successful.\n", + "residual_ratio = 1.099384e-17\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 11:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 11 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 5.593925E-24\n", + "Starting checks for alpha (primal) = 5.54e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.900e-17 (tol = 1.000e-07) r (rel) = 4.945e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 5.536510e-01:\n", + " New values of barrier function = 4.3172665918063944e-04 (reference 4.4383941446602901e-04):\n", + " New values of constraint violation = 5.6500485713650850e-17 (reference 2.7057620467091370e-17):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 12:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 3.147200e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 5.351565e-06 snrm = 3.830228e+00 ynrm = 3.454497e-06\n", + " Perform the update.\n", + "sigma (for B0) is 3.647805e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 12:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 12 8.7777014e-05 0.00e+00 1.93e-06 -7.7 5.68e-01 - 9.39e-01 5.54e-01f 1 sigma=9.90e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 12 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 2.1128514827613649e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999847061958458e-01\n", + "\n", + "||curr_x||_inf = 9.9269074254289646e-01\n", + "||curr_s||_inf = 3.1125173758473780e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 8.2925995310985465e-04\n", + "||curr_z_L||_inf = 5.5510583861884247e-07\n", + "||curr_z_U||_inf = 2.5162498793861936e-06\n", + "||curr_v_L||_inf = 8.2926183453545103e-04\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 12:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 8.7777014205619642e-05 8.7777014205619642e-05\n", + "Dual infeasibility......: 1.9310599295723908e-06 1.9310599295723908e-06\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 3.6625901774115287e-08 3.6625901774115287e-08\n", + "Overall NLP error.......: 1.9310599295723908e-06 1.9310599295723908e-06\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 12:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 3 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 4.075420e-01\n", + "Barrier parameter mu computed by oracle is 8.789274e-09\n", + "Barrier parameter mu after safeguards is 8.789274e-09\n", + "Barrier Parameter: 8.789274e-09\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 12:\n", + "**************************************************\n", + "\n", + "residual_ratio = 3.389405e-16\n", + "Factorization successful.\n", + "residual_ratio = 2.069211e-17\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 12:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 12 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 5.809344E-23\n", + "Starting checks for alpha (primal) = 4.29e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.925e-17 (tol = 1.000e-07) r (rel) = 4.970e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 4.287589e-01:\n", + " New values of barrier function = 2.2236974805033534e-04 (reference 2.3085699915389052e-04):\n", + " New values of constraint violation = 2.1632882659695929e-16 (reference 5.6500485713650850e-17):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 13:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 1.542398e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 2.315770e-06 snrm = 2.738706e+00 ynrm = 2.016783e-06\n", + " Perform the update.\n", + "sigma (for B0) is 3.087484e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 13:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 13 7.5639338e-05 0.00e+00 2.93e-07 -8.1 3.60e-01 - 7.50e-01 4.29e-01f 1 sigma=4.08e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 13 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 8.7892737430615351e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999806894007037e-01\n", + "\n", + "||curr_x||_inf = 9.9999998101301035e-01\n", + "||curr_s||_inf = 2.8253250421433731e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 5.5271654158366276e-04\n", + "||curr_z_L||_inf = 1.5806351019790904e-07\n", + "||curr_z_U||_inf = 3.1585860841021594e-06\n", + "||curr_v_L||_inf = 5.5271370496960204e-04\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 13:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 7.5639338012802059e-05 7.5639338012802059e-05\n", + "Dual infeasibility......: 2.9308063523552921e-07 2.9308063523552921e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 2.4306152807359659e-08 2.4306152807359659e-08\n", + "Overall NLP error.......: 2.9308063523552921e-07 2.9308063523552921e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 13:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 4 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 1.922145e-01\n", + "Barrier parameter mu computed by oracle is 2.370645e-09\n", + "Barrier parameter mu after safeguards is 2.370645e-09\n", + "Barrier Parameter: 2.370645e-09\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 13:\n", + "**************************************************\n", + "\n", + "residual_ratio = 7.921762e-17\n", + "Factorization successful.\n", + "residual_ratio = 3.217258e-17\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 13:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 13 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.004053E-21\n", + "Starting checks for alpha (primal) = 7.02e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.543e-17 (tol = 1.000e-07) r (rel) = 5.594e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 7.024951e-01:\n", + " New values of barrier function = 1.0173332444787992e-04 (reference 1.1521550481585263e-04):\n", + " New values of constraint violation = 2.0838365755171395e-16 (reference 2.1632882659695929e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 14:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 3.966060e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 6.358785e-06 snrm = 5.288576e+00 ynrm = 2.597870e-06\n", + " Perform the update.\n", + "sigma (for B0) is 2.273509e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 14:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 14 5.9500382e-05 0.00e+00 2.94e-07 -8.6 5.65e-01 - 4.18e-01 7.02e-01f 1 sigma=1.92e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 14 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 2.3706453464152239e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999970691936479e-01\n", + "\n", + "||curr_x||_inf = 9.9999995557784849e-01\n", + "||curr_s||_inf = 2.1014804531052428e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 4.4102964736513433e-04\n", + "||curr_z_L||_inf = 1.0510144320550275e-07\n", + "||curr_z_U||_inf = 2.8623304964562342e-06\n", + "||curr_v_L||_inf = 4.4103521359022515e-04\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 14:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 5.9500382156498893e-05 5.9500382156498893e-05\n", + "Dual infeasibility......: 2.9390926419814768e-07 2.9390926419814768e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 1.4419656232729822e-08 1.4419656232729822e-08\n", + "Overall NLP error.......: 2.9390926419814768e-07 2.9390926419814768e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 14:\n", + "**************************************************\n", + "\n", + "Staying in free mu mode.\n", + "The current filter has 4 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 1.000000e+02\n", + "Barrier parameter mu computed by oracle is 7.926865e-07\n", + "Barrier parameter mu after safeguards is 7.926865e-07\n", + "Barrier Parameter: 7.926865e-07\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 14:\n", + "**************************************************\n", + "\n", + "residual_ratio = 1.323454e-14\n", + "Factorization successful.\n", + "residual_ratio = 1.115951e-15\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 14:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 14 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.020983E-27\n", + "Starting checks for alpha (primal) = 8.96e-03\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.367e-17 (tol = 1.000e-07) r (rel) = 5.416e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 8.962402e-03:\n", + " New values of barrier function = 1.3459815043010175e-02 (reference 1.4181175584509374e-02):\n", + " New values of constraint violation = 1.3408192741856673e-16 (reference 2.0838365755171395e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 15:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 9.655825e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 2.737012e-06 snrm = 4.103109e+00 ynrm = 2.011375e-06\n", + " Perform the update.\n", + "sigma (for B0) is 1.625738e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 15:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 15 7.0708911e-05 0.00e+00 7.88e-07 -6.1 1.08e+02 - 1.60e-02 8.96e-03f 1 sigma=1.00e+02 qf=13y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 15 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 7.9268650929104817e-07\n", + "Current fraction-to-the-boundary parameter tau = 9.9999970609073585e-01\n", + "\n", + "||curr_x||_inf = 9.9999972620626976e-01\n", + "||curr_s||_inf = 2.7530631909999081e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 7.9264470760534054e-04\n", + "||curr_z_L||_inf = 3.6793286125936317e-07\n", + "||curr_z_U||_inf = 3.3840413322241640e-06\n", + "||curr_v_L||_inf = 7.9266156458749314e-04\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 15:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 7.0708911268401221e-05 7.0708911268401221e-05\n", + "Dual infeasibility......: 7.8827386556891253e-07 7.8827386556891253e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 3.6793276420098123e-07 3.6793276420098123e-07\n", + "Overall NLP error.......: 7.8827386556891253e-07 7.8827386556891253e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 15:\n", + "**************************************************\n", + "\n", + "Switching to fixed mu mode with mu = 1.6260778144823957e-08 and tau = 9.9999998373922183e-01.\n", + "Barrier Parameter: 1.626078e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 15:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 1.123118e-17\n", + "Factorization successful.\n", + "residual_ratio = 1.778592e-18\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 15:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 15 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.925856E-24\n", + "Starting checks for alpha (primal) = 2.81e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.355e-17 (tol = 1.000e-07) r (rel) = 5.404e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 2.808779e-01:\n", + " New values of barrier function = 3.4492840146736439e-04 (reference 3.4536639794680328e-04):\n", + " New values of constraint violation = 1.0561145599545518e-16 (reference 1.3408192741856673e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 16:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 1.941760e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 2.027273e-08 snrm = 6.169946e-01 ynrm = 3.586992e-07\n", + " Perform the update.\n", + "sigma (for B0) is 5.325366e-08\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 16:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 16 7.0037779e-05 0.00e+00 3.03e-07 -7.8 6.91e-01 - 6.55e-01 2.81e-01f 1 Fy A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 16 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 1.6260778144823957e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999998373922183e-01\n", + "\n", + "||curr_x||_inf = 1.0000000068425474e+00\n", + "||curr_s||_inf = 2.7298053821905725e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 6.7582362595528574e-04\n", + "||curr_z_L||_inf = 2.8401168782898505e-07\n", + "||curr_z_U||_inf = 3.0585395471179582e-06\n", + "||curr_v_L||_inf = 6.7583214759629111e-04\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 16:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 7.0037778623259896e-05 7.0037778623259896e-05\n", + "Dual infeasibility......: 3.0266326294118364e-07 3.0266326294118364e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 2.3554896737735246e-07 2.3554896737735246e-07\n", + "Overall NLP error.......: 3.0266326294118364e-07 3.0266326294118364e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 16:\n", + "**************************************************\n", + "\n", + "Remaining in fixed mu mode.\n", + "Barrier Parameter: 1.626078e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 16:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 2.179247e-15\n", + "Factorization successful.\n", + "residual_ratio = 6.802861e-18\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 16:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 16 <--\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 3.458089E-26\n", + "Starting checks for alpha (primal) = 2.95e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.476e-17 (tol = 1.000e-07) r (rel) = 5.526e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 2.945346e-01:\n", + " New values of barrier function = 3.4393488656618326e-04 (reference 3.4492840146736439e-04):\n", + " New values of constraint violation = 5.8180999081003382e-17 (reference 1.0561145599545518e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 17:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 8.766042e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = -2.999276e-07 snrm = 1.384476e+00 ynrm = 1.216035e-06\n", + " Skip the update.\n", + "Number of successive iterations with skipping: 1\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 17:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 17 6.8969782e-05 0.00e+00 7.87e-07 -7.8 2.98e+00 - 6.08e-01 2.95e-01f 1 Fy AWs\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 17 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 1.6260778144823957e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999998373922183e-01\n", + "\n", + "||curr_x||_inf = 9.9999999574573328e-01\n", + "||curr_s||_inf = 2.7209387033486182e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 6.3151349170585919e-04\n", + "||curr_z_L||_inf = 1.0346085029701024e-07\n", + "||curr_z_U||_inf = 2.8973925022172645e-06\n", + "||curr_v_L||_inf = 6.3151916724826806e-04\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 17:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 6.8969781888316459e-05 6.8969781888316459e-05\n", + "Dual infeasibility......: 7.8709759613971154e-07 7.8709759613971154e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 8.8718473852048702e-08 8.8718473852048702e-08\n", + "Overall NLP error.......: 7.8709759613971154e-07 7.8709759613971154e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 17:\n", + "**************************************************\n", + "\n", + "Remaining in fixed mu mode.\n", + "Barrier Parameter: 1.626078e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 17:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 4.540383e-16\n", + "Factorization successful.\n", + "residual_ratio = 2.547272e-19\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 17:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 17 <--\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 9.972109E-26\n", + "Starting checks for alpha (primal) = 2.80e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.086e-17 (tol = 1.000e-07) r (rel) = 5.133e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 2.804694e-01:\n", + " New values of barrier function = 3.4339948909169399e-04 (reference 3.4393488656618326e-04):\n", + " New values of constraint violation = 1.6249818873305399e-16 (reference 5.8180999081003382e-17):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 18:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 18 6.8317630e-05 0.00e+00 3.97e-07 -7.8 1.44e+00 - 1.00e+00 2.80e-01f 1 Fy A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 18 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 1.6260778144823957e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999998373922183e-01\n", + "\n", + "||curr_x||_inf = 1.0000000053109623e+00\n", + "||curr_s||_inf = 2.7198896670573142e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 5.9826160939990803e-04\n", + "||curr_z_L||_inf = 7.9325355294146254e-08\n", + "||curr_z_U||_inf = 2.2418829706615098e-06\n", + "||curr_v_L||_inf = 5.9826471336189293e-04\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 18:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 6.8317630446442925e-05 6.8317630446442925e-05\n", + "Dual infeasibility......: 3.9676785277369727e-07 3.9676785277369727e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 4.5913924324028463e-08 4.5913924324028463e-08\n", + "Overall NLP error.......: 3.9676785277369727e-07 3.9676785277369727e-07\n", + "\n", + "\n", + "\n", + "\n", + "Number of Iterations....: 18\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 6.8317630446442925e-05 6.8317630446442925e-05\n", + "Dual infeasibility......: 3.9676785277369727e-07 3.9676785277369727e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 4.5913924324028463e-08 4.5913924324028463e-08\n", + "Overall NLP error.......: 3.9676785277369727e-07 3.9676785277369727e-07\n", + "\n", + "\n", + "Number of objective function evaluations = 20\n", + "Number of objective gradient evaluations = 19\n", + "Number of equality constraint evaluations = 0\n", + "Number of inequality constraint evaluations = 20\n", + "Number of equality constraint Jacobian evaluations = 0\n", + "Number of inequality constraint Jacobian evaluations = 19\n", + "Number of Lagrangian Hessian evaluations = 0\n", + "Total CPU secs in IPOPT (w/o function evaluations) = 2.256\n", + "Total CPU secs in NLP function evaluations = 2.173\n", + "\n", + "EXIT: Solved To Acceptable Level.\n", + "v=0.4\n", + "w=0.7\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "rm: cannot remove '/home/fenics/shared/templates/RES_OPT/TEMP*': No such file or directory\n", + "WARNING: The requested image's platform (linux/amd64) does not match the detected host platform (linux/arm64/v8) and no specific platform was requested\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.825e-17 (tol = 1.000e-07) r (rel) = 4.873e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "\n", + "List of options:\n", + "\n", + " Name Value # times used\n", + " acceptable_tol = 0.001 1\n", + " file_print_level = 5 1\n", + " hessian_approximation = limited-memory 7\n", + " max_iter = 100 1\n", + " output_file = /home/fenics/shared/templates/RES_OPT/solution_V=0.2_w=0.1.txt 1\n", + " print_level = 6 2\n", + "\n", + "******************************************************************************\n", + "This program contains Ipopt, a library for large-scale nonlinear optimization.\n", + " Ipopt is released as open source code under the Eclipse Public License (EPL).\n", + " For more information visit http://projects.coin-or.org/Ipopt\n", + "******************************************************************************\n", + "\n", + "This is Ipopt version 3.12.9, running with linear solver mumps.\n", + "NOTE: Other linear solvers might be more efficient (see Ipopt documentation).\n", + "\n", + "Number of nonzeros in equality constraint Jacobian...: 0\n", + "Number of nonzeros in inequality constraint Jacobian.: 10201\n", + "Number of nonzeros in Lagrangian Hessian.............: 0\n", + "\n", + "Hessian approximation will be done in the space of all 10201 x variables.\n", + "\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Scaling parameter for objective function = 1.000000e+00\n", + "objective scaling factor = 1\n", + "No x scaling provided\n", + "No c scaling provided\n", + "No d scaling provided\n", + "Moved initial values of x sufficiently inside the bounds.\n", + "Initial values of s sufficiently inside the bounds.\n", + "MUMPS used permuting_scaling 5 and pivot_order 5.\n", + " scaling will be 77.\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Least square estimates max(y_c) = 0.000000e+00, max(y_d) = 9.999011e-01\n", + "Total number of variables............................: 10201\n", + " variables with only lower bounds: 0\n", + " variables with lower and upper bounds: 10201\n", + " variables with only upper bounds: 0\n", + "Total number of equality constraints.................: 0\n", + "Total number of inequality constraints...............: 1\n", + " inequality constraints with only lower bounds: 1\n", + " inequality constraints with lower and upper bounds: 0\n", + " inequality constraints with only upper bounds: 0\n", + "\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.900e-17 (tol = 1.000e-07) r (rel) = 4.949e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 0:\n", + "**************************************************\n", + "\n", + "Limited-Memory approximation started; store data at current iterate.\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 0:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 0 1.4705639e-04 0.00e+00 1.00e-04 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 0 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 1.0000000000000000e+00\n", + "Current fraction-to-the-boundary parameter tau = 0.0000000000000000e+00\n", + "\n", + "||curr_x||_inf = 9.2386388778686523e-01\n", + "||curr_s||_inf = 2.2181296172274489e-02\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 9.9990110040229219e-01\n", + "||curr_z_L||_inf = 1.0000000000000000e+00\n", + "||curr_z_U||_inf = 1.0000000000000000e+00\n", + "||curr_v_L||_inf = 1.0000000000000000e+00\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 0:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.4705639051276883e-04 1.4705639051276883e-04\n", + "Dual infeasibility......: 1.0002114978358012e-04 1.0002114978358012e-04\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 9.9000001999999998e-01 9.9000001999999998e-01\n", + "Overall NLP error.......: 9.9000001999999998e-01 9.9000001999999998e-01\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 0:\n", + "**************************************************\n", + "\n", + "Setting mu_max to 4.999766e+02.\n", + "Staying in free mu mode.\n", + "The current filter has 1 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 1.000000e-06\n", + "Barrier parameter mu computed by oracle is 4.999766e-07\n", + "Barrier parameter mu after safeguards is 4.999766e-07\n", + "Barrier Parameter: 4.999766e-07\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 0:\n", + "**************************************************\n", + "\n", + "residual_ratio = 2.231605e-16\n", + "Factorization successful.\n", + "residual_ratio = 1.115803e-16\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 0:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 0 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 0.000000E+00\n", + "Starting checks for alpha (primal) = 1.00e+00\n", + "trial_max is initialized to 1.000000e+04\n", + "trial_min is initialized to 1.000000e-04\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.621e-17 (tol = 1.000e-07) r (rel) = 4.667e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 1.8794893658976082e-02 (reference 1.8795059259517077e-02):\n", + " New values of constraint violation = 3.1225022567582528e-16 (reference 0.0000000000000000e+00):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 1:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 4.867590e-07\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 5.774963e-11 snrm = 4.170723e-05 ynrm = 2.111060e-06\n", + " Perform the update.\n", + "sigma (for B0) is 3.319912e-02\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 1:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 1 1.4705642e-04 0.00e+00 1.30e-06 -6.3 4.87e-07 - 9.90e-01 1.00e+00f 1 sigma=1.00e-06 qf=13y \n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 1 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 4.9997659095840198e-07\n", + "Current fraction-to-the-boundary parameter tau = 9.8999999999999999e-01\n", + "\n", + "||curr_x||_inf = 9.2386350532892458e-01\n", + "||curr_s||_inf = 2.2180963008126948e-02\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 1.0036028936224520e-02\n", + "||curr_z_L||_inf = 1.0000988632117624e-02\n", + "||curr_z_U||_inf = 1.0001935573413556e-02\n", + "||curr_v_L||_inf = 1.0036701407285520e-02\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 1:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.4705641727446358e-04 1.4705641727446358e-04\n", + "Dual infeasibility......: 1.3010612777680092e-06 1.3010612777680092e-06\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 9.9004935394333589e-03 9.9004935394333589e-03\n", + "Overall NLP error.......: 9.9004935394333589e-03 9.9004935394333589e-03\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 1:\n", + "**************************************************\n", + "\n", + "Switching to fixed mu mode with mu = 4.0000153326789677e-03 and tau = 9.9599998466732098e-01.\n", + "Barrier Parameter: 4.000015e-03\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 1:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 9.798059e-16\n", + "Factorization successful.\n", + "residual_ratio = 2.041262e-17\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 1:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 1 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.500076E-28\n", + "Starting checks for alpha (primal) = 9.14e-02\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.619e-17 (tol = 1.000e-07) r (rel) = 4.665e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 9.135549e-02:\n", + " New values of barrier function = 1.0832927347178341e+02 (reference 1.4919040163438481e+02):\n", + " New values of constraint violation = 5.2388648974499574e-16 (reference 3.1225022567582528e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 2:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 1.236794e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 1.073723e-06 snrm = 3.109402e+00 ynrm = 2.341267e-06\n", + " Perform the update.\n", + "sigma (for B0) is 1.110550e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 2:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 2 1.5706600e-04 0.00e+00 3.77e-02 -2.4 1.35e+00 - 3.52e-01 9.14e-02f 1 FNhj y \n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 2 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 4.0000153326789677e-03\n", + "Current fraction-to-the-boundary parameter tau = 9.9599998466732098e-01\n", + "\n", + "||curr_x||_inf = 8.9279525079999800e-01\n", + "||curr_s||_inf = 8.8714232126248932e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 1.1149088522973198e-01\n", + "||curr_z_L||_inf = 4.6646778568752101e-02\n", + "||curr_z_U||_inf = 1.0915251060010137e-02\n", + "||curr_v_L||_inf = 1.0862810916454405e-01\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 2:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.5706600049314055e-04 1.5706600049314055e-04\n", + "Dual infeasibility......: 3.7717549553117360e-02 3.7717549553117360e-02\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 8.8935856668390118e-03 8.8935856668390118e-03\n", + "Overall NLP error.......: 3.7717549553117360e-02 3.7717549553117360e-02\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 2:\n", + "**************************************************\n", + "\n", + "Remaining in fixed mu mode.\n", + "Reducing mu to 2.5298366740050934e-04 in fixed mu mode. Tau becomes 9.9599998466732098e-01\n", + "Barrier Parameter: 2.529837e-04\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 2:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 2.179391e-16\n", + "Factorization successful.\n", + "residual_ratio = 8.513248e-19\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 2:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 2 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 4.382301E-25\n", + "Starting checks for alpha (primal) = 2.01e-02\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.561e-17 (tol = 1.000e-07) r (rel) = 4.606e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 2.014679e-02:\n", + " New values of barrier function = 6.8406535339280161e+00 (reference 6.8515050922684582e+00):\n", + " New values of constraint violation = 4.5606964385602744e-16 (reference 5.2388648974499574e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 3:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 1.656571e-03\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 1.797414e-09 snrm = 3.777210e-02 ynrm = 7.911232e-08\n", + " Perform the update.\n", + "sigma (for B0) is 1.259813e-06\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 3:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 3 1.5747890e-04 0.00e+00 9.07e-04 -3.6 8.22e-02 - 9.76e-01 2.01e-02f 1 Fy \n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 3 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 2.5298366740050934e-04\n", + "Current fraction-to-the-boundary parameter tau = 9.9599998466732098e-01\n", + "\n", + "||curr_x||_inf = 8.9227696779121723e-01\n", + "||curr_s||_inf = 3.4489828888516596e-07\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 8.0265653362666001e+00\n", + "||curr_z_L||_inf = 2.2214740931772239e-03\n", + "||curr_z_U||_inf = 7.8398010105133926e-04\n", + "||curr_v_L||_inf = 8.0264965095918352e+00\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 3:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.5747890013272979e-04 1.5747890013272979e-04\n", + "Dual infeasibility......: 9.0709423432018673e-04 9.0709423432018673e-04\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 1.3429330132109587e-03 1.3429330132109587e-03\n", + "Overall NLP error.......: 1.3429330132109587e-03 1.3429330132109587e-03\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 3:\n", + "**************************************************\n", + "\n", + "Remaining in fixed mu mode.\n", + "Reducing mu to 4.0238216787501142e-06 in fixed mu mode. Tau becomes 9.9974701633259944e-01\n", + "Barrier Parameter: 4.023822e-06\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 3:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 6.462265e-17\n", + "Factorization successful.\n", + "residual_ratio = 1.362771e-19\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 3:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 3 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.506687E-22\n", + "Starting checks for alpha (primal) = 1.00e+00\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.932e-17 (tol = 1.000e-07) r (rel) = 4.981e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 1.0748875623798022e-01 (reference 1.0895871734549814e-01):\n", + " New values of constraint violation = 4.2181923135908937e-17 (reference 4.5606964385602744e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 4:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 2.257619e-02\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 4.832449e-07 snrm = 4.442187e-01 ynrm = 1.316924e-06\n", + " Perform the update.\n", + "sigma (for B0) is 2.448914e-06\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 4:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 4 1.6383222e-04 0.00e+00 9.51e-08 -5.4 2.26e-02 - 1.00e+00 1.00e+00f 1 Fy A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 4 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 4.0238216787501142e-06\n", + "Current fraction-to-the-boundary parameter tau = 9.9974701633259944e-01\n", + "\n", + "||curr_x||_inf = 8.8364387020362267e-01\n", + "||curr_s||_inf = 8.4303043730527649e-07\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 7.2036774678150187e-02\n", + "||curr_z_L||_inf = 2.7078609281192705e-05\n", + "||curr_z_U||_inf = 2.0240430427977121e-05\n", + "||curr_v_L||_inf = 7.2036774876892196e-02\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 4:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.6383221527421568e-04 1.6383221527421568e-04\n", + "Dual infeasibility......: 9.5059421167469660e-08 9.5059421167469660e-08\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 2.2234823091003274e-05 2.2234823091003274e-05\n", + "Overall NLP error.......: 2.2234823091003274e-05 2.2234823091003274e-05\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 4:\n", + "**************************************************\n", + "\n", + "Remaining in fixed mu mode.\n", + "Reducing mu to 8.0715713319459585e-09 in fixed mu mode. Tau becomes 9.9999597617832126e-01\n", + "Barrier Parameter: 8.071571e-09\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 4:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 7.748155e-15\n", + "Factorization successful.\n", + "residual_ratio = 1.920912e-18\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 4:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 4 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.066023E-20\n", + "Starting checks for alpha (primal) = 1.00e+00\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.842e-17 (tol = 1.000e-07) r (rel) = 4.890e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 3.7719746139176606e-04 (reference 3.7912027948436668e-04):\n", + " New values of constraint violation = 1.2391033227146346e-18 (reference 4.2181923135908937e-17):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 5:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 3.679047e-02\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 8.493101e-08 snrm = 1.758487e-01 ynrm = 6.934237e-07\n", + " Perform the update.\n", + "sigma (for B0) is 2.746554e-06\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 5:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 5 1.6154231e-04 0.00e+00 1.48e-07 -8.1 3.68e-02 - 9.54e-01 1.00e+00f 1 Fy A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 5 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 8.0715713319459585e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999597617832126e-01\n", + "\n", + "||curr_x||_inf = 8.8448176139761503e-01\n", + "||curr_s||_inf = 9.2687821942481365e-07\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 5.5834070059706861e-03\n", + "||curr_z_L||_inf = 1.2369686400773544e-06\n", + "||curr_z_U||_inf = 1.4733644379664898e-06\n", + "||curr_v_L||_inf = 5.5834070033907585e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 5:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.6154230559720372e-04 1.6154230559720372e-04\n", + "Dual infeasibility......: 1.4772270514867860e-07 1.4772270514867860e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 1.0163765993962152e-06 1.0163765993962152e-06\n", + "Overall NLP error.......: 1.0163765993962152e-06 1.0163765993962152e-06\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 5:\n", + "**************************************************\n", + "\n", + "Remaining in fixed mu mode.\n", + "Barrier Parameter: 8.071571e-09\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 5:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 2.354619e-14\n", + "Factorization successful.\n", + "residual_ratio = 2.919430e-18\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 5:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 5 <--\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 2.758936E-23\n", + "Starting checks for alpha (primal) = 1.00e+00\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.407e-17 (tol = 1.000e-07) r (rel) = 4.451e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 3.5869742365835021e-04 (reference 3.7719746139176606e-04):\n", + " New values of constraint violation = 3.7142818253574697e-17 (reference 1.2391033227146346e-18):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 6:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 3.517985e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 7.002066e-06 snrm = 2.227109e+00 ynrm = 5.148725e-06\n", + " Perform the update.\n", + "sigma (for B0) is 1.411702e-06\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 6:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 6 1.3870512e-04 0.00e+00 4.90e-07 -8.1 3.52e-01 - 4.96e-01 1.00e+00f 1 Fy A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 6 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 8.0715713319459585e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999597617832126e-01\n", + "\n", + "||curr_x||_inf = 8.9170465467028626e-01\n", + "||curr_s||_inf = 2.0188616011069191e-06\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 3.8581185243911236e-03\n", + "||curr_z_L||_inf = 6.2329692853758058e-07\n", + "||curr_z_U||_inf = 9.6391865836145831e-07\n", + "||curr_v_L||_inf = 3.8581284486378211e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 6:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.3870512353646771e-04 1.3870512353646771e-04\n", + "Dual infeasibility......: 4.8973753314882331e-07 4.8973753314882331e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 5.1594021882408290e-07 5.1594021882408290e-07\n", + "Overall NLP error.......: 5.1594021882408290e-07 5.1594021882408290e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 6:\n", + "**************************************************\n", + "\n", + "Switching back to free mu mode.\n", + "The current filter has 2 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 3.100209e-01\n", + "Barrier parameter mu computed by oracle is 3.067659e-08\n", + "Barrier parameter mu after safeguards is 3.067659e-08\n", + "Barrier Parameter: 3.067659e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 6:\n", + "**************************************************\n", + "\n", + "residual_ratio = 5.258785e-14\n", + "Factorization successful.\n", + "residual_ratio = 1.152811e-17\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 6:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 6 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 2.219683E-22\n", + "Starting checks for alpha (primal) = 8.54e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.152e-17 (tol = 1.000e-07) r (rel) = 4.193e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 8.539200e-01:\n", + " New values of barrier function = 9.2581153549170883e-04 (reference 9.7480164851010604e-04):\n", + " New values of constraint violation = 9.6876852243368838e-17 (reference 3.7142818253574697e-17):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 7:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 3.127310e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 7.392181e-06 snrm = 3.692675e+00 ynrm = 4.801309e-06\n", + " Perform the update.\n", + "sigma (for B0) is 5.421138e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 7:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 7 1.5711830e-04 0.00e+00 1.14e-06 -7.5 3.66e-01 - 9.59e-01 8.54e-01f 1 sigma=3.10e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 7 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 3.0676586125868578e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999948405978123e-01\n", + "\n", + "||curr_x||_inf = 9.9999993711555135e-01\n", + "||curr_s||_inf = 7.0507958477153509e-06\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 3.9119818195015423e-03\n", + "||curr_z_L||_inf = 4.7219400259507321e-07\n", + "||curr_z_U||_inf = 1.3301917285452153e-06\n", + "||curr_v_L||_inf = 3.9119987123555685e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 7:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.5711830200493092e-04 1.5711830200493092e-04\n", + "Dual infeasibility......: 1.1361686079532655e-06 1.1361686079532655e-06\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 9.8941265082249117e-08 9.8941265082249117e-08\n", + "Overall NLP error.......: 1.1361686079532655e-06 1.1361686079532655e-06\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 7:\n", + "**************************************************\n", + "\n", + "Switching to fixed mu mode with mu = 2.7223188723064175e-08 and tau = 9.9999997277681130e-01.\n", + "Barrier Parameter: 2.722319e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 7:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 6.733096e-16\n", + "Factorization successful.\n", + "residual_ratio = 6.287097e-18\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 7:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 7 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 6.301810E-24\n", + "Starting checks for alpha (primal) = 7.44e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.155e-17 (tol = 1.000e-07) r (rel) = 4.196e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 7.438886e-01:\n", + " New values of barrier function = 8.1515920969987658e-04 (reference 8.3927637785765848e-04):\n", + " New values of constraint violation = 4.1275915519702056e-17 (reference 9.6876852243368838e-17):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 8:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 3.859246e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = -5.991395e-07 snrm = 4.413591e+00 ynrm = 3.490860e-06\n", + " Skip the update.\n", + "Number of successive iterations with skipping: 1\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 8:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 8 1.5437821e-04 0.00e+00 1.01e-06 -7.6 5.19e-01 - 9.12e-01 7.44e-01f 1 Fy AWs\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 8 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 2.7223188723064175e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "\n", + "||curr_x||_inf = 1.0000000013884498e+00\n", + "||curr_s||_inf = 8.4078287509868032e-06\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 2.9391150279237587e-03\n", + "||curr_z_L||_inf = 3.4194439061141347e-07\n", + "||curr_z_U||_inf = 1.1241262938801358e-06\n", + "||curr_v_L||_inf = 2.9391212138849486e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 8:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.5437820644055507e-04 1.5437820644055507e-04\n", + "Dual infeasibility......: 1.0067114366532949e-06 1.0067114366532949e-06\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 4.8520182360606052e-08 4.8520182360606052e-08\n", + "Overall NLP error.......: 1.0067114366532949e-06 1.0067114366532949e-06\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 8:\n", + "**************************************************\n", + "\n", + "Remaining in fixed mu mode.\n", + "Barrier Parameter: 2.722319e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 8:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 1.134608e-15\n", + "Factorization successful.\n", + "residual_ratio = 1.199844e-18\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 8:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 8 <--\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 5.080217E-26\n", + "Starting checks for alpha (primal) = 6.25e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.460e-17 (tol = 1.000e-07) r (rel) = 4.505e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 6.251177e-01:\n", + " New values of barrier function = 8.0443889257491739e-04 (reference 8.1515920969987658e-04):\n", + " New values of constraint violation = 3.9254894907553295e-17 (reference 4.1275915519702056e-17):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 9:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 3.811753e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 2.356081e-06 snrm = 2.798376e+00 ynrm = 3.566785e-06\n", + " Perform the update.\n", + "sigma (for B0) is 3.008694e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 9:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 9 1.5295135e-04 0.00e+00 8.44e-07 -7.6 6.10e-01 - 6.47e-01 6.25e-01f 1 Fy A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 9 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 2.7223188723064175e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "\n", + "||curr_x||_inf = 1.0000000014545833e+00\n", + "||curr_s||_inf = 9.1761617469650529e-06\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 2.8522554071847019e-03\n", + "||curr_z_L||_inf = 3.3900761536175374e-07\n", + "||curr_z_U||_inf = 9.5186430741765041e-07\n", + "||curr_v_L||_inf = 2.8522628154606087e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 9:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.5295135006259956e-04 1.5295135006259956e-04\n", + "Dual infeasibility......: 8.4449396122111269e-07 8.4449396122111269e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 3.4444459929521884e-08 3.4444459929521884e-08\n", + "Overall NLP error.......: 8.4449396122111269e-07 8.4449396122111269e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 9:\n", + "**************************************************\n", + "\n", + "Remaining in fixed mu mode.\n", + "Barrier Parameter: 2.722319e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 9:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 5.377459e-16\n", + "Factorization successful.\n", + "residual_ratio = 5.901217e-18\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 9:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 9 <--\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 3.744540E-26\n", + "Starting checks for alpha (primal) = 6.42e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.802e-17 (tol = 1.000e-07) r (rel) = 4.850e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 6.416571e-01:\n", + " New values of barrier function = 7.9758113511439667e-04 (reference 8.0443889257491739e-04):\n", + " New values of constraint violation = 1.2704138956098898e-16 (reference 3.9254894907553295e-17):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 10:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 4.232701e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 4.041249e-06 snrm = 3.861719e+00 ynrm = 4.579124e-06\n", + " Perform the update.\n", + "sigma (for B0) is 2.709906e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 10:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 10 1.4843773e-04 0.00e+00 2.61e-06 -7.6 6.60e-01 - 9.46e-01 6.42e-01f 1 Fy A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 10 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 2.7223188723064175e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "\n", + "||curr_x||_inf = 1.0000000000589839e+00\n", + "||curr_s||_inf = 9.7743300618451697e-06\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 2.6837173175875088e-03\n", + "||curr_z_L||_inf = 3.2134023460517586e-07\n", + "||curr_z_U||_inf = 5.4028900333374227e-07\n", + "||curr_v_L||_inf = 2.6837267051078649e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 10:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.4843773331278846e-04 1.4843773331278846e-04\n", + "Dual infeasibility......: 2.6085318329257540e-06 2.6085318329257540e-06\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 1.4249962233056421e-07 1.4249962233056421e-07\n", + "Overall NLP error.......: 2.6085318329257540e-06 2.6085318329257540e-06\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 10:\n", + "**************************************************\n", + "\n", + "Remaining in fixed mu mode.\n", + "Barrier Parameter: 2.722319e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 10:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 6.255437e-16\n", + "Factorization successful.\n", + "residual_ratio = 9.740116e-19\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 10:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 10 <--\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.475848E-25\n", + "Starting checks for alpha (primal) = 4.10e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.993e-17 (tol = 1.000e-07) r (rel) = 5.043e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 4.100400e-01:\n", + " New values of barrier function = 7.9513275660310020e-04 (reference 7.9758113511439667e-04):\n", + " New values of constraint violation = 6.6509027018407663e-18 (reference 1.2704138956098898e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 11:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 4.128123e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 2.408930e-06 snrm = 1.492650e+00 ynrm = 3.883168e-06\n", + " Perform the update.\n", + "sigma (for B0) is 1.081205e-06\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 11:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 11 1.4703404e-04 0.00e+00 1.23e-06 -7.6 1.01e+00 - 5.59e-01 4.10e-01f 1 Fy A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 11 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 2.7223188723064175e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "\n", + "||curr_x||_inf = 1.0000000014065347e+00\n", + "||curr_s||_inf = 9.9003405003506122e-06\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 2.6917071891186463e-03\n", + "||curr_z_L||_inf = 2.9938493621947088e-07\n", + "||curr_z_U||_inf = 4.5481733906074267e-07\n", + "||curr_v_L||_inf = 2.6917186962841515e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 11:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.4703403987189042e-04 1.4703403987189042e-04\n", + "Dual infeasibility......: 1.2323063397704064e-06 1.2323063397704064e-06\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 7.5551841170247104e-08 7.5551841170247104e-08\n", + "Overall NLP error.......: 1.2323063397704064e-06 1.2323063397704064e-06\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 11:\n", + "**************************************************\n", + "\n", + "Remaining in fixed mu mode.\n", + "Barrier Parameter: 2.722319e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 11:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 5.749126e-16\n", + "Factorization successful.\n", + "residual_ratio = 1.750567e-18\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 11:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 11 <--\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 7.783225E-27\n", + "Starting checks for alpha (primal) = 6.46e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.634e-17 (tol = 1.000e-07) r (rel) = 4.680e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 6.461774e-01:\n", + " New values of barrier function = 7.9183800201424468e-04 (reference 7.9513275660310020e-04):\n", + " New values of constraint violation = 6.2565241615991640e-17 (reference 6.6509027018407663e-18):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 12:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 3.533840e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 6.292786e-07 snrm = 1.457286e+00 ynrm = 1.834871e-06\n", + " Perform the update.\n", + "sigma (for B0) is 2.963147e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 12:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 12 1.4705477e-04 0.00e+00 7.78e-07 -7.6 5.47e-01 - 7.61e-01 6.46e-01f 1 Fy A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 12 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 2.7223188723064175e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "\n", + "||curr_x||_inf = 1.0000000003797616e+00\n", + "||curr_s||_inf = 1.0216885760132184e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 2.6324848272244804e-03\n", + "||curr_z_L||_inf = 3.0766355124056562e-07\n", + "||curr_z_U||_inf = 6.8432984829646557e-07\n", + "||curr_v_L||_inf = 2.6324945713303931e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 12:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.4705477146526895e-04 1.4705477146526895e-04\n", + "Dual infeasibility......: 7.7832444069418973e-07 7.7832444069418973e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 5.9639062026560725e-08 5.9639062026560725e-08\n", + "Overall NLP error.......: 7.7832444069418973e-07 7.7832444069418973e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 12:\n", + "**************************************************\n", + "\n", + "Remaining in fixed mu mode.\n", + "Barrier Parameter: 2.722319e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 12:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 4.623051e-16\n", + "Factorization successful.\n", + "residual_ratio = 1.885439e-18\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 12:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 12 <--\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 5.895139E-26\n", + "Starting checks for alpha (primal) = 4.60e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.619e-17 (tol = 1.000e-07) r (rel) = 4.665e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 4.595809e-01:\n", + " New values of barrier function = 7.8850096602701091e-04 (reference 7.9183800201424468e-04):\n", + " New values of constraint violation = 1.3349239248727773e-17 (reference 6.2565241615991640e-17):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 13:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 3.850562e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 1.704929e-06 snrm = 2.086038e+00 ynrm = 2.163729e-06\n", + " Perform the update.\n", + "sigma (for B0) is 3.917977e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 13:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 13 1.4939779e-04 0.00e+00 5.16e-07 -7.6 8.38e-01 - 8.94e-01 4.60e-01f 1 Fy A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 13 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 2.7223188723064175e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "\n", + "||curr_x||_inf = 1.0000000022667506e+00\n", + "||curr_s||_inf = 1.0637360960505911e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 2.4481980658413128e-03\n", + "||curr_z_L||_inf = 2.9094526043824776e-07\n", + "||curr_z_U||_inf = 6.6248234625351159e-07\n", + "||curr_v_L||_inf = 2.4482004857085925e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 13:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.4939779200170430e-04 1.4939779200170430e-04\n", + "Dual infeasibility......: 5.1614828927523612e-07 5.1614828927523612e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 8.5439547895813522e-08 8.5439547895813522e-08\n", + "Overall NLP error.......: 5.1614828927523612e-07 5.1614828927523612e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 13:\n", + "**************************************************\n", + "\n", + "Remaining in fixed mu mode.\n", + "Barrier Parameter: 2.722319e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 13:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 1.465565e-15\n", + "Factorization successful.\n", + "residual_ratio = 7.328570e-19\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 13:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 13 <--\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.834115E-26\n", + "Starting checks for alpha (primal) = 1.00e+00\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.924e-17 (tol = 1.000e-07) r (rel) = 4.973e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 7.8610156016912960e-04 (reference 7.8850096602701091e-04):\n", + " New values of constraint violation = 1.1089016532274398e-16 (reference 1.3349239248727773e-17):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 14:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 3.569959e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 1.792604e-06 snrm = 2.688494e+00 ynrm = 2.847658e-06\n", + " Perform the update.\n", + "sigma (for B0) is 2.480082e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 14:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 14 1.5362607e-04 0.00e+00 9.18e-07 -7.6 3.57e-01 - 7.89e-01 1.00e+00f 1 Fy A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 14 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 2.7223188723064175e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "\n", + "||curr_x||_inf = 9.7187915272562730e-01\n", + "||curr_s||_inf = 1.1275987654546383e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 2.4180325983607903e-03\n", + "||curr_z_L||_inf = 2.7759047785231994e-07\n", + "||curr_z_U||_inf = 7.4115007882390933e-07\n", + "||curr_v_L||_inf = 2.4180311918402231e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 14:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.5362606843249261e-04 1.5362606843249261e-04\n", + "Dual infeasibility......: 9.1812912389470822e-07 9.1812912389470822e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 4.2377847446537874e-08 4.2377847446537874e-08\n", + "Overall NLP error.......: 9.1812912389470822e-07 9.1812912389470822e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 14:\n", + "**************************************************\n", + "\n", + "Remaining in fixed mu mode.\n", + "Barrier Parameter: 2.722319e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 14:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 5.046691e-17\n", + "Factorization successful.\n", + "residual_ratio = 1.136027e-18\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 14:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 14 <--\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 1.776198E-20\n", + "Starting checks for alpha (primal) = 1.00e+00\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.670e-17 (tol = 1.000e-07) r (rel) = 4.716e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 7.8481025801009109e-04 (reference 7.8610156016912960e-04):\n", + " New values of constraint violation = 1.8214427091167024e-16 (reference 1.1089016532274398e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 15:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 3.940773e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 1.671400e-06 snrm = 1.474771e+00 ynrm = 2.364038e-06\n", + " Perform the update.\n", + "sigma (for B0) is 7.684777e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 15:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 15 1.5476349e-04 0.00e+00 5.01e-07 -7.6 3.94e-01 - 1.00e+00 1.00e+00f 1 Fy A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 15 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 2.7223188723064175e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "\n", + "||curr_x||_inf = 9.5330070549010015e-01\n", + "||curr_s||_inf = 1.1406858507713101e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 2.3840871588389442e-03\n", + "||curr_z_L||_inf = 2.6881424074940881e-07\n", + "||curr_z_U||_inf = 5.8414507488241126e-07\n", + "||curr_v_L||_inf = 2.3840836745169887e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 15:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.5476349153543129e-04 1.5476349153543129e-04\n", + "Dual infeasibility......: 5.0054583110348449e-07 5.0054583110348449e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 5.1864101498357542e-08 5.1864101498357542e-08\n", + "Overall NLP error.......: 5.0054583110348449e-07 5.0054583110348449e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 15:\n", + "**************************************************\n", + "\n", + "Remaining in fixed mu mode.\n", + "Barrier Parameter: 2.722319e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 15:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 7.509304e-17\n", + "Factorization successful.\n", + "residual_ratio = 7.412612e-19\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 15:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 15 <--\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 3.363582E-20\n", + "Starting checks for alpha (primal) = 1.00e+00\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.929e-17 (tol = 1.000e-07) r (rel) = 4.978e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 7.8341992465810837e-04 (reference 7.8481025801009109e-04):\n", + " New values of constraint violation = 5.3949222476520897e-17 (reference 1.8214427091167024e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 16:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 1.981181e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 1.179498e-06 snrm = 1.429086e+00 ynrm = 1.935569e-06\n", + " Perform the update.\n", + "sigma (for B0) is 5.775381e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 16:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 16 1.5154537e-04 0.00e+00 2.11e-07 -7.6 1.98e-01 - 1.00e+00 1.00e+00f 1 Fy A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 16 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 2.7223188723064175e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "\n", + "||curr_x||_inf = 9.6172656812131052e-01\n", + "||curr_s||_inf = 1.1419635363106410e-05\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 2.3818039943382650e-03\n", + "||curr_z_L||_inf = 2.8810337578962471e-07\n", + "||curr_z_U||_inf = 6.2851405946942329e-07\n", + "||curr_v_L||_inf = 2.3818047330917014e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 16:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.5154537407026223e-04 1.5154537407026223e-04\n", + "Dual infeasibility......: 2.1135137784591342e-07 2.1135137784591342e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 2.8616723661402496e-08 2.8616723661402496e-08\n", + "Overall NLP error.......: 2.1135137784591342e-07 2.1135137784591342e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 16:\n", + "**************************************************\n", + "\n", + "Remaining in fixed mu mode.\n", + "Reducing mu to 9.0909090909090910e-10 in fixed mu mode. Tau becomes 9.9999997277681130e-01\n", + "Barrier Parameter: 9.090909e-10\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 16:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 1.966589e-15\n", + "Factorization successful.\n", + "residual_ratio = 8.261866e-18\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 16:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 16 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 4.206227E-22\n", + "Starting checks for alpha (primal) = 4.96e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.105e-17 (tol = 1.000e-07) r (rel) = 5.155e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 4.955224e-01:\n", + " New values of barrier function = 1.4810246367516608e-04 (reference 1.7264618689086475e-04):\n", + " New values of constraint violation = 8.8040604537611977e-18 (reference 5.3949222476520897e-17):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 17:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 2.759149e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 1.212184e-05 snrm = 2.951106e+00 ynrm = 5.285112e-06\n", + " Perform the update.\n", + "sigma (for B0) is 1.391871e-06\n", + "Number of successive iterations with skipping: 0\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 17:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 17 1.2565776e-04 0.00e+00 2.55e-07 -9.0 5.57e-01 - 4.03e-01 4.96e-01f 1 Fy A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 17 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 9.0909090909090910e-10\n", + "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "\n", + "||curr_x||_inf = 1.0000000041267956e+00\n", + "||curr_s||_inf = 9.5678170729163958e-06\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 1.7680920402422696e-03\n", + "||curr_z_L||_inf = 2.0383468974066809e-07\n", + "||curr_z_U||_inf = 6.5789430451231059e-07\n", + "||curr_v_L||_inf = 1.7681117532479319e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 17:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.2565776316551998e-04 1.2565776316551998e-04\n", + "Dual infeasibility......: 2.5516656650561394e-07 2.5516656650561394e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 1.7812930228906213e-08 1.7812930228906213e-08\n", + "Overall NLP error.......: 2.5516656650561394e-07 2.5516656650561394e-07\n", + "\n", + "\n", + "\n", + "\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 17:\n", + "**************************************************\n", + "\n", + "Switching back to free mu mode.\n", + "The current filter has 2 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 4.125976e-01\n", + "Barrier parameter mu computed by oracle is 6.804585e-09\n", + "Barrier parameter mu after safeguards is 6.804585e-09\n", + "Barrier Parameter: 6.804585e-09\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 17:\n", + "**************************************************\n", + "\n", + "residual_ratio = 1.941033e-15\n", + "Factorization successful.\n", + "residual_ratio = 2.425043e-18\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 17:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 17 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 8.998556E-26\n", + "Starting checks for alpha (primal) = 9.52e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.410e-17 (tol = 1.000e-07) r (rel) = 5.464e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 9.521680e-01:\n", + " New values of barrier function = 2.9116927903389189e-04 (reference 2.9365731757830004e-04):\n", + " New values of constraint violation = 1.0633651619830486e-16 (reference 8.8040604537611977e-18):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 18:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 18 1.1906625e-04 0.00e+00 2.82e-07 -8.2 2.41e-01 - 9.98e-01 9.52e-01f 1 sigma=4.13e-01 qf=12y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 18 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 6.8045847875022476e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999974483343346e-01\n", + "\n", + "||curr_x||_inf = 9.9999998617316055e-01\n", + "||curr_s||_inf = 9.0481780820836682e-06\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 8.1313752757328553e-04\n", + "||curr_z_L||_inf = 9.4352784989153782e-08\n", + "||curr_z_U||_inf = 5.2997432177651092e-07\n", + "||curr_v_L||_inf = 8.1313496326127247e-04\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 18:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.1906625426859106e-04 1.1906625426859106e-04\n", + "Dual infeasibility......: 2.8191370018366373e-07 2.8191370018366373e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 1.2457836678372943e-08 1.2457836678372943e-08\n", + "Overall NLP error.......: 2.8191370018366373e-07 2.8191370018366373e-07\n", + "\n", + "\n", + "\n", + "\n", + "Number of Iterations....: 18\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.1906625426859106e-04 1.1906625426859106e-04\n", + "Dual infeasibility......: 2.8191370018366373e-07 2.8191370018366373e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 1.2457836678372943e-08 1.2457836678372943e-08\n", + "Overall NLP error.......: 2.8191370018366373e-07 2.8191370018366373e-07\n", + "\n", + "\n", + "Number of objective function evaluations = 19\n", + "Number of objective gradient evaluations = 19\n", + "Number of equality constraint evaluations = 0\n", + "Number of inequality constraint evaluations = 19\n", + "Number of equality constraint Jacobian evaluations = 0\n", + "Number of inequality constraint Jacobian evaluations = 19\n", + "Number of Lagrangian Hessian evaluations = 0\n", + "Total CPU secs in IPOPT (w/o function evaluations) = 1.963\n", + "Total CPU secs in NLP function evaluations = 2.144\n", + "\n", + "EXIT: Solved To Acceptable Level.\n", + "v=0.2\n", + "w=0.1\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "rm: cannot remove '/home/fenics/shared/templates/RES_OPT/TEMP*': No such file or directory\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "\n", + "# Let's first optimize the generated designs, to see how far away they are from optimal designs\n", + "opt_designs = []\n", + "opt_histories = []\n", + "for i in range(n_samples):\n", + " opt_design, opt_history = problem.optimize(\n", + " starting_point=gen_designs[i][0],\n", + " config=conditions[i],\n", + " )\n", + " opt_histories.append(opt_history)\n", + " opt_designs.append(opt_design)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Eyeball testing the model" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# Render the optimal designs next to the generated designs to get a visual idea\n", + "for i in range(n_samples):\n", + " fig_opt, ax_opt = problem.render(opt_designs[i])\n", + " fig_gen, ax_gen = problem.render(gen_designs[i][0])\n", + " ax_opt.set_title(\"Optimal\")\n", + " ax_gen.set_title(\"Generated\")\n", + " plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Qualitative evaluation" + ] + }, + { + "cell_type": "code", + "execution_count": 44, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "WARNING: The requested image's platform (linux/amd64) does not match the detected host platform (linux/arm64/v8) and no specific platform was requested\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.392e-17 (tol = 1.000e-07) r (rel) = 3.425e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "\n", + "List of options:\n", + "\n", + " Name Value # times used\n", + " acceptable_tol = 0.001 1\n", + " hessian_approximation = limited-memory 7\n", + " max_iter = 0 1\n", + " print_level = 6 2\n", + "\n", + "******************************************************************************\n", + "This program contains Ipopt, a library for large-scale nonlinear optimization.\n", + " Ipopt is released as open source code under the Eclipse Public License (EPL).\n", + " For more information visit http://projects.coin-or.org/Ipopt\n", + "******************************************************************************\n", + "\n", + "This is Ipopt version 3.12.9, running with linear solver mumps.\n", + "NOTE: Other linear solvers might be more efficient (see Ipopt documentation).\n", + "\n", + "Number of nonzeros in equality constraint Jacobian...: 0\n", + "Number of nonzeros in inequality constraint Jacobian.: 10201\n", + "Number of nonzeros in Lagrangian Hessian.............: 0\n", + "\n", + "Hessian approximation will be done in the space of all 10201 x variables.\n", + "\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Scaling parameter for objective function = 1.000000e+00\n", + "objective scaling factor = 1\n", + "No x scaling provided\n", + "No c scaling provided\n", + "No d scaling provided\n", + "Moved initial values of x sufficiently inside the bounds.\n", + "Moved initial values of s sufficiently inside the bounds.\n", + "MUMPS used permuting_scaling 5 and pivot_order 5.\n", + " scaling will be 77.\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Least square estimates max(y_c) = 0.000000e+00, max(y_d) = 9.999010e-01\n", + "Total number of variables............................: 10201\n", + " variables with only lower bounds: 0\n", + " variables with lower and upper bounds: 10201\n", + " variables with only upper bounds: 0\n", + "Total number of equality constraints.................: 0\n", + "Total number of inequality constraints...............: 1\n", + " inequality constraints with only lower bounds: 1\n", + " inequality constraints with lower and upper bounds: 0\n", + " inequality constraints with only upper bounds: 0\n", + "\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.529e-17 (tol = 1.000e-07) r (rel) = 3.563e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 0:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 0 2.8915034e-05 0.00e+00 1.00e-04 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 0 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 1.0000000000000000e+00\n", + "Current fraction-to-the-boundary parameter tau = 0.0000000000000000e+00\n", + "\n", + "||curr_x||_inf = 9.9000000989999992e-01\n", + "||curr_s||_inf = 9.9999900000000003e-03\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 9.9990102485197330e-01\n", + "||curr_z_L||_inf = 1.0000000000000000e+00\n", + "||curr_z_U||_inf = 1.0000000000000000e+00\n", + "||curr_v_L||_inf = 1.0000000000000000e+00\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 0:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 2.8915033667732908e-05 2.8915033667732908e-05\n", + "Dual infeasibility......: 9.9997412027374999e-05 9.9997412027374999e-05\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 9.9000001989999997e-01 9.9000001989999997e-01\n", + "Overall NLP error.......: 9.9000001989999997e-01 9.9000001989999997e-01\n", + "\n", + "\n", + "\n", + "\n", + "Number of Iterations....: 0\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 2.8915033667732908e-05 2.8915033667732908e-05\n", + "Dual infeasibility......: 9.9997412027374999e-05 9.9997412027374999e-05\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 9.9000001989999997e-01 9.9000001989999997e-01\n", + "Overall NLP error.......: 9.9000001989999997e-01 9.9000001989999997e-01\n", + "\n", + "\n", + "Number of objective function evaluations = 1\n", + "Number of objective gradient evaluations = 1\n", + "Number of equality constraint evaluations = 0\n", + "Number of inequality constraint evaluations = 1\n", + "Number of equality constraint Jacobian evaluations = 0\n", + "Number of inequality constraint Jacobian evaluations = 1\n", + "Number of Lagrangian Hessian evaluations = 0\n", + "Total CPU secs in IPOPT (w/o function evaluations) = 0.233\n", + "Total CPU secs in NLP function evaluations = 0.129\n", + "\n", + "EXIT: Maximum Number of Iterations Exceeded.\n", + "Optimization complete: v=0.5, w=0.5\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "rm: cannot remove '/home/fenics/shared/templates/RES_SIM/TEMP*': No such file or directory\n", + "WARNING: The requested image's platform (linux/amd64) does not match the detected host platform (linux/arm64/v8) and no specific platform was requested\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.834e-17 (tol = 1.000e-07) r (rel) = 4.878e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "\n", + "List of options:\n", + "\n", + " Name Value # times used\n", + " acceptable_tol = 0.001 1\n", + " hessian_approximation = limited-memory 7\n", + " max_iter = 0 1\n", + " print_level = 6 2\n", + "\n", + "******************************************************************************\n", + "This program contains Ipopt, a library for large-scale nonlinear optimization.\n", + " Ipopt is released as open source code under the Eclipse Public License (EPL).\n", + " For more information visit http://projects.coin-or.org/Ipopt\n", + "******************************************************************************\n", + "\n", + "This is Ipopt version 3.12.9, running with linear solver mumps.\n", + "NOTE: Other linear solvers might be more efficient (see Ipopt documentation).\n", + "\n", + "Number of nonzeros in equality constraint Jacobian...: 0\n", + "Number of nonzeros in inequality constraint Jacobian.: 10201\n", + "Number of nonzeros in Lagrangian Hessian.............: 0\n", + "\n", + "Hessian approximation will be done in the space of all 10201 x variables.\n", + "\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Scaling parameter for objective function = 1.000000e+00\n", + "objective scaling factor = 1\n", + "No x scaling provided\n", + "No c scaling provided\n", + "No d scaling provided\n", + "Moved initial values of x sufficiently inside the bounds.\n", + "Moved initial values of s sufficiently inside the bounds.\n", + "MUMPS used permuting_scaling 5 and pivot_order 5.\n", + " scaling will be 77.\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Least square estimates max(y_c) = 0.000000e+00, max(y_d) = 9.999011e-01\n", + "Total number of variables............................: 10201\n", + " variables with only lower bounds: 0\n", + " variables with lower and upper bounds: 10201\n", + " variables with only upper bounds: 0\n", + "Total number of equality constraints.................: 0\n", + "Total number of inequality constraints...............: 1\n", + " inequality constraints with only lower bounds: 1\n", + " inequality constraints with lower and upper bounds: 0\n", + " inequality constraints with only upper bounds: 0\n", + "\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.142e-17 (tol = 1.000e-07) r (rel) = 5.189e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 0:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 0 6.8136479e-05 1.94e-03 1.00e-04 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 0 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 1.0000000000000000e+00\n", + "Current fraction-to-the-boundary parameter tau = 0.0000000000000000e+00\n", + "\n", + "||curr_x||_inf = 9.9000000989999992e-01\n", + "||curr_s||_inf = 9.9999900000000003e-03\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 9.9990105427563769e-01\n", + "||curr_z_L||_inf = 1.0000000000000000e+00\n", + "||curr_z_U||_inf = 1.0000000000000000e+00\n", + "||curr_v_L||_inf = 1.0000000000000000e+00\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 0:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 6.8136478539192145e-05 6.8136478539192145e-05\n", + "Dual infeasibility......: 9.9989669015254634e-05 9.9989669015254634e-05\n", + "Constraint violation....: 1.9428233460940195e-03 1.9428233460940195e-03\n", + "Complementarity.........: 9.9000001989999997e-01 9.9000001989999997e-01\n", + "Overall NLP error.......: 9.9000001989999997e-01 9.9000001989999997e-01\n", + "\n", + "\n", + "\n", + "\n", + "Number of Iterations....: 0\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 6.8136478539192145e-05 6.8136478539192145e-05\n", + "Dual infeasibility......: 9.9989669015254634e-05 9.9989669015254634e-05\n", + "Constraint violation....: 1.9428233460940195e-03 1.9428233460940195e-03\n", + "Complementarity.........: 9.9000001989999997e-01 9.9000001989999997e-01\n", + "Overall NLP error.......: 9.9000001989999997e-01 9.9000001989999997e-01\n", + "\n", + "\n", + "Number of objective function evaluations = 1\n", + "Number of objective gradient evaluations = 1\n", + "Number of equality constraint evaluations = 0\n", + "Number of inequality constraint evaluations = 1\n", + "Number of equality constraint Jacobian evaluations = 0\n", + "Number of inequality constraint Jacobian evaluations = 1\n", + "Number of Lagrangian Hessian evaluations = 0\n", + "Total CPU secs in IPOPT (w/o function evaluations) = 0.226\n", + "Total CPU secs in NLP function evaluations = 0.131\n", + "\n", + "EXIT: Maximum Number of Iterations Exceeded.\n", + "Optimization complete: v=0.4, w=0.7\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "rm: cannot remove '/home/fenics/shared/templates/RES_SIM/TEMP*': No such file or directory\n", + "WARNING: The requested image's platform (linux/amd64) does not match the detected host platform (linux/arm64/v8) and no specific platform was requested\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.361e-17 (tol = 1.000e-07) r (rel) = 5.415e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "\n", + "List of options:\n", + "\n", + " Name Value # times used\n", + " acceptable_tol = 0.001 1\n", + " hessian_approximation = limited-memory 7\n", + " max_iter = 0 1\n", + " print_level = 6 2\n", + "\n", + "******************************************************************************\n", + "This program contains Ipopt, a library for large-scale nonlinear optimization.\n", + " Ipopt is released as open source code under the Eclipse Public License (EPL).\n", + " For more information visit http://projects.coin-or.org/Ipopt\n", + "******************************************************************************\n", + "\n", + "This is Ipopt version 3.12.9, running with linear solver mumps.\n", + "NOTE: Other linear solvers might be more efficient (see Ipopt documentation).\n", + "\n", + "Number of nonzeros in equality constraint Jacobian...: 0\n", + "Number of nonzeros in inequality constraint Jacobian.: 10201\n", + "Number of nonzeros in Lagrangian Hessian.............: 0\n", + "\n", + "Hessian approximation will be done in the space of all 10201 x variables.\n", + "\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "Scaling parameter for objective function = 1.000000e+00\n", + "objective scaling factor = 1\n", + "No x scaling provided\n", + "No c scaling provided\n", + "No d scaling provided\n", + "Moved initial values of x sufficiently inside the bounds.\n", + "Moved initial values of s sufficiently inside the bounds.\n", + "MUMPS used permuting_scaling 5 and pivot_order 5.\n", + " scaling will be 77.\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Least square estimates max(y_c) = 0.000000e+00, max(y_d) = 9.999011e-01\n", + "Total number of variables............................: 10201\n", + " variables with only lower bounds: 0\n", + " variables with lower and upper bounds: 10201\n", + " variables with only upper bounds: 0\n", + "Total number of equality constraints.................: 0\n", + "Total number of inequality constraints...............: 1\n", + " inequality constraints with only lower bounds: 1\n", + " inequality constraints with lower and upper bounds: 0\n", + " inequality constraints with only upper bounds: 0\n", + "\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.597e-17 (tol = 1.000e-07) r (rel) = 5.653e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 0:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 0 1.2998394e-04 0.00e+00 1.00e-04 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 0 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 1.0000000000000000e+00\n", + "Current fraction-to-the-boundary parameter tau = 0.0000000000000000e+00\n", + "\n", + "||curr_x||_inf = 9.9000000989999992e-01\n", + "||curr_s||_inf = 9.9999900000000003e-03\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 9.9990105260724160e-01\n", + "||curr_z_L||_inf = 1.0000000000000000e+00\n", + "||curr_z_U||_inf = 1.0000000000000000e+00\n", + "||curr_v_L||_inf = 1.0000000000000000e+00\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 0:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.2998393571133284e-04 1.2998393571133284e-04\n", + "Dual infeasibility......: 1.0001132391701528e-04 1.0001132391701528e-04\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 9.9000001989999997e-01 9.9000001989999997e-01\n", + "Overall NLP error.......: 9.9000001989999997e-01 9.9000001989999997e-01\n", + "\n", + "\n", + "\n", + "\n", + "Number of Iterations....: 0\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.2998393571133284e-04 1.2998393571133284e-04\n", + "Dual infeasibility......: 1.0001132391701528e-04 1.0001132391701528e-04\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 9.9000001989999997e-01 9.9000001989999997e-01\n", + "Overall NLP error.......: 9.9000001989999997e-01 9.9000001989999997e-01\n", + "\n", + "\n", + "Number of objective function evaluations = 1\n", + "Number of objective gradient evaluations = 1\n", + "Number of equality constraint evaluations = 0\n", + "Number of inequality constraint evaluations = 1\n", + "Number of equality constraint Jacobian evaluations = 0\n", + "Number of inequality constraint Jacobian evaluations = 1\n", + "Number of Lagrangian Hessian evaluations = 0\n", + "Total CPU secs in IPOPT (w/o function evaluations) = 0.225\n", + "Total CPU secs in NLP function evaluations = 0.131\n", + "\n", + "EXIT: Maximum Number of Iterations Exceeded.\n", + "Optimization complete: v=0.2, w=0.1\n", + "Initial optimality gaps: [array([0.0004409]), array([0.0004189]), array([1.70724543e-05])]\n", + "Cumulative optimality gaps: [0.00262544707661, 0.0026488498937400004, 0.00036407371150999987]\n", + "Final optimality gaps: [array([-7.5917667e-07]), array([1.8115146e-07]), array([-1.09176857e-05])]\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "rm: cannot remove '/home/fenics/shared/templates/RES_SIM/TEMP*': No such file or directory\n" + ] + } + ], + "source": [ + "# Now we can also use metrics to get a quantitative idea of how good the generated designs are\n", + "# Here we compute the average optimality gap, the cumulative optimality gap, and the final optimality gap\n", + "objective_values_opt = [problem.simulate(opt_designs[i], config=conditions[i]) for i in range(n_samples)]\n", + "optimality_gaps = [metrics.optimality_gap(opt_histories[i], baseline=objective_values_opt[i]) for i in range(n_samples)]\n", + "\n", + "iog = [optimality_gaps[i][0] for i in range(n_samples)]\n", + "cog = [np.sum(optimality_gaps[i]) for i in range(n_samples)]\n", + "fog = [optimality_gaps[i][-1] for i in range(n_samples)]\n", + "\n", + "print(f\"Initial optimality gaps: {iog}\")\n", + "print(f\"Cumulative optimality gaps: {cog}\")\n", + "print(f\"Final optimality gaps: {fog}\")" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": ".venv", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.12.9" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} From 62596e72f660bfdc36465259b323d199ee8df315 Mon Sep 17 00:00:00 2001 From: Florian Felten Date: Wed, 9 Jul 2025 12:13:10 +0200 Subject: [PATCH 2/5] Update readme --- README.md | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 9892c19e..c4cf28bb 100644 --- a/README.md +++ b/README.md @@ -10,9 +10,9 @@ This repository contains the code for optimization and machine learning algorith ## Coding Philosophy As much as we can, we follow the [CleanRL](https://github.com/vwxyzjn/cleanrl) philosophy: single-file, high-quality implementations with research-friendly features: -* Single-file implementation: every detail is in one file, so you can easily understand and modify the code. +* Single-file implementation: every training detail is in one file, so you can easily understand and modify the code. There is usually another file that contains evaluation code. * High-quality: we use type hints, docstrings, and comments to make the code easy to understand. We also rely on linters for formatting and checking our code. -* Logging: we use experiment tracking tools like [Weights & Biases](https://wandb.ai/site) to log the results of our experiments. +* Logging: we use experiment tracking tools like [Weights & Biases](https://wandb.ai/site) to log the results of our experiments. All our "official" runs are logged in the [EngiOpt project](https://wandb.ai/engibench/engiopt). * Reproducibility: we seed all the random number generators, make PyTorch deterministic, report the hyperparameters and code in WandB. ## Install @@ -31,3 +31,7 @@ pip install -e ".[all]" ## Dashboards The integration with WandB allows us to access live dashboards of our runs (on the cluster or not). We also upload the trained models there. You can access some of our runs at https://wandb.ai/engibench/engiopt. WandB dashboards + +## Colab notebooks +We have some colab notebooks that show how to use some of the EngiBench/EngiOpt features. +* [Example hard model](https://colab.research.google.com/github/IDEALLab/EngiOpt/blob/main/example_hard_model.ipynb) From 8995f9f143623345fe086122dcebeb46f9ad853f Mon Sep 17 00:00:00 2001 From: Florian Felten Date: Wed, 9 Jul 2025 14:16:58 +0200 Subject: [PATCH 3/5] Cleaning up --- example_hard_model.ipynb | 3911 +++++++++++++++++++------------------- 1 file changed, 2008 insertions(+), 1903 deletions(-) diff --git a/example_hard_model.ipynb b/example_hard_model.ipynb index 1bdb7538..4181f25d 100644 --- a/example_hard_model.ipynb +++ b/example_hard_model.ipynb @@ -20,7 +20,7 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 1, "metadata": {}, "outputs": [ { @@ -132,8 +132,8 @@ "Requirement already satisfied: grapheme==0.6.0 in ./.venv/lib/python3.12/site-packages (from alive-progress->pymoo>=0.6->engiopt==0.0.1) (0.6.0)\n", "Requirement already satisfied: wrapt<2,>=1.10 in ./.venv/lib/python3.12/site-packages (from Deprecated->pymoo>=0.6->engiopt==0.0.1) (1.17.2)\n", "Collecting pyoptsparse@ git+https://github.com/mdolab/pyoptsparse#egg=7e862db49bbc8bde2b49dee686cce68c8d2ab9f0 (from engibench[all]->engiopt==0.0.1)\n", - " Cloning https://github.com/mdolab/pyoptsparse to /private/var/folders/fw/69hmbxm946l1xycg8w29nw0800l9gd/T/pip-install-f89x5s5_/pyoptsparse_d30cd54894d5404ba70e7fae5d361a3e\n", - " Running command git clone --filter=blob:none --quiet https://github.com/mdolab/pyoptsparse /private/var/folders/fw/69hmbxm946l1xycg8w29nw0800l9gd/T/pip-install-f89x5s5_/pyoptsparse_d30cd54894d5404ba70e7fae5d361a3e\n", + " Cloning https://github.com/mdolab/pyoptsparse to /private/var/folders/fw/69hmbxm946l1xycg8w29nw0800l9gd/T/pip-install-zxekdyn0/pyoptsparse_f7cbacf77ad341919fb2519347802ce5\n", + " Running command git clone --filter=blob:none --quiet https://github.com/mdolab/pyoptsparse /private/var/folders/fw/69hmbxm946l1xycg8w29nw0800l9gd/T/pip-install-zxekdyn0/pyoptsparse_f7cbacf77ad341919fb2519347802ce5\n", " Resolved https://github.com/mdolab/pyoptsparse to commit 9d417c740693dc706fef8ff2c064837fea04ec50\n", " Installing build dependencies ... \u001b[?25ldone\n", "\u001b[?25h Getting requirements to build wheel ... \u001b[?25ldone\n", @@ -182,8 +182,8 @@ "Requirement already satisfied: pure-eval in ./.venv/lib/python3.12/site-packages (from stack_data->ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (0.2.3)\n", "Building wheels for collected packages: engiopt\n", " Building editable for engiopt (pyproject.toml) ... \u001b[?25ldone\n", - "\u001b[?25h Created wheel for engiopt: filename=engiopt-0.0.1-0.editable-py3-none-any.whl size=16674 sha256=d7d123eb09c72b58f08ef781c4a39dabdce219f28c2c741d1bf51cb9663f2b3c\n", - " Stored in directory: /private/var/folders/fw/69hmbxm946l1xycg8w29nw0800l9gd/T/pip-ephem-wheel-cache-6qjf1rk_/wheels/b6/44/39/3c3f0715f4343d6ef24ce1d8f1606494f927246a0d3d312dcc\n", + "\u001b[?25h Created wheel for engiopt: filename=engiopt-0.0.1-0.editable-py3-none-any.whl size=16828 sha256=d2bc822e152627edd5203df85f86232675450ffc27a4116b36a094823796fa5a\n", + " Stored in directory: /private/var/folders/fw/69hmbxm946l1xycg8w29nw0800l9gd/T/pip-ephem-wheel-cache-catulojk/wheels/b6/44/39/3c3f0715f4343d6ef24ce1d8f1606494f927246a0d3d312dcc\n", "Successfully built engiopt\n", "Installing collected packages: engiopt\n", " Attempting uninstall: engiopt\n", @@ -193,19 +193,31 @@ "Successfully installed engiopt-0.0.1\n", "\n", "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m24.2\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m25.1.1\u001b[0m\n", - "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n" + "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n", + "Note: you may need to restart the kernel to use updated packages.\n" ] } ], "source": [ - "!pip install -e ." + "%pip install -e ." ] }, { "cell_type": "code", - "execution_count": 28, + "execution_count": 2, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/gymnasium/spaces/box.py:235: UserWarning: \u001b[33mWARN: Box low's precision lowered by casting to float32, current low.dtype=float64\u001b[0m\n", + " gym.logger.warn(\n", + "/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/gymnasium/spaces/box.py:305: UserWarning: \u001b[33mWARN: Box high's precision lowered by casting to float32, current high.dtype=float64\u001b[0m\n", + " gym.logger.warn(\n" + ] + } + ], "source": [ "# Some imports we will need\n", "import os\n", @@ -218,7 +230,7 @@ "from engiopt import metrics\n", "from engiopt.diffusion_2d_cond.diffusion_2d_cond import beta_schedule\n", "from engiopt.diffusion_2d_cond.diffusion_2d_cond import DiffusionSampler\n", - "import wandb\n" + "import wandb" ] }, { @@ -230,7 +242,7 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 3, "metadata": {}, "outputs": [ { @@ -239,13 +251,13 @@ "(
, )" ] }, - "execution_count": 8, + "execution_count": 3, "metadata": {}, "output_type": "execute_result" }, { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfQAAAGgCAYAAABcyZVrAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/GU6VOAAAACXBIWXMAAA9hAAAPYQGoP6dpAACOQUlEQVR4nO2de5xVZfX/13AZBoFhuDgzgOAgXhBEUUBEMFPH8Ebe8y6hoSmkOL8yMQEzFbO+SpZpmteUJEvNW5rhLRUFUUpT8QZB1IBmx1FERmbO7w9q9ud5b2YfDjMDnDPr/Xrx4lln7dln72ffzl6fZ62nIJ1Op81xHMdxnJymzZbeAMdxHMdxmo4/0B3HcRwnD/AHuuM4juPkAf5AdxzHcZw8wB/ojuM4jpMH+APdcRzHcfIAf6A7juM4Th7gD3THcRzHyQP8ge44juM4eYA/0B3HcRwnD9hiD/Trr7/eKioqrKioyEaOHGnz58/fUpviOI7jODlPwZao5T5nzhw7/fTT7cYbb7SRI0farFmz7N5777XFixdbaWlpxr+vr6+3f/7zn9alSxcrKCjYDFvsOI7jNCfpdNo++eQT6927t7Vp0zLvlp9//rnV1tY2y7oKCwutqKioWdbVYqS3AHvvvXd60qRJDXZdXV26d+/e6ZkzZ27U3y9fvjxtZv7P//k//+f/cvzf8uXLW+Q5s2bNmnR5eXmzbWd5eXl6zZo1LbKtzUU728zU1tbawoULberUqQ2ftWnTxiorK23evHkb/Ju1a9fa2rVrG+z0f4MKR5tZ+/9+9i9Zvg/+/huwd5T20/Dx99eh+4lxSOir+15oV0n7PaxnAOyf9BbjzbPh/TQ0R9/d0Lz69dD1Ef5S93XH4tB3ZU1o/13al2A9fT8+Ap9MkXYHS2adtJ8PXUeH3zTzyaj9ItbyBewKaV8OX/ePK/DJT6TdKb6Jm511sO+Pmj1+EXh+gUW1Bz/EWipg/1yv6H8fBe8E2En90hZ20q0CJ+Vp5zQ0Zz4Yul7GX+p1cRF83S/AB7+OmpdWh66/JGzdTrCv3hEfLLxQjKFwfg57rbRxndplgVXb9ZOG9pVYkue23lr26xv63l8e2j+T9mqs5yzYw1TJ3KVn6NwnPJt+8WbU/jfWcwrs7feN2re/EPp+j2U/k/ae0q41s1vMrEuXLtYS1NbWWnV1tS1fvsSKi4sz/0ECNTU11rdvf6utrd2q39I3+wP9ww8/tLq6OisrKws+Lysrs7feemuDfzNz5kz7/ve/H/u8vUUPdN2RQizH25Ye2m3g46Eqbte4sw7L6veyY7lNxRphKqYXttxbuX18tHbW1UKN4LL6Lbykiovb4xPtxUwntD6RsCw6pqhxl6VhJ28vQ3adGmlvKfhAl73BceqIJfVIZDyvdF2Jx3BDtsJvSrpVYIvla3mmcIv0nIwdU56wcojpSto6LlvM3yrBF/GOwPNKbd4BwgOpgd6EXTGz8EjwVGa/6DHnD4POsIuDGwJWjH7Qo8jtjR0b6XCer0lnzoZeBVpaNi0uLm7yAz1X2OwP9E1h6tSpVlUVvf+u/7XU15ZYtAPvy/L65mkWf9fYX9rvw9cbtumv41TowktCsC6uN3br/EANHobRodnltka/k28m8hJj+38c+t7AsvrCXs/tsxLY5dLO5hfqyNDEn/5T2vw5x5uV3gw+ga9H7G1J4zS8zW0J+ECXs3D7nwaeBe+GS74mbe43b5CrpdM6xb6Tsats+iWLW4V8DSNIvC702oypqP+EnYqai+Hiua0PvR5cL5/DQb8wlsY+1PNsLXzhuV7Y5omG9ru4wJbiL1PSbosb2DNY9hVp8+cH4wnB63FwDZtZ0arA1LdyRhjfgV0hHc6+x+kb36b/wp5tOdY1w7dtvq1tCpv9gd6zZ09r27atrVy5Mvh85cqVVl5evsG/6dChg3XokCnM6ziO4zik9TzQN3vaWmFhoQ0bNszmzp3b8Fl9fb3NnTvXRo0atbk3x3Ecx8lr1jXTv62fLRJyr6qqsvHjx9vw4cNt7733tlmzZtnq1attwgQO2klmnUU6q4Z1PsNyj8JeIG1qQ/vCrpH4UTHifBirEoTNGGaKRfmCEByDzQj7dY+aDB+mYGvI/XH4GP5UeaG7EYaw1zXSzgSG7aBjtF8YYk+yY79E14XhQ2u3qdvbUnAbUlEzg4LxeSNts7hUEp5nDJxm2iaFt4akZeGTOHBJhi3QMO9K+LqhX16UC5vXHq95DeXHtmHDgcBG4H6rjsUhivim4VGzDGU2FuEv/yTtl+DjdaCyyy7wYTxdePnV/jX0oX/1b5/Caq6Gfb/sOgezUhbq0kib++U0nS3yQD/hhBPsgw8+sOnTp1t1dbUNHTrUHnvssdhAOcdxHMdpGnXW9B/1sVeyrZItNihu8uTJNnny5C319Y7jOE6rwDV0x3Ecx3FyiJxIW2uMHhbltaqGlsJy1GpUf2MeL+VMzZwphrj1WmgG2hHzbUtgh/lGTD47KDQHRc3hKNTxSmgG+htT3Lhvqr8Vd+UGNgX9NQvVF6IvUwqT0D6NpQFygEDp1v6LWvoFuUeUeJP6iOdZaKPgy+Z6y5BKLoPg4r6pyn8+fHvdFNp6lVBDJ9tKOzbUdi9+kJS+xz77h7SpdqNwS1Q7y845OnSl8JevSpv7xvtXibSpmVdwV/R6QwEY3hA0nfe3WJT3Gb33cRzHdrAPlPYe0l5j4diBlqP1vKHn9APdcRzHcZJpPQ90D7k7juM4Th7gb+iO4zhOHlNnTR+l7qPcW5xvWqSlqtbNipFLYWtpQpY0ZBnDb0p7EEoyLsKyKWkzX5wTRAQL1GCLi5GRu33URCHVsIKshfvDfqDeGuhv21oGNjXkhHx2XBe6TdSKqR3qeAfuS6wjtgoNPSkXPtXoX3E4g4bR2EdJfdZxCdTN/tn0SROW3SFq7h96YpnxMkVNoCObxUeWxMsTR1Cb13oSx3DhA/hBNiVwpRjFkttCV/9+oX3UwQ3NwVOeCFzfnRUuepe0qV/z1NZc7jHcvCEJf0whfEFopqTNxxfrH+h5xvEsI2D/WItcyKCKmnVmZzOJvUVoPWlrHnJ3HMdxnDwgp9/QHcdxHCeZ1jMoLqcf6Ad/XGDF/5sv8lkJyCG1y8Lppu01iQLfgUU5s5GGCDPNBqbhT87a9hXYQbx7KXy7LwptmfmvFMX0jkR0Xku/MlRXAntIkjORTCe3+hFyR6dlk7aWGE7iBM5b3QXI7ZENRt3STGF1hX2iodFiTm3Wv7F5r7KF+wJbztHy3ULXOcik03DttVgrMxE1y2oH+BhFP1u/43A4qVtldRucEzW/B1fZstC+Vrb42tLANeT1sFTx0ZK/tQirZSnVEmkzvB3THpZKe07oug+zMd4nbc6YRvRY8BzkTHj3yYEc9FzUZnHplsMf6I7jOI6TB7SeB7pr6I7jOI6TB/gbuuM4jpPHtJ5R7jn+QP+LNSRxfElUnS+lwsV+fF1gDrnyhob2D6CD3RWawTSCLPRIjVpTqajx7QM7iI0wV27350MbM5AqVEWTUpw4VWyQSofyo3E2NY0JU0wmTJ+aDbG/o+C6RUJk2Xzn0qiZcHzNkvuIqVyBHI80y/jZks30qUlgPbpRJaGr246hfZyItb8OXTHtWEuKng7fObALvirGgXDGUjS1X9gn6LO5UTLoe9hgponul3o7Mm5DEdxpoYa+l2joTHHl7aGukbaZxTvtj1Hz6SWhi+OHNE2Qq2GXDUzYhhRs7SY9/Gttc+Ehd8dxHMdxcogcf0N3HMdxnCRazxu6P9Adx3GcPMYf6DlCV4uStDUzkqUcLwvNi7/W0Ox4YJjBOhE5q3uJNvsbrBVpnYHaloKPenv5y2LMg3MEEpM1dxtaMWdEVDe1LWrqHZKcMZLKmCaBRPlmSoWOlQGNaeiq0HF7m3Lab+qFzb+TCgcJh9ssueQpCbqXom4msT4gm/3EmaYC7NLQteYfoa11H3hq8JTU0g0nwVfAQg8qWdfCx37pl3RSIllbalzcH3pilVW/uD1qHzgESnhVOHFz4dHRRu6BFT8XmsHpwlIDw1aE9oo3o/adWJaldVPSLoGPJXwvlDanZf4D7KXS1orXzVUVwYnI8Qe64ziO4yThb+iO4ziOkwd42lqO0Nk2bqYkBoWkHuU+iHfPHRWYw8ZF7XqECzlTm4a/Wf7wUdgaVR/CerNHw2YOifAybA3tZ3UKxqYv46mRFHJPWhadhjhbNuFkJWO6TlYBvc3x65vfITVQse1JgXHud6JNnYcphDZA2k1IUyMa4obiwvNVT/1Y9V6gaZfdeL5Sp3gtYUWckSwx5J4KzUVRk+FuhrBvlfYB/y/0FYyBDjAhao5DyP2p0AzSZykDFr0Z2pINF0u7hdIThMP3gO8I2DvrlIAopXt2KuGL5OSuqTebFkutdJpCjj/QHcdxHCcJD7k7juM4Th7gD3THcRzHyQP8gZ4jtLNN2wX9m4rQNfSx0P7BIQ3NHSaErjFYq+rm1fCxrKUyBKkmtgC2rGw1tMLlWJS6mMI0oGBkAed7jY1NyCZtTf3JaWubOtQkpr1npaFvBWVhl0V7kMZB45Zn00fBMY6J0jwrs+mHLK4znRMVtShTWFTHC2QaTxHsGzuFg1Z03zvBR0E4sfRrKjQTxrPUwNbxAjfBd/al+EBKUJdjLtivQUTXBDimrXJcj94umL3H8tR7SfsU+Ebvgg/Okvah8O3Km4mK7HKgaurMuvLAOU0hxx/ojuM4jpOEv6E7juM4Th7QetLWfHIWx3Ecx8kDWukbertG2mZmEIu+/q2GZo+7fxq4hv4pXFSrTf4Za30LtpaJPAPVJWPJuiLxMZeUVSxVJ2P2fVfY5WqMgDP216p2ss+4bML0qcwX3kSoB8YEzERddEuc9lDGZfADu4TTSia9G3BcRPALPcWlmfSr/ZKpT5LecODThPG+oav3u6E9VNo8t1nNNyXtFRDc63Ah6CW1PTq0OHG8BUcwpEKzJGpSKaZUr2UAHobvSNRHLZfSGDY89B2FjrlbxlxwDA1t7X5I87GhBId0F+NrcJ4KW2923Tj38qewdQf0XNnUKhTZUmdNf8POjTf0VvpAdxzHcVoHrUdD95C74ziO4+QBOf6G/qlFv0kY9lX46yrp1xbXc0bUnBWG3MfsFi6p1SaZPsLQuBT+tL8h5jqYsUbJNmIonxlvSg/Ye8Huo9t/ZMKKzCws4doBvqS+T85T0wqemX5d1jfSNrMNpBMx7Lcl0PMMBV0TQu6UE5ICk+yzYA4vSjkx4Sebcr6N/d0GbD2oyI0ahqh/iez8k1grQ/D6pwxhl8HWfohlZMZIOE7M/ZNQ83CkkzHbVC9jzLVmv4U9WWu4IuRuiGh3kJA7JRcGhrWE689YLpcx+LHS7g4f1Rrd9yeRd4np4aq/iPQwPV8TMgCbmdbzhp7jD3THcRzHScJHuTuO4ziOk0P4G7rjOI6Tx3jIPUdYa5FG21K7UhI1B/8s8BSePjmwR94ZtTmVIvUi1divhe9CpMNpFtvjWJYyqarZLO34DdimUzr2oaCG/KJAk+4JX9IUttDXIfp1atyVSKyw61apoSvYHhlUwWAeNXP1s48KYSenrS2Erb2YlHqYCSyrgwK42u1Cc4Bos32x46+EZpAGhkskJgfruU852LblB7r9LI+LK1mqmB7yy9D1d6THqYa+FGvljMlnST8UYlDFbGR+6pWZaTpdjt1R0n8M7QKdD5Z6O24078mKb8OiPG66OzqeIZZ62mL4A91xHMdx8oDW80B3Dd1xHMdx8gB/Q3ccx3HymNbzhp7jD/RPzazgv20V67KZ3jObAzU6NK8KzX3ui9rLIZlSL9I8dFSBjOXfqrTIGTEpdalGxdKOfU7HB1/vJwbEzr9hq1R3LN0fK9reGgd6e9Hbgan6JuVWyuKqD8YqyMbmjdUD0JQc66ag60qFLhGEm1INlyG2QGNnPYMVqI/bR0XfZsxD12NBSfqD0FydRfVP7acUfCy7WiHtgv5wohxtuK9LQ9e/ng1tLWv77dB19o9Du0i6l3nni2AfL+1yCN+vhWZQ3rWLJaNjdcbiRCsPTWsr4jzvKxyzopvIcrM8n7XktNYL2FyFXz1tzXEcx3GcnCLH39Adx3EcJ4l1ll0OTWPr2PrJ8Qf6xxaFQpJSp7Ihi7KwvRBzP+OihuaR14UuRvk0tQPZL/a+NQ7LuaL6rGkw/Owd4fwB1yaFIec+FLq+g0UlXcdueA/OYbD1tEJRTsQIVSJgyJ2XoIbyYmlrrNgZ/yCBzXGxIvdIwuFNCbkzNBoEBlNw8rD10QV4ZjWhT1RfQrrTfyBFrZR2plKgGk5kqJnXwUCtTjwQzljIXYHg9Ru4X5D2EPjCLFYb//uoPQhZoN8KTZsn7UxhU918lnNmRp7eZ3hfYTnabM5Dvf54nQ6CrRVlNb1wtcW7t2VoPQ90D7k7juM4Th6Q42/ojuM4jpNE63lD9we64ziOk8e0nlHuOf5A/8jWl381C9OjstmtbA4014t5Ds+KmoUQqPbBfKpDpNQjtSyWa9T0DmqFl8MeoqL0dDj7Qd1aKLr5oaHrZghqx70atbvd8AhW/DXY2k9IjikJTVVukyZhNQsvqZiGHvsgqfTr5vq1rd+zMnSJYEzdKxsdjKk/etjWIJWvY6wOaEra7ECe61lcU3ogcR9EddRgpEPsEAIdL0AZnHJ2sADnVo3VghWhP410zUdDc42US+3I+q0Uj+V7OJyFYwCSyqByCMB50j6Zgyiwr+/JrMc3YNEXYOsZyhEo1Nf1vOO7L6ervWTvDS9cs87ic846TSLHH+iO4ziOk8Q6a/pwMQ+5O47jOM4Wxh/ojuM4jpMH+AM9R3jfzDr+t91HPqca21KaetfQHLxv1N4WCtWi0ExJmzIYbdWvKuAbwoX3aKRtZrYMav2ZUfMCiGSvh2bQo6e9jQkSd7YESkJzm8ZN7grRTYzlLMdKv2ZSZBujKRcuzzNd1z9Cl2xepilRk6C2qf3CLukY67TUhjfIzDKPaFCw3yXSxs4llarNNA5Z/dRpC3FeBR3DgQaxL1oaNSks45LRXO62GBbRHbZOvMrSrxhSE2wie57jBU7W4UIj4RwamgNWRO0f3h76ZuNPdZrmFHy0F0ubFYa5b7PnR+1K+TxT3QEne3L8ge44juM4SdRZ00ep+yh3x3Ecx9nCeNpajvCiRQHKwfI55xFqClmUgtX0rR3C2N0aVP7USB5DT4wIJk1etQox11JdGWNhvw7NByQm+BQWZdpKULnyOTh3ZrhWy/DiFGvfuMn95iWk35KCL163MmlGvaTQeNJ6SKb1qP330JWQtpZEUp+Yhf0Sq3Aai6JrXVYecX6T1lLNkNKWUM+X26unaybJJen4r4a+0Ea6uyMXjoXc/xY1GXLHNaR/ynRTZmBpCPst+Ji+pyVbmdLG2cx+KPeSI5BBOhildu3/Rc22Y0PXaf8X2iNlgzEpHoUz+5O058AH5cFukrb2EfvAaTo5/kB3HMdxnCTWWTTNdlPWsfXjD3THcRwnj2k9D3SfnMVxHMdx8oDcfkOvnRPVTCzU/I0SLNiUNLYkuB4pzHpE6On4YywqAhK1TlamVJ2R6WSPwz5NRXYKd6hqKVUsY9o89cxAYV0KZ+KvV6imSCHKZsqERA09hh7zbI73pqa7beh7EtLWaqIms6qoZieV2aTSrUMJevDn+vawg/K42L6glLJZeE1leFtpt3vU3uGv4fb9JVxUz21qx9xX7RckTsZO9U7SHshOYr98Ktv4Knw4HVLSfhKLcmhJrNKuwENxmLR3ge/hBJuS/2HYiLP1wv4VFr49NHeWe9TO94e+GmjzY6TN85VdqGVt6xtptyyt5w09tx/ojuM4jpNI63mge8jdcRzHcfIAf0N3HMdx8pg6a/obuuehtzw3WCSVfucycXB6T2roLRU+kVKw++wduvadH5i7SOI3NSjmnWoq7L/hewn2aSkx3g19NdAvVXfMVPizRI2Y8J2Ufw1tNpYvvvGo5hZbTSyJuTM/SCApZz0phz2Lyye9KrQlb5rHNAVbbyXcTR6nQIfeC84R3CjV0HkmsW5w0pgETlUrNSEODTX0HaDN6rm+E9ZSAlv1dpzasToKQcp1Jzgp3mpyNJLL38ayOgyF+jWvW2UH2CfA/rbGSiGw910S2joNKsusssTsF9JRkyfCSU39SmkfGLqKcXoMe2PDbTOz1Y3PFBycv5+Y2S9sc9Ac9/vcCLnn9gPdcRzHcRJpPQ/0ZtfQZ86caSNGjLAuXbpYaWmpHXXUUbZ48eJgmc8//9wmTZpkPXr0sM6dO9uxxx5rK1eyvpDjOI7jOBtLs7+hP/PMMzZp0iQbMWKErVu3zi6++GL7yle+Ym+88YZ16rQ+9nXBBRfYI488Yvfee6917drVJk+ebMccc4w9//zz2X3ZtRZJI2slpHnJYizIKYk2lUy/0rQ7vxa6jghD7v0kTMVZmxhO1lAe09ZQUDScqgsxQWStxco7KoxSBqk0sfQnov2EOH8qNLViZyaVSn99MsUpqJ1pZmY9EtaUFFanAEIxQnsmQwlUhfV9RUfhMeVx0X4pgY8pj0Fo90g4Ox6AD+Q6qX02dBWehmV3tMZh0qMwPjQLkFY15M6ozZ7mN+q38CgthR2cH0l5oGZm86LmGpyuUAiCNFFeezx/tQD1GPiOhh0sPDB07YcbwieiYj2B1bDyq2RH2gqcaH1Ys3WqtE/bPfSdxnP780baZp1itvztf5Y1NNvXWHz6yBah9byhN/sD/bHHHgvs22+/3UpLS23hwoX2pS99yT7++GO75ZZbbPbs2XbggeuFmttuu8123XVXe/HFF22fffZp7k1yHMdxWi3NMaAtNwbFtXja2scfr//N2L37+p/JCxcutC+++MIqK6OZcQcOHGj9+vWzefPmbXAda9eutZqamuCf4ziO4zgRLfpAr6+vtylTptjo0aNtt93WV1Grrq62wsJCKykpCZYtKyuz6uoNh+5mzpxpXbt2bfjXt29sHinHcRzH2QDrmunf1k+LjnKfNGmSvf766/bccyyKmB1Tp061qqqqBrumpsb69u1rf/yP2Tb//axyWrR84SVnYQ1MyYnNI5lANgdSuxMK4KFYVOYf7IO6lfsjTUUldhbofA32T6Te5EmoPfkolk1Jmz3SG/Z+agy1DIiGtgo5LdgmHR9AXZSoel1BZ0zXV1Eym+lSU7Ax761tJ+1MqXHyvfzpLDvLaTg5661COZiZaaV62p3Ov94uNN+TvKX7sOh3KLCeIu0O8K2A/buoWYgz6arwBOghOXsDkW3KfVN5O2naWDNo6BxOwVwvEaKpmf8J9vvSZhCWwzh05M7B8HGMSnDycxDNMaF5mFw0hy3Csjx5tPu5ERxjUTwowcmCtHp9cW94vcndpdt7UbvtZ2Z2srU868ws3cR1tPKQ++TJk+3hhx+2p556yrbbLrqRlJeXW21traVSqWD5lStXWnn5hucx79ChgxUXFwf/HMdxHGdr5vrrr7eKigorKiqykSNH2vz58xOXnzVrlu2yyy7WsWNH69u3r11wwQX2+ecbP79Esz/Q0+m0TZ482e6//3578sknrX///oF/2LBh1r59e5s7d27DZ4sXL7Zly5bZqFGjmntzHMdxnFbNlgm5z5kzx6qqqmzGjBn2yiuv2B577GFjx461VatWbXD52bNn20UXXWQzZsywN99802655RabM2eOXXzxxRv9nc0ecp80aZLNnj3bfv/731uXLl0adPGuXbtax44drWvXrnbmmWdaVVWVde/e3YqLi+1b3/qWjRo1yke4O47jOM1M84XcOSC7Q4cO1qEDJaj1XHPNNTZx4kSbMGGCmZndeOON9sgjj9itt95qF110UWz5F154wUaPHm0nn7xehqioqLCTTjrJXnqJknHjNPsD/YYb1hcm/PKXvxx8ftttt9nXv/51MzO79tprrU2bNnbsscfa2rVrbezYsfbzn/886+962SI1T3N3x0//LFzwMmp8AxLWms0vsaRlu4bmrv1Cu/syawxqccOlDRU0NlXh1dL+PXyU11Q3Z9lNKmjlMjNsrCpoDMmGRS58LTR0HT7A9GCyjbTZDzaIH3D6TyVJU18KHxKT7SBplyV8B76nM4NhUU1R5p0njSWgWjmcC2jJzn4oP2xzQ/O70maRgoG1oT1ODY4dQOnXW+Vv63DAJxaG9k+iZSuQ2L0H8qZVkFuKLWD+dUAK9oLQ1BR8auacllV7hdI8q10cJ20ep1I+A3Sc727wHQZbv6i4FM7BsHVlvO+VwNY7Ao8xzzz1Z5qeWv1dE5bb+uGA7BkzZtill14aW662ttYWLlxoU6dGyf1t2rSxysrKRrO59t13X7vrrrts/vz5tvfee9v7779vjz76qJ12GmtCNE6z92g6nfmXUFFRkV1//fV2/fXXN/fXO47jOI5QZ01/Q1//A3z58uXBGK7G3s4//PBDq6urs7Ky8Ed/WVmZvfUWfyau5+STT7YPP/zQxowZY+l02tatW2ff/OY3swq5+/SpjuM4Th5T10z/LDY4u7EH+qbw9NNP25VXXmk///nP7ZVXXrH77rvPHnnkEfvBD36w0evIvZiH8HeLKp2qsjH+Jix42S34YLq0M3XBpuYfMgx1eGiWy5xJCDWycK0GLRnm4+xbGrZmSVGGqXVGKobYD2bemsYPC3aGM6EQ58uhhzNS6b5BKImhPRrLh4jJAEmhvaTSr8+Erk8fCu3O+s0MbyaVgkUaY/u3G5rMUuJkYDq5XVKk1szM9lcDss/t4Qjb5yW7jOfcGb/GB+M0rN6H3xoif7sKMezSPyGUf7O0cd86EPVRfyNtnkeJcg2zZt8PzSelzRTCtbD1+uMQXiZgHaavS4zHMyevUtr7w9dtX3ygK+M5SKlJQ+OcJpF7l3SvY9pW0gyFREdp1zXSbknWWdPfXXlVJtOzZ09r27ZtbI6SpGyuadOm2WmnnWbf+MY3zMxsyJAhtnr1ajvrrLPse9/7nrVpk3kf/A3dcRzHcZqRwsJCGzZsWJDNVV9fb3Pnzm00m+uzzz6LPbTbtl3/I2xjpGyzHH9DdxzHcZxkNv8buplZVVWVjR8/3oYPH2577723zZo1y1avXt0w6v3000+3Pn362MyZM83MbNy4cXbNNdfYnnvuaSNHjrR3333Xpk2bZuPGjWt4sGfCH+iO4zhOHrNlHugnnHCCffDBBzZ9+nSrrq62oUOH2mOPPdYwUG7ZsmXBG/kll1xiBQUFdskll9iKFSts2223tXHjxtkVV1yx0d+Z0w/0Ty3SkFWNeQ/TkQ74273hB4PPE6Mkw7dsqobOroWItr9o6L8IXZ8gC0gVqB1CVyytRtVsnsJDYOtIglLO78ioUKCpL4UzBVsGBUCvZIqW/iW1ZP4m1fECxRw8MBR2oB1mOs0l6ek9aOZ3YdEZerBQkzNRQ4fW2SnS0OtYXTYBJH1ZMUX1YCwBpiO+NDRl5tJY35+BoQRhCVyWAcV+vxI1/w9L7vib0J6YEuPG0Fd8UmiPEW2eyYQp2JqZdgjLAr8Smirz8/xM6t7j4DuMmV46MIWpZywF3U2vbGqsPK/0ql9qyejN5N/wsQKZ7gDHSXDn9L7I9XB79W91WZ51+cfkyZNt8uTJG/Q9/fTTgd2uXTubMWOGzZgxY5O/L6cf6I7jOI6TTJ1tyht2SFPT3jYP/kB3HMdx8ph1ZlbQxHXkxgPdR7k7juM4Th6Q02/on1uknauGjrRuG8A5EQfrEknaUCaSuo8+6GLjZMa4g8P6wHthe3UqSJaFpXqlJTCZHc7M+FJNU+aKmJCrWvi+yCUeDNFfJ3mFbMdc88YyVM3iGnqw/eyIYpY5zTS1qSK1ki8LPbV3hnZhiRyr89nDzAHWLUY9z22ig7xx41fXE/sFXgJb51d9NuztJ3Fh6NS7XUKXrUHF1o7BVTUaS+PMkoPK0+jPsIv/GLVPuBrOr4bm2aKho7yBzoBqZqF2//cHQx+VW01T5xFlSeQJ0h7Hyr8U1Q+QNgewUKx/RS6w5Rh4kjTZFhPwefNbmrAsixiMkDbrUCyFrQcAYxK4DXXS4W117Mtme+ltPW/oOf1AdxzHcZxkWs8D3UPujuM4jpMH5PQb+lqLwrQaRmM0izN+BSHWTGUsgy5qSncxAeaEqHnczYGnLSLaA9+N2n1ZoxNo9IvhzmraEp8vQwyTYclCDeV+A87BlCkkxQkx9lBcCItPMuS+DexAtIhNdMZ6mUnHCjHM2qjm6DKE2H+Lv6yaKcb5L8KbNIsffCIZdH0zdGX1K5udpDwcmk/CrdcJ0+EojXQMYq483gi56zZhRYwI/1LaeyJtbWdM8NX22Kh90e9C3yKsV68DpnZyX1PSZjYkK7SOU5mKFVkpA+nOUiNgmFrrNDPuz/WqjSkUV+P+oPtOyW0wryHt79fgw4Vwr9w7OEMdy/IqveXGUtv4Ys1Lur7pL9i58YKe2w90x3Ecx0mk3pqetdbUv99M+APdcRzHyV+iydKato4cwDV0x3Ecx8kDcvoN/XOLdkAjIixwGBfVRZS2lXSCEml3amyh/5JNGpuUz+TUn9SvZM7GThDC9kANTM08eQereRf2o9JmBguzVgarrher2Mi8GhHykCoDyS8mFyolsAM9cyCXzjSVqYJaq9dGTVZ6nQd7kJwuhwTFc82CcRFmljjhq1T6rHg2dPEs4ySXiWgHo3wrzweVt/kCEn8hWSHtpDwqC06mHdDVHNehGu8N8F3LD2SYxM54FemLkKjK18zWopasMLLKcSdp0Y4Lfg9nrFxuxH9w4rMf1F0B35CEXM/af4SuBaEZ3PqYghe7wPSwYtzRrR+Htl4nSLKLjb9o34hvU4tqZ00rekPP6Qe64ziO4yTSijR0D7k7juM4Th7gb+iO4zhO/uIh99xAIymJOY2xXF1V0aiSMdFbBe4M00YGqlDSVJpmZttFTeaDUmhWIaxr6CrF1JBHiHjIcpio5hmUvESVzbjOqLVBY3EdlH6tEVUNiefsbeptSgnsYKjBnlx6O34gUK1bGJpzouarWJL6oPbZITcjDjeR2rL2Ig7c0KhJbbMEtp4OscgfxzNoEjDGV3Aoid6jEsvsmll4jLmfOP4ykOMIdCiPv6ZnPwffHJwcR/whamPG4dhxUipgc6ZgHUXD4SsvwdbrpAwHowuvWwESdOz5UKHrge8/OMb1optTM38dtq4rdkx5GEU3fxIDbliPQe+SmUpM6+1Xr4LNNnmqh9wdx3Ecx8klcvoN3XEcx3ESqbemh8xz5A09px/obSwKMSQkCIW5XGYWznz1Xuj6D3Jluml9x4uxnqQEmExJGTIzV8/dQ1fqr4FZJ6FGpuTxPEtJmyUuk6KzTFvrDjvIpeKKuVWLpI2wKaOSKpUw7Eu7RI0d4ExMKeSxmB1YdRIWZkiYpWqD0C7TliYyBbJE2jhXRDKAahILuWoYLXZfSsgYfAv5bkzf0nOHJU+5DeH2M8i6KDR3jJoH7xi6tkUoV0u//iF0ceK7IMxO+Yhygm4tiwL/hOpMKmpeA/XgN1h0kbQzhYz1/M00k6CGpZn9xmtRvzcFH4+MZLzG7h0jUIdX0xp5LChF6DbwUuTtVlNgddk1G/ieFqEVaegecnccx3GcPCCn39Adx3EcJ5FWNCjOH+iO4zhO/tKKQu45/UAvtGgHVDsYxAWZkxWolreELgp3IySX42ROxDga9rpG2mZxvV0139ND117fDsylIuRxqkKm66gORj2Ymt9SaVODjFVWVQE2liOIZBlNVYJozm1SCThT+dEg8Su2gUnjGSCM/jWcu1KPKiVp9lmgQzNHKPaBTpmKS21o1OSYD+5JkhYby4eSPCamXFFD12uGmnlBbHpanWYY/fmvp0JbuwEX41AM1rhQ/hRZdjHdVovPso9o677FygvzIEvF3irkZxWifx+SNhNc2b86fITaJjV0Pc+4HqLLZnrOqEzO0r8cH6Dr4v2A6aV6epwP33gOCtGazTIgoKbW7OxfWcvTih7orqE7juM4Th6Q02/ojuM4jpOIa+iO4ziOkwe0opB7Tj/Q21u0A5pHO5BCwpeK8YFogPeEGa2rZoVLlkq6uJ18Jdbzu4Sto3LHM0I19KGh6/+F5gCZ3vMFpDrzW1SyznQOqk72MHz7cWFNlGV91MORyy/C/jKI0NTmkqYGpS7aTQ/yYNbzTTqVkYH7YGiqok6tkD/Mgz7lgAB7BPZYaWP7iiNxuVOHcDLNLugUPZ3ZX/9G//YQTZplQXmu6BkYS+vfix9o8jbm7GQy8X1RMw25vQAJ75pznUk71pzqUfDxClf9ndMG34spXY/XggyYEXfyFaG9rVxgLK1MTV37m9uHQsCBpk6Jn7Ub9GzmsknnK5eNzSqdAG+piTn4vC7U1sEam632a+shpx/ojuM4jpNI2poeMk83x4a0PP5AdxzHcfIXD7nnBh0tCrVrOM6Gc8mxsB+PmpeGnuux5A4Snht/I4Jf32TOUGdpL03wmYUpTQjAtftZaE+f3NAcOyl0sQqrpkCxwiXTmDTkxqgpI65DJMo6gvkvtUiek3qpDKMyZKwZcNyXWPnRYIeYMpg08x3SyVBbU72xFCegodE12JmO/0EQtpsGOSkgjIya5WHIvTdKcmpofEXosh/DHigK0ivwMcKpiWi8QuIf6Dn6fOj6c2i+LWF2nnNvYCP0ULCcK4v5agnXn3JZ1Ef9lVyqnCnscdj7/jFq9zkBzktD8wSRw7bBviwPzeBc4Yx6nLdRlQiG2JmaqnIC+4zoc4jXIs+HeLnfCKYULpX2tfA9g+ui+7yoPUjaaxK+z9k0cvqB7jiO4ziJ+Bu64ziO4+QBrShtzQvLOI7jOE4ekNNv6L0t0l2DCpPUwawiNJ//UUPzSeSaQA4MNMvx98P5zRQ+UJ2c6iFyZexMaQ+Ab1honhtN4Vr6lxcC13E3hYtqAlnSZJ5moW6egm8O7KA8KcVCVsQV8ZDTchL94dsh6TtjHwzNsGbVr5FOhjEAmlXD1B6iv4CZ4tbxOXwwLiUG90aO+W6hZyQ0dDXDo292F+wSaVOLpUa6h7QPZh7VEbB1dMGnuBDCIQDB984LXbHtT9KAOaWvDo2hZs4+PPnZqM1jiu4NNr8P8zenwP5e1Bx3Xeh6Dx2u5wc3tzfsAjn5t4UGXYJl2S8Kv0dHblBDZ7/oNhXjVe8hvKHqKB+mBT4JW8dC6HiAzZa15iF3x3Ecx8kD/IHuOI7jOHmAa+iO4ziO4+QSOf2Gvqetz0U3MztOHRO45NLQvCFqPhR6YvJwkD3M/OvYFKmaeYps10dWhfbhqnD/BOthzvp5UXN6qEJyGsadRcDqA3GLmpnqWcxZpi6mOuNhFGdjycZRE0VCY9ugvyiZd7wH7LA+KedoJJIM/R9M74nSlKrlZfohnlSiM1ZrdZyqtT3hFE19n9AzFpK/apvMQ+eErVoZgbL4CNjBpL3HwTmAKq/knnMsCQZr6LiJpCqgZmEkk9/IchIj1eAUvhii0PabUXsiz0/WaNXBBew0FkeQS9EqQ9cA1DcIDg6qD8cuXOm0AojL5fjbcu3vTPVy5QLrwROWRSF033ExjsHAA70FsHuTjrFq/LFZmFuKemt6yDxH3tBz+oHuOI7jOIl4yN1xHMdxnFwip9/QT7BoFqNCCbFZt73DBf91b2DW3R21WdIwsTxi7Fcau0/svyLEjhnU7H4Jcv8SuWdWBVuKdPY5OnQdh/inxC07IQR8MOq7viP7w5mXEkOlDN1RipBcJKYpMfKolMCOTfg1RI1YUhuQuCRCoSzZmk00TiOlsb9jKDcIRg6GT8QcTHVWjLqgo0UH2g+5cpRGNIzJcqNnwR6tqV6xNDWc24vkSLJOMELEGqVOYVFeQhqC3R++s2EPKROD8fh9YY+R9vctGX2t6ckYO/QQDfyP7hO6RvPCUIGER4riiV5hqdD1KYLTHzXSNgvKLptZGFbn5iXpX5AEuuF7xkuq6nikLSbqbLIrNfVmv2IOYUvgo9wdx3EcJw9oRQ90D7k7juM4Th7gb+iO4zhO/tKKBsXl9AO98Ayzwv9JXuPVszRc8P9CU8tlsjwqCUqXltDLKTEljQ2ZUo9BXy0U+8DuvwqdV3PuStVfMW1oe2joKWmzRiTk93PlT/+JE5aZSYF0y/wi6GBLxWZKCyecVcWSZSshLUM3ZWofBUER+bARHCfBDKIkEkNasTqmmrp4Mnxy6XGlmMO1jjVmBZbL1VWx/77MhXVMAo/pkmWhPVvaHBgB3VYlVfY1SwHrOIlL4Cs/AB+oyM4BFtxZPfeL+8HJhMiB0i6DryTBpo8dnHR7Zcprgo+nutr9MsWCk75nc6E5cA9GzZq1Zl2viy3d7HjI3XEcx3GcXCKn39Adx3EcJ5FW9IbuD3THcRwnf0lb0zXwdHNsSMuT2w/04yySZ3aUz59ADvhtoanT+6WwSuqpJWpsyw1g94lCDN0WhWAD7X7bH4W+Id84NfxgZ00oHxr6KDzLAIH3MGNrCRbtIaUej4C4zSkQ1f45xgcMCc1gjAJTVJMqdnI9PbbDB0HtT/Y9NfS5URPjF5LSbzNpUDpqghnLsST7f0lx3V4QxnX7MRQjjTxePfpLsZakKShjYwO4sArazGfmgZPc86UYM8Fz+z5pM82YU7geKe1yHm/k4wf9xO1jQQk9kL0xHmAI7EFSALpjKVbEkr1qcwNZa6BC2j3g4/ibdo20N0Q2t+2WusUnrTfJ9zVpf2pmrqE3J66hO47jOE4ekNtv6I7jOI6ThKet5Qh7WVT7VSsnIsT+BMKJWqmU4dcS2EH5zEwTfJnEuN8KPcxo0ojhHfD9+Ex88GcJHxtS2nZHzcsvopyiu0JPLCpZxBwygaVgVSK4DD4qEdqnTFtimFqzpQ7lRjB7r7PuK9NxkIBY+0TUXhq6eG1mk7amUepYeIv9qR3ei/UwS6Imqthy+/TcYQZb0n0mBXsNFu6oBkPuD4bmWzJz2A2hy/4MW8PsvL6YthacZ4zH32mNshodwe/poRlk1HKYtaYn8EeQ65bD1hN622dD35jQDPSEAaxN+xXYWoeX4XnmrWm4PqH8dJNozrB+Y3LCZnr8eMjdcRzHcZxcIrff0B3HcRwniVb0hu4PdMdxHCd/cQ09R+jYzaxjwfr2ShEBkXP1KP5M1Vb+8GIWWJApFatFSh33vQ1/icUzhlLSprb99nOhvfPCiyJjGCeZPC80x0caem9khGAWUZNZOWPbR+27LsFHDVr7sBN8LKypyuKXmSF0EuwgLShxgtdwDAMGBDTlx7bua+wa59SVOljjkPfglH3B2Iy2SN9qI/I7ZWYWG9XZMrl9HbfBB9ovrNH7+9D8mbT/hEV5JBRq5kze+6W0338z9JVg2doEHy/NkdIR/ZhOuBy2DGhJzwtduBSDfWW13NHM39MSuXuhXm5f2HoODMV6ijEddFADl6lynFZYclMTU+U2ZG+sLxN6n9TRDmu5oNNEcvuB7jiO4zhJtKKQe4sPirvqqqusoKDApkyZ0vDZ559/bpMmTbIePXpY586d7dhjj7WVKzNNk+I4juM4WVJv0UN9U//lSMi9RR/oCxYssF/84he2++67B59fcMEF9tBDD9m9995rzzzzjP3zn/+0Y445piU3xXEcx2mN1DfTvxygxULun376qZ1yyil288032+WXX97w+ccff2y33HKLzZ492w488EAzM7vtttts1113tRdffNH22WefLL6lrTX8JhFxaw1e9t8PzUADTtJ/zZCyyulI7VPYi6ImEmOpJer5wRx1VFa1nWeI8fB98H41NH8S/Uab+MvwLPwEubuq3TMNmboo+1DZCfYUaVPRK4E9UHVzzp95EKe9VC1uboLPwoOMY8Gc5WyiaTrWgOtJ43QoCA7s66FTc42ZdoxhEkPubnwt78DWr+R+pXH8C/SgvhL6foYbmI6/YI0C6vp6mbBGAf9Wt5dnNtEaBrwUj4MdjN2gVIuysTUyRuEPoSsou2sWqtA8bDGJWusScOABL3o9XzkwYtv5od1XbF58g2BrDj7z7wsOxgd67x0AH69k7WHmySdp86sbaTvNQYu9oU+aNMkOP/xwq6ysDD5fuHChffHFF8HnAwcOtH79+tm8efO4GjMzW7t2rdXU1AT/HMdxHCcjTQ23N4cGv5lokTf0e+65x1555RVbsIC/b82qq6utsLDQSkpKgs/LysqsunrD42Vnzpxp3//+91tiUx3HcZx8xtPWNp3ly5fb+eefb0888YQVFTEGtWlMnTrVqqqqGuyamhrr27evBaMVJITFsB4rcmr6C9OqGFgKUlNis62lYD/PBRpgWFJDI4w3MI2t7pGo3XYRpmYbOgxLXxM1b58SeKq+FS5ZLdIEZ0FjRFCj/pzoihVxj9JOGwnnjrArpB1LU2PY73dR81/otV47h7YeyJLQVY9Z6DSMnum6/aSRtlk8BN8xyAXkvGMyBVwKLvTZADmhJ/wx9DETTaulvgYfS/YeKJuAJCq7H7ZeU5SlRsDW8DfL+b4L+xfSfhk+Zpdp9lkKPgZv1R6Bk5nr1dcOTMwXW6+mWcayWHmRS6Zi7ZLQRalEzyXeNbfB37aViHvSBHpm4f2tK3wdOzwRftBDbN7rOBOehvo56RxL7apdrDJajjwlc4hmf6AvXLjQVq1aZXvttVfDZ3V1dfbss8/az372M3v88cettrbWUqlU8Ja+cuVKKy/n43Q9HTp0sA4dKCw5juM4TgZaUdpasz/QDzroIHvttfDdYMKECTZw4ED77ne/a3379rX27dvb3Llz7dhjjzUzs8WLF9uyZcts1KhRzb05juM4TmvGH+ibTpcuXWy33XYLPuvUqZP16NGj4fMzzzzTqqqqrHv37lZcXGzf+ta3bNSoUVmOcHccx3Ec539skUpx1157rbVp08aOPfZYW7t2rY0dO9Z+/vOfb8KaPjWz/5Z+FWGMqT2phDWUwGbWR4HqwRQPqYtS1xUoJmgKDrVYanyqtw3kXKtDr8YHX4uax38pdH0eTvdYPkcM5KVtizKc+0mbOig1/ztEs9wLYuwQlnc9VdqxX8EYk3CzKP380vPeDu2di6P2oPC4dMAOUHdUuEmqk1NfjU3DGoihHPApO8CyqzyBRa/sg42tgqa+UkTV2VgNpz39tbSp//OcLJH2YfBNgD20vxgDQ98Q6NlTpRtYopmHWG2mUVKT1qrHscsW6L5yetokKBV/gNuBjjXhIeY1rv3P1CNq6pyCWOHf6unC87Mt0vnayLEpwnEqwcFQiZ33Ng4lUHtbW9bQZtJvi+GD4pqXp59+OrCLiors+uuvt+uvv35zfL3jOI7TWvlfpbimriMH8PnQHcdxHCcP8MlZHMdxnPzFQ+45wqdfRDEG0a9YziYFW7Uk5mbuBTvIsWTSOlU+1Z2Q2F2Cv1Rtizng1C8DrZbpzH9Fzc7dpY5lkORt8TxvTS6GZsb8Vh1KwNxn6oGaW8zJXo9EDvg+SQm4z6Nnroia/0Y/9KAQfq2UoxwTaujFD4aL6mFlyIqROh4rpZAVMFW8XbMq9LUX+63QZQ/D1uPEOTshWG4rNXyTKpGaxY+xwv3WY34AfENZiEAvItYhGBqafWTfJ94S+lbg8lJN+rdYLacG1lLGrKnAUyU29kHg8dY89ZnwUTtWXTwFH6vR1jXSNkveviSfWXYhWF0X/466vZ5bTChOWlZPlaRrqVlpRaPcPeTuOI7j5C9bsPTr9ddfbxUVFVZUVGQjR460+fPnJy6fSqVs0qRJ1qtXL+vQoYPtvPPO9uijHC7aOLn9hu44juM4WyFz5syxqqoqu/HGG23kyJE2a9YsGzt2rC1evNhKS0tjy9fW1trBBx9spaWl9tvf/tb69Oljf//732Nl0pPI7Qf6hxbFrlZEHzMqzZQcDbmxwiFTUYIZihij/DRMAwviagjPFyFHI+kHH32FSU7m77wvodwuCPNSTxgjbdT+rHs1tIsaaW9okzRCTOkpVi5TJQ3mDP06NB+VA8sodRUTJiZLoBW1ScsRlyqRjWQ4lodc9ydW2LgPbA25s16u6ijc71RoPikbUY2Teyn+VLPYKN3wXOfkWwpLoGoI+xb6sE1DxB7B6csmwf62tL8RuvpcC1ty0YowoyLD3Rqt5/VfArtM2gxh8/JaKm32b8oaJ5ME276R9tYCw+N1CT72oV4neg4mST7NyhbS0K+55hqbOHGiTZiwPrHzxhtvtEceecRuvfVWu+iii2LL33rrrfbRRx/ZCy+8YO3brz8LKioqsvpOD7k7juM4+Uszhtw56+fatRwNsZ7a2lpbuHBhMKtomzZtrLKystFZRR988EEbNWqUTZo0ycrKymy33XazK6+80urqNj7e7w90x3Ecx9kI+vbta127dm34N3Mmh0au58MPP7S6ujorKysLPk+aVfT999+33/72t1ZXV2ePPvqoTZs2zf7v//7PLr/88o3evtwOuTuO4zhOEs04yn358uVWXBxVoWzOScPq6+uttLTUbrrpJmvbtq0NGzbMVqxYYT/60Y9sxowZmVdguf5AX2qRVi3CLbUtonrbUPjKmYKj0wRSuEXGWJADhykFy6D5qa5EzYkaWqAP8sRkTo7WwGQJ0TGwj5D22NDVDVGh7vI9JVgNt1+HD1C2L+2PD3RlrI8JDf1xafM37hMQ5A6+SwzsG4/NjlLmlpmJlLc1pMX0PevLDwTmVWmuHztwp9DsKoLwfViU055q0iLDb8Nh/1Tapcjf/AVy3DQVkac9y66qTroTOrASs/+erCmEj4Q+u3jf0B4T7e1+0OKLcK5rCWdeMqhGG4wloBbPoQ+aUchxBrzv6PfyEGdK9dpYeIyzkXr5t20baW/M3za2HrOwT/W4fGZmv8/wPc1C2pquoafX/1dcXBw80BujZ8+e1rZtW1u5MrzxJ80q2qtXL2vfvr21bRv14K677mrV1dVWW1trhYWZzxIPuTuO4zhOM1JYWGjDhg2zuXPnNnxWX19vc+fObXRW0dGjR9u7775r9fXRr4+3337bevXqtVEPczN/oDuO4zj5zBbKQ6+qqrKbb77Z7rjjDnvzzTftnHPOsdWrVzeMej/99NNt6tSpDcufc8459tFHH9n5559vb7/9tj3yyCN25ZVX2qRJTA1pnNwOuTuO4zhOElsobe2EE06wDz74wKZPn27V1dU2dOhQe+yxxxoGyi1btszatIneqfv27WuPP/64XXDBBbb77rtbnz597Pzzz7fvfve7G/2duf1Af9MiMVoSZZl3Sk1HNb4j4IuJvilps2APE94PbHw9ByJl/SZpfxC6YvnNgaZOoY4in+itNc+FrmLquCpE7wdfZWgOvTNqM5ecQwtUM4NcHU+GVi0ZSb9LP4ItbXYDhxIEgnsZfNCo9xMN/c7QFcst1mMTU9IowOr+/Ak+FWCZEI7t1c3luc1zR/2cNrQCdqmWkUWJ1pMw7e1SaTO1nNuQknbSVMBmZu9IP8zYHc6XMULgS0dH7T+GGzjiO+GiI6Q27CpkFnGcRCcdP1AS+gbjRNsp4T7DsS8aJG1LJw+Onju8AfAGpnHVpC/lsrH5U2GrsJ9Uv5Xfm6lOrO7bv6NmTa3ZRM7xm2dMnjzZJk+evEEfZyE1Mxs1apS9+OKLm/x9uf1AdxzHcZwkWlEtd3+gO47jOPmLP9BzhL9aFN5JyFVjREgrdI5g7hEXltBtNcLm72HRfSR1pu3Roa/HrqE9SMK8DFmzJGIQ2mMMk3G/d6Pma3B9hhXvf2PULmSNy4rQLJftH/Nm6GPqlH5N7LAwNq45UNCpVoRmMFvYavhi15tKEQgfM+aqEgITStjdGj2MjSjlzkoa3mqkAeqx2YdSCELwqowg+zE2a5dGQrkvzMgMYuVwFp8Z2udLvVeqJpgULVhtprC0qlazkeJ28ulY+AE5kH1wgd2Fq/H7f21oli7CejjtnN4DSuCDjFIoYepChqEpuWhYvRvj0DvC1gPQEz7aCRqBcco/tbnBtPVxwHg8HxXtNtJH+29Rs+Yzs9lnWIvTiqZP9VHujuM4jpMH5PYbuuM4juMk4SF3x3Ecx8kD6q3pD+QcCbnn9gP9RWsQDV4T/Y1SJo9FoOPxQD8YmgtEEH48dMW+R7XDo56Ac1BoJk2RmIKtGmrRktBHWzVLZtUxvUjZ5Y+h3Y9SnJQ1rYCLaUCSmRLT1wf9I7S7i12AVB5WrtX1UptlKdhVopMWYd+KsW+Fmq4DUZrjGVTmLWX9Tg5akA6nTK6LfoQTdDdMXasVcDnUgZqZjgdgxVsUUg3h2Abkc5ZXRO0qntsJc63WoAN5Dqo7VjmX41s0m+fYv8F5cmgOOE/asYlugfp5S+Tf6kHPRjvOtOzG+pr6t5u6bHP9rSayftroUs6mkdsPdMdxHMdJohUNivMHuuM4jpO/uIaeG/zg/SgApllKTCdjaPzP0j4ZIdakycwYLuSyOlnYHxFN6j0/tDUUze2j/f+kzZQhRiX1hyTTvrjeHgnr7Y7t7yKpaqz2xdQuTSm7DT5OqKZR9npUhmM6nx4L9v3dsDVqzT4qwr7p9jJqzvS4BdI+FedOX+SU6fnybugKJAJuXw/YuizPwe1gnyXtMxiPR4W8oPNZuYzx79Mk7eqSoXBiqjahOBBKzIqDmnNmVisHvRZ/3Jk7sJu0h8HH1C5N2mOPJqVrba5wd0vRUtvQXOvNpq+dbPEedRzHcfIXD7k7juM4Th7QikLuXljGcRzHcfKAnH5DX2ZR+pdmzlAr5o8r9TOtisuqTEqJj8URNaVpaRbbwGgOFT7Vwpmelc328ntUk6aOy31TUKEzsbwnNWiWCdXt53poa4oT+ygFe4E1DvtM18t9I7osU9GSlmX6m8IURpZ31X7gtveBfYYeyN3gpIau0jJKE8frxA6V9kHwcU491a+Z1IhbTuE6afNsYVqTLJsxvUy/t6XSs7aG2+fWrpmTdY20W5BW9Ia+NZyRjuM4jtMyuIbuOI7jOHlAK6oU5xq64ziO4+QBOf2GfrxF2m+JfP4wlkvBrpD2XvBRf18qbWq6JbCZyqu8DFt/8DEHfCRs9VOLZQ64zkaaaerKvo20N4SWHE3Bx1kjVbodAh+3Sbf/FfioZ2v/8riVJHwP+4zjEHT8BX/hVsDWY5OCjzn2OksnFV7NHx8IH18mtITvUvhYa+At6bS+qH3QCWV3bU9pY8rW2IXwtqzsA6w4hWVVvqZu32cHfLC/tCncM7e8RNrZTAWaDZl0XV0vl83p22n+UmdNf3V1Dd1xHMdxtjCtSEP3kLvjOI7j5AE5/Yb+lb5mxf/9SbKHxCU5I9UbsDVs+gvWzkTO0EMSr+XsZZhAzUZJuxC+E2FrutyO8F0Du1xjzcjseQ/lR6dIexHWwxSnM6V9AnwMo/+ftH+ZYdlvS/tgZjQhRrxMMpOmYdEnYWv09ufw9UF09j8fRm2G7hly/5a0KWEMh/0LPVgoVTsVts6SxuQt7e8L4SvAz+yn5e2AfcRz/WJpHwffLphRbZhqOSks/CfYomOteSp0USrRS4jpkIOwxf32FPscLDweduHxYlAjKLGNJynljdMM0k6abS2bsH9zzr62sWwNt/stlLbmIXfHcRzHyXE85O44juM4Ti7hb+iO4zhO/uIh9xzhy9YgVve5M/q4JKnOpiEjh8t+LTTHPRe1/wMRvRuzajRHC3VBt0HIRs8vllllGpsdIG3kfbV/JLR1Iktq5gfAnqjbPwJOzF26y5KoTV00kd6wkZPX79GovceHoY+pfgpT8LjebjK2oBuF8cWhqfo2jwW17yB3DilZB/8htDWNjafZvtIu4IHBzWPIs1GbSV/U0PU7eZxi1Vy1E6GvM/dzjvTZfViUY0v0FGUqJ0+zs2Se2wGvw8m5Yvs8L8bvQt9fcYHpwBnuGw+GXjS8+Epgt0/wcZZWXVchLwSOnCmTNm8stHWDuRHU/JPGB3SArY+DTOMBNnXKWT071nLBlqEVPdA95O44juM4eUBuv6E7juM4ThJpa/qgtnRzbEjL4w90x3EcJ3+pM7OCZlhHDpDbD/Q9zazjf9uiX7aHZpZULnUR8s6HMmn51KjZjQntXa1xIMZ+gl+Iuk1IX7bXYA9VbW7P0LctNPQSaTNP/mTYum8x/Q8auuqk7M+PYWuO/cFL4Twd9r+jZp+E8QBmofpG7biUtWsrpP1o6EpBQ9drlftGGTc4WOjQPaGh6yalsJpAoeSBwkYUiYZegUVZA0C3l6crq7kGG4U+mo0+0toDcMX6TLX7Mvi4qwP2FuNMOPtgFMCbcmHfhmXvD82ad6N2CotyXEQPlZKpoXPnUgk+oveALrgp0dZd5UAO2u0baW9oWRVVk2aYNQsvuHL4OHhD60vwoPbkeIHB0i6RdqYObCZa0QPdNXTHcRzHyQNy+w3dcRzHcZJoRYVlcvuBvrNFYSMJlXVimgpQ9+/hG/oEPjhU2qPgQ7je/hw1/4NoEoNLen6gmqv9hdukcdT9Ql/H3UJ7pKT+MGo25Kv44Ahpz4MPfagReIZuOYOahsNXLwl9nRjmk20YhJA7o58a2ud3xqYs0zDg46GLp4fuDyNrse9RvQaZR92ODu0REgaGghHuG8OZ0GA0qspKxSyAqmlrPD3RDbaXhKVZvvXXsN+RNvuI2zRU2ifBd/BX8MGV0ubMbE9A/7ohatYgxM4ywTqxHJOz+DU7SfZUe8xIxxn0JIs1JsfUwm4rF3l7HNNC6mwJ6+GzJCmsms2yLE+t/VQCHxW5sgRfF1xhXcXWVFre91oMD7k7juM4jpNL5PYbuuM4juMk4SF3x3Ecx8kDWlHIPbcf6L0sqmYoog6TJqgrqGbK1J63Pw3tnVUzpSDIP5YpJ5n9Rl1Mzw8uuxR28D2s54k5Mk9T+QrlUO17sDWNBfuyAlUZtRu4vUT7l1OVDliKD2SKzCFXh65B0N91Gyg7x+Y5Va0eYidT3nR/+EOcYx/+oyVlKcZPCM2TROelfl2hKULcGdw8CqVi5344P4fiT0Vmjn3nb2HrWA2OFWA5V9Xx94LvSNhn6MKcEpXXkHIt7DtDc5Hky3FfFsBOkKhjqX56qvC+TZ1cbZ4bTbnna7bZ5np2JGW4UV9nxpvq7VyW91v9W03l5D3RaTq5/UB3HMdxnCTqrem/kjzk7jiO4zhbmHpresg9Rx7oPsrdcRzHcfKA3H5D72Fmxf9ti3C+CxYrhq26bgq+mPamgmvnfqFv6LLQFlGIudqcKFAjQNSSmD8ciOoU9ZD7HCSI7gsfc7XvkvYLoYvlZ7VfGL0qga3TdMamOWXd0FI5OpfXBK5TTwkX1Vzpfsi/j+2rLgytOyl/mPvGY6P90I3zu14Vmj2mRO2Tb8Kyh0mbJyw34sCoOZDiNnZmuewrdWXOIsvc8yQ0d/tU+I7vjw+0VgJrv1L8lnEnS1GAgTUiNAecef08pjomIFMlVYVdnxSppRZPW/+WYxR4P2iuIqhJ+5bp7U1fQjNdBwq3ndug69V+2EyFX5tnUIIPinMcx3GcLYw/0B3HcRwnD2hFGnpuP9BLLIqnSz4Es4A4a1dWqSdBLA/FVAt7YuEoiMl0raQyh0mTKcVg/sjukAF212/GwgvDkLbNiZrVCOUympyUBrQ9bMlEs34MJ3PnamSbTg6nbRr8SphLN1i2NzZrWwGSqaqjY7EaGx8r55oFQcoeZIrYQT9P2qxjqzVbd0Uw9APcPfRvcZxWQE6Q6sOx9EJW3dXjxj5JKo8bk4QoA+jmzwldLyLW/LC0KQFQItBLMVPqpF7znIivBLbuG/eb160eihHwMZ1P7yXvwJf0PdlIBCRp8jX6GIJPWpbnTndpx0u/Nm5r/6222CR5ThPJ7Qe64ziO4yThIXfHcRzHyQNaUcjd09Ycx3EcJw/I7Tf0gm5mBf/96dU7EkozaToKtbiPuUCgv7K7hoVmp0gFpM6YpNty+zi9YyDO7c7ar8fCloSeD/8QupgHJKIl01IoB6v6Tk2PpXaHSqlSTjEa+6Wr09UeOyD0/Ri9tr/kEFKw5HgBSY9jSlN32CxdqfD8+Le006+HvoI/YuHz5ZvOg5DfWb8VYzH+DoVV0rmWIbXrenylbgLPOWb63SHt8s6h7wKUmH2wkb8zM1uEY9pHNPWlWJZy+wpp89pLSpXimwiHKOwvbVTkjZ06mpnKrDrq+qrHs6rtYKbopaLmexg7QA1dI7ol8HHYjMJ+oNat6+rEeWT5RWrzIqGtw4kq4OOgGrWlH2pWWzzttiVojrfrHHlDz+0HuuM4juMkUWdm6SauI0ce6B5ydxzHcZw8oEUe6CtWrLBTTz3VevToYR07drQhQ4bYyy9HiVDpdNqmT59uvXr1so4dO1plZaW98w4TOxzHcRynidQ3078coNlD7v/5z39s9OjRdsABB9gf/vAH23bbbe2dd96xbt26NSxz9dVX23XXXWd33HGH9e/f36ZNm2Zjx461N954w4qKkhQj0t0aFN3ySKOkypy0xhRsTq0ZislcenBoisDG/FWeD7pNyG63r3AbgpzrSXAy0fuZqElN9ynYIrKWNO4yszBfn9om9cAFor92fyT0UW/vqCLlgIdC59Bvhfa4pWJA5A3UbQuSmJlxwgq4Kg9y7EAKtv7s5L7szLz089dF7c6nwSnbX4ts3OdCc40UBZiHtfArdbwArwNqx+U6vgEa6dT5oa3HGJsXO/76lsDznsdC851ZL4JjS0qkzWuaueY6peto6r/QeAfrgBfsDI9x4n2dF01F1BwAH+1gh6jF94Gt288BN7wdDJF26c5wjk744+3gK4GtR4tHg48V9f8tatZ8ZvGCEi1AKwq5N/sD/Yc//KH17dvXbrvttobP+vePij2n02mbNWuWXXLJJXbkkesvuzvvvNPKysrsgQcesBNPPDG2zrVr19ratdFoipqamtgyjuM4jtOaafaQ+4MPPmjDhw+3448/3kpLS23PPfe0m2++ucG/ZMkSq66utsrKyobPunbtaiNHjrR58/j+sZ6ZM2da165dG/717cvf447jOI6zAeqa6V8O0Oxv6O+//77dcMMNVlVVZRdffLEtWLDAzjvvPCssLLTx48dbdfX6oGZZWRhXKisra/CRqVOnWlVVVYNdU1Pz34d6Z2sIuUveCgNADN3prxjOihZT8jUElw5LkVpBRWgfEDU7XRG6mDGiqXXD4dt5FD74ppZ33R/Of4TmEolh3xW6ahEr1fDsS1grpQcNuTP6xAnUviNthlEHwT5Swsn7nA/nM3fjgyulvQ6+31ljMOLK0sBMeVKYtqa9zXMnFnsOGAlbEraWwoVcKY0IrwhdMUlA7zu8Dhi5DaKoQ0JXKXL9jlsStbnfPHc0msxtYN/vleDDJgVpdyzZnJiamIKTaVVyUg5GzdPeyLPT1DquJhbu3kfazBlkqFwvjFL2BK95lfoq4GPSblEjbbP47b/dRvpINo8RPQuTCmI3I562tunU19fb8OHD7cor199899xzT3v99dftxhtvtPHjx2f46w3ToUMH69CBj0THcRzHyUC9NV1Db+rfbyaaPeTeq1cvGzQofA/bddddbdmy9XOHl5evHwK2cmVYemXlypUNPsdxHMdxsqPZH+ijR4+2xYvDIOzbb79t22+/PkDVv39/Ky8vt7lz5zb4a2pq7KWXXrJRoxhrdhzHcZwm4Glrm84FF1xg++67r1155ZX2ta99zebPn2833XST3XTTTWZmVlBQYFOmTLHLL7/cdtppp4a0td69e9tRRx2V5bd9blHaWvRpMXSwCghuqqmz3CS146VyICsoFu4DTeqgKMllFwiqjD3oNpwNXyAVm5nZZdKmDjY3NKPkAluDyq8sY6l68H3wcSyBlnulwsct0pEQXA9Lf6oe2+XZ0Df4JyiXqmlgTBmkutw3ytHrBrXmC5Th1O3PNFWlpuzFSsbGfh5rSVeeAaI0c+gIMvB0+6hfJ5UUZsoVx2oE21sJHzpixE1Rm0NS/w/2k9Jm6twRsGdodhTlX6ps2uHoiAEU0fVAcZBEBWwNKCJntBvOlW56b2FZY86n2m93MQ6Ck+evdgQLuPIKUzvTLXxTb/HNud7GtPlsJoZtAnXW9MlZciTk3uwP9BEjRtj9999vU6dOtcsuu8z69+9vs2bNslNOOaVhmQsvvNBWr15tZ511lqVSKRszZow99thjWeagO47jOI7zP1qklvsRRxxhRxzB3+IRBQUFdtlll9lll13W6DKO4ziO02T8Dd1xHMdx8oDmmA/dH+ibgfTiqKNVJ0PK777Iv1YpHLNRxlKCtbRmxZ/g3Ie50Gc1tIb2vjTwHIgcZU1LrfhO6LMvX4APdFpRaMUf3hvaUpuHeiuVOa23x7EDkA4D3fxS+MbCnintX8PHVG0pVBtTmQf/Eh+cr0eOed086NKGmNx2iTVKplGiKuMyxz6mAQc6KZ0fRk0O5IA4rznVGFWQlYa+D3euWNrUmRFgq5ABGF1eDn1nhaa9J21K20nbGytSQNFfc7cxTiZW+EPlWQr5PNF0UEBHjhDZA7YWDsZ0v7EV6xXXlBxwks1tW+9Rm/p3G/rbTV2v05L4kXAcx3HyFw+5O47jOE4e4A/0HOE5i6JaWh3xpHCxMQi5a5j3LawyBVuDvCc/A+clKLuqgXRMivbtS7Ho16R99b5wHgNbDxNy5zZc/t7MzEp7wkZo9zVJ52MpTUZntczlIQxhnhqaF0vq3MdYMdPjNCS7AL5Vr4d2afpXkVFQFTpZ2HQfiSdvG07m8wVC7hoGzjQ7mGZDxXIyYrVAVVhJCLlSG8GXqkyxPHTFStMqsaQg1kDWneHCB8D+c9TswRnJYGsgOiwfFT9dn5BLaCT0mWLu7BhpHwgfL6FiDZ3zwJTA1guFBXIrEpaliJV0O2UIm+JDcxUL5/dsKtyXbFLMkuQEPSOSzl5nU8jtB7rjOI7jJJG2nHnDbir+QHccx3HyluaYLC1HJlvzB7rjOI6Tv/gDPVe43aL0FdUHzw1F3uL7Q61mD0k/exSrZJrNUjUouAcJOmZmo6PmeXCxXuZpmlTEAjudYWuq2vWhi8Kzit+QA9MfhrZOBksZl5q6pkutgPTVh6VLp0fNK6eFLvavlgmlZPoU7BMeF+MQaoXssxOiZvnNgafWNh5eyNpPzJyKTYkZzKdJDV00VGQipt8N7dnS5lS1PE4KUxHfxhiKnXWMAmYGti/vHtrH/TVqI32zEzR0VaxZbpilgK+VNoa+2ECMfRmh1x9rCL8Gu7fsfQ/0xHahabvqRTMGTs6JqseROjgusEAv5kUCOy0XFQ8qZxlNSZspjzWwk+Y9Jlpql9vAfFM9X3hQuU16oeis2dlciM5GkdsPdMdxHMdJoDnmVsmRuVn8ge44juPkL60p5N7s06c6juM4jmN2/fXXW0VFhRUVFdnIkSNt/vz5G/V399xzjxUUFGQ9A2lOv6F/8NtIxSrVnyZ3oRjp9+4PzOGiAbLiJUtrBlJSrG4ldTGh88TQPu1vWEALZrIsKL9oTtT8IYSnG0LzPyLjcV9+C1s16hR8/EW6VNo/g+/bD4Z2D8kRbvvH0HcLpqc8U9ocokBpzrTk6CHsoxLYUgq24mbbVBhq0+EDxfDZXvxAxVpeap9GTeTbs8Kwlj/AzKqxqqbKUtgcqXGynCCHzYHzXPTv3ntH7ZHhTano1XBRHS7CQqqU6lVj5zn3Ddh95dwuXwQnB2AslTb1YFYNPkKO8h6Yw/dz2IsbafM7zcILMFOhh6Q5fDnARW87vMiTXiW5XtYlKJE2dPtlGNfxsLRZPpubpLuqxXFZXrql2FIh9zlz5lhVVZXdeOONNnLkSJs1a5aNHTvWFi9ebKWlpY3+3dKlS+3b3/627bfffll/p7+hO47jOHlLXTP9y5ZrrrnGJk6caBMmTLBBgwbZjTfeaNtss43deuutjW9rXZ2dcsop9v3vf9922IE/hzPjD3THcRzH2QhqamqCf2vXbjjOUFtbawsXLrTKysqGz9q0aWOVlZU2b17j5T0vu+wyKy0ttTPPPLPRZZLI6ZD7Q2bW8b/tQ++OPu931yPhgl8+PjB33iWaoawCYTNmZ+hMV7EqkIbaqkHcbDR8w2Brcg9TsBCefyQqeVp7Uei6CX+pYWtGIRE1CyJ3jMYxWUf5A2xmqfz8u2IgHFuAjbhVuuXHWFF3A0G8OVPAriJq7hp6MJlZAH+J8xev+tsyU46h3Ph8bIKkNCEOzZC7Jl0xVW4IbP1Nz0rFKBocnK2jEFnu9t7b4QcDLo/ah4Uh97Ibw0U1IXP/0BVL3tOsS4bjmbYYKBrIRFv1ZmirisGjMIwXhsaIea/9M2xJnVzxaehKWeOwSCzP7WJNGUMn1eG60M1nRhv7V7+X5057xMbbSyog0w3vhv2itJmKyvC0btOO0m6uIrWZqLemD2r73z717RvmH8+YMcMuvfTS2PIffvih1dXVWVlZWfB5WVmZvfVWLP/ZzMyee+45u+WWW2zRokWbvJ05/UB3HMdxnCSaU0Nfvny5FRdHo2c6dOiw4T/Ikk8++cROO+00u/nmm61nT74objz+QHccx3GcjaC4uDh4oDdGz549rW3btrZyZTg90cqVK628vDy2/HvvvWdLly61cePGNXxWX7/+Z0S7du1s8eLFNmDAgNjfEdfQHcdxnLxlSwyKKywstGHDhtncuXMbPquvr7e5c+faqFGjYssPHDjQXnvtNVu0aFHDv69+9at2wAEH2KJFi2Kh/sbI6Tf01y3SQ1VLqroGqV1VUPK+F2noY04PXdSD9lHjSG7BYNiqWLG+JFFliclI00NTdPNfYsknYau0yKQ6Tlaov+aGw3cPbN1aDtd4AfYFIm9fizQ1w9SllurX0Pz2ycsCVxrTaQbirEHAjCHHYkToSUr1ygpmGxYcig+013ipyWACaLos2arHjdrreNgnSB9NxYCQ32BZTTeibn88+/6Skqg9LtyKgl1CMXYPGZeyB1aDhFKbIe2X4eMYAB11MhJDKDj2JSXtw+CLDTzQC+Hx0FWLASN6vXFqWGrUmhLLdzIu+4Xsz2rsG/VsHQvB+xU19BJrHCZ+pqSdafyN3m+Z/cYJaLUfdIzHWovfv1qCLVVYpqqqysaPH2/Dhw+3vffe22bNmmWrV6+2CRMmmJnZ6aefbn369LGZM2daUVGR7bbbbsHfl5SUmJnFPk8ipx/ojuM4jpPElspDP+GEE+yDDz6w6dOnW3V1tQ0dOtQee+yxhoFyy5YtszZtmjdI7g90x3Ecx2kBJk+ebJMnT96g7+mnn07829tvvz3r7/MHuuM4jpO3tKZa7jn9QF9pkRYVqOZMzq7CXIunHdDQPGNimO36BfSrCWpM4RZw1CEVrCQ0C3NR6PpdqAiukqRa6l7bw9YqkdQVeVKqJP1t+LpR/BTRbw+InSz9qD165T9C38XICbe/iW4+OyyXW3AhSrYGYh17glmtciyGhZnnXRLmbWQ+Pu3gCLNucKz2QNJUmyI0Ix84hSX1uFGv5NgH/WAESvLCDPRXjoM4/iF8cIkqxhhFcdyPAnPn34tB8RjT/e4lOdbMQ8essnabtLm9A2FrznoFBx70hi11TN9GPj7HHejmU79mdVcdUcPjljSOg2Nd+D1qMw+dxKb43UhY5YFnr3bhqfBhWJL10ywsuSRq6s1+wptUC9CaZlvzUe6O4ziOkwfk9Bu64ziO4yTRnJXitnZy+oH+mUU7oAdsFcq5lqbDkKAVyHxh14Qh97Nn4kt0PEPnS+Fk7c92jbTN4iFhDZbNDV2o0amRXabgMKSmqT6MZjHQrOH6gxkKj6dKNjAIIXeW1lSB42H4SpD/cq6mfr2IEPvQoxvfiIz9q/5jA09BzzAnq+2H1ijs3yCCHKuPy8BvwuW1LiqtuhqxWoZu9WbCsrWx7FRJ0dsNMXaGfTVZEspIPP4dFBU+PHQxxqq6D3KuFqCMKScsUyhgaaiZE5Cx/PAYNXjih5Mv2qOyEZxJEGJdIIcw3M2bvoY/M0k5STTlYaSh8kyTranN/uU26LoYcu/H60IPhsbq15rZ1dbitCYN3UPujuM4jpMH5PQbuuM4juMk0ZoGxfkD3XEcx8lbWlPIPacf6Gst6mjVHV/Hcgfehg/OkPa5/UJf77D8qB2los+Y0JfYfZk03pS0UeQSImqB6NkDmT8CofwlyS6i/kv9NdCDWcaUaWuiM7IXOGuoap3UIJkGVCQVXM9g1td8iJ07XypGiSWj/Y+Co4NCDb2LpCqxzzjtZSAP7svvLOMHAo6/dEwKSybdPKiRFTL/SapEVsCVlCpVw21AZd22QaFT5KLtjGtoRXQNvfhc6LoW37NI2iXw8RRMSZtTwXJCSi2R/Dn25RMI93OkzaEDTCFT7ZjnRjb6ZdIbH89BZt1p75fAR+1bU/+ooe8Fu1LaLD88B7beds6F7zj0b4nYejfNlHLnZE9OP9Adx3EcJwl/Q3ccx3GcPMA1dMdxHMfJA/wNPUeotQ3/cuLUf7E5+s54TwxMQHkU5k+0Y6RN1Yy6eDbI5KbvIVGWQpgCUW8Z5nDUXeWkrEn6cBpaZ8EULCzJ8N1uCF37oE6k6m/UOt+DrbNTliOZ+LCLsfBvtcjsQXDyVFYbJXqRzL+DaOjUmWkHGjrnAo0pmroNOFeWRs0U/irp5hHLX2ay9k5RsxCSfhHn+xSY+86hGp2CT3gdYBrhlyMNndorpwLV/TkOvktY5kHOs8OwwTzP5kmbue7sQ14nyg4JNqvaUuvWQ8Njyrx53QZuH2d7PULaFbioV6BftP95+FnlYR+dpRMi+he479wlbU57y/7WftHrh+ec03Ry+oHuOI7jOEmkrekh83RzbMhmwB/ojuM4Tt7iIfccRH+BxTqf0xWZ5rEMg4/5WxWbvE0hDGL+LWoia43xwzrJw2PI+s6EP2XknpODaSiMW9eRdWPPkfaBoWu3P4S2uil/MI3tNWkzbFr5u9Au/JeksfWqwtJJIfeeoeuC0Nz/oqj9KNbC4PJOagxg4k/S5YSQ+ztRM9OsXYkwV0rzglCCswQxVw3t8g2G1VLDfmCcH8FnOXd4vPmX+0n7EtYQPjI0V/NkFygnJYW72d/qZ/Hei2AfpRcNY+yMlacaaZvZfyBTaeldbh+7JQizDwp9fXDBDcFMfkrsPNNO2z90HY1rXE7f2HWbgq3nUom0myJYOhsmbx7ojuM4jkP8Dd1xHMdx8oDWlLbmk7M4juM4Th6Q02/ohbbh6VNLuGBsjsk+0mbyCfTWQD1sSndBNLPfR01kyv0btWu1rCXL2iLbLEhNoaTXBzbTYQKovbXbPWqP/WvgqoC+piU79wtdsdWqXrgAPmpz+2m63GVMNuJxVKDcFs4K17v9lIb2EIi+1BmLd1Trq/AmlftFoUv5HqYTcTyDEvsFzlcHnQMTpWn7PhvaSTozz9bkcx/XTIeoWY4OZBrYlWp8G87XQvMOaVOb5/iQILULPg5ZUZsafywzUU9u1jzmBaVaN9LAur3UuL0a41di54P0b2wOVGjqY+QGwfsBxx0E4jfOncHQ5ifIjYhXHsvGqobeF58/zW1oATzk7jiO4zh5QGt6oHvI3XEcx3HyAH9DdxzHcfKW1jQoLqcf6O0t2gHVg7bnggfyA802ZZYnsyPbNdLeEOqn8vWP0Fy4KmqjHmZ1aAYnE5Vj2qpXMaO+ErZKgB0pqCWVn/1KaHbaLrS3lV09FH/6DmyVEqm9xTT0h8W47Hl4UX40EczT+oOoefTpoYslOsN6mUPhTDo/MIenaJIrQk/szNFwX6z0K+egVM0aedx7XBXaOnSD642HGJP2DWfaPlHzzD+FrmG83FREL4FvZmjKqJNYrjOnWtVSA6VfCn2HYizBRGlTm0eX2XFPRe0hr9EJW6tGT4KPttRL7YT7QSduVNJFjuu4o3TMgay7m4Kt+vuu8GG8wJdlwMuXeaHyQtZzUk6smrTZzZy3twWot6aHzP2B7jiO4zhbmNb0hu4auuM4juPkATn9ht7RoghToXw+iAtyWqEggYuJKkkFCbPpLq7nb6GpYapU6GIaiEbVGDVL+uXIFKFLGFbX0qD0lXBtsu+79gtdI5cFZjnUBQXZMPaGtBnlZ3jedLI1uwVOxMqNU3Ul+E6LTpBDbrg/9HEjvqEGpjOLoefAh6FLctW439m8DaxBflnHpWLsEx6nIyw8Tr+WNkOSvCrC9E2e2+hPORTDqH8dAVvzmK4NXXMww5eeVtxenr6lWgIXN4R+yP3cSXIpOVPYfbA1QnwwDumYG0O7WC/WE7CiU2EPE11gGGr2xo7G5420zWLSTiDKvQsfxT39nt3gwzadrHeppHRNs/DckSTBmnVmXf9sLU1rGuWe0w90x3Ecx0miNT3QPeTuOI7jOHmAv6E7juM4eUtrGhSX0w/0nhZp51r6seMBWLAj5ssMBOKkaTfNspvkL6HUpyHNSss7IvWkB3KI6hL0Vk5zqcR00B1hl0ibWmdv2IFOiiShsbcFZg9J7fkCtV454ajOQEnpnQrfe3JVDXgCKz6Y2qFub6ZjOD5qzoKGPg+L7qwnV6ZzJUFDFzE2loqWAEN/LKXbR2sB7xoWLi3+5s2BrZovp7ntwalBE0vrQkM/Sdo84BS7fxM103eHLlREtqQMp7jmL7AkKwaiDJkftZmBxX75o7R5k+eu7SXr7cGT+w3Yx0ku3SEvwnksbL3+kDMa64lsbvG6LNfDs1SPeabvVHuMtFebmWvozYmH3B3HcRwnD8jpN3THcRzHSaI1vaH7A91xHMfJW9LWdA083RwbshnI6Qf6QIvUm6AyISXz2CSISYpbUunXTMsqKSw6P7R13lCKb/g5qIoa89CTcpjfgu/KN0N7hLQPXoSFYxq6atTDQhc0dBOzKzRoTnNZIm1WuMQskkGO8AAKrAezCC6nwU1CBjHsPS507fUQlt0/YT1JGjq2Tw4U5eqks5PTucY0dC21OhHjCq4JzfNEQ3/KAGv2Mg85ABp6Oxln0AVrxhAFkylxMQuvLYWtaensoxJukoqJTKlGKej95dL8S+iK9a+O63gSPh5H3aYSnMxtX8bCemD/gJEx7X8d2mVic4zCQVRRZXxIbL5X6u+aL55JFy9K8CXZOvdrNuOTnI0hpx/ojuM4jpOEh9wdx3EcJw/wtLUcYZyZdflvu2JvdTBvrY9tPNmkrXFZLcOIpCuG2HRGIuTGvIYpvjSbh9kuPNG0kiaDbwjcaRFGewPR2fNnYeELJNjf6+LQ1+/g0B7yREOzCCF3BvKoNigMd+rkVocxvyhW1jIpRJwkqxwOF9O1KhLWk2SnQpccnIrQE5v/T88kFvr8ALZp2toaHHGkb3acFdVaPeyXWM942MFW8rxnmV05ygvguiM0/yahaEpEKdj6lsTzJpZUpyFsZlxBgRv4s6h9BK4DCjkvSJuSEFWgCmlTaqpI4QM9uRH3fxEXuV4XLHNdcQAW/n+ifx3+CJbGdHw2QNol8NHWY94JPl7lauvNDbV9W4jW9IbuaWuO4ziOkwfk9Bu64ziO4yThb+hNoK6uzqZNm2b9+/e3jh072oABA+wHP/iBpdPRwP90Om3Tp0+3Xr16WceOHa2ystLeeSc2t5bjOI7jNIn6ZvqXCzT7G/oPf/hDu+GGG+yOO+6wwYMH28svv2wTJkywrl272nnnnWdmZldffbVdd911dscdd1j//v1t2rRpNnbsWHvjjTesqCixiGPA9l81K/6fmDZZPWdjyabspv5tNmkWC0OTFQ5FMKyGZs4pGzVtjboys8v+n7QL4fsZbNXjMQun9UZ+1PFSotPOZ39Ci9sz0tALOoSuDvgibqNCvTjQLFmTM5ZwpEJppuOvfqby8G9LMqyrMXDuyI7zGCZ9A9MUl8JeJZ1UegOcVdiG8+Wbd4AifHBpFltFJAHxudCz5tXQ/qiRtllcYdWbKjV0HrXg5OGKmcUq9qDfha6TQjOQ5jmMg6M4ZkubGnoRcjRrxP5T6IqlquqQABaYruCgCt3gD1eFvk/CUsDBNcUbQhfY20ibTzvWo9Zbeol+vznNTLM/0F944QU78sgj7fDD1w8uqqiosF//+tc2f/76ZM90Om2zZs2ySy65xI48cv2D4M4777SysjJ74IEH7MQTT4ytc+3atbZ2bXSG1dQkVXV2HMdxnPV4yL0J7LvvvjZ37lx7++23zczsL3/5iz333HN26KHrK1UsWbLEqqurrbKysuFvunbtaiNHjrR58zgTxnpmzpxpXbt2bfjXt2/fDS7nOI7jOEq9RQ/1Tf3XakPuF110kdXU1NjAgQOtbdu2VldXZ1dccYWdcsopZmZWXb0+CaesrCz4u7KysgYfmTp1qlVVVTXYNTU1/lB3HMdxHKHZH+i/+c1v7O6777bZs2fb4MGDbdGiRTZlyhTr3bu3jR8fS27dKDp06GAdOnSIO86zKAVyH80951ygzbWbmXLUNYkVc0E+GJrLZDrFh7GWZ2CrusmQyr6wT9tVDGjzL0Am1VT497AeSJ92/BNinM+pSjEv6z7SLgldhStDOylExDBXoKlTPLS/wdZjk01pSmaBc9kOCb4ksKxokJQnaatmStlxMWydKbSUZVarmIcs03KOmwvfQbA3cP01gOsgLaMzXg9dlLNVXk3BxzEUCjOfWWmiTs79thxvO2730D7irw3NCtSfrYWQf3TC9nEUhw4XuBy+EbB1TADz24lOnnocnd+ArScEA6B3wX7BGoeDAOREXIPBA6yVocdcz+3YJdxCeGGZJvCd73zHLrroogYtfMiQIfb3v//dZs6caePHj7fy8vUlIFauXGm9evVq+LuVK1fa0KFDm3tzHMdxnFaMa+hN4LPPPrM2bcLVtm3b1urr1//G6d+/v5WXl9vcudEbQU1Njb300ks2atSo5t4cx3Ecx2kVNPsb+rhx4+yKK66wfv362eDBg+3VV1+1a665xs444wwzMysoKLApU6bY5ZdfbjvttFND2lrv3r3tqKOOyu7LRgw2K/5fQPIUcbAUJclmt5NS1eiT3JN7w9yN/yCGrZG932MtDKNqaI+hxT1gB7VfMcxgD8TyNNSfwmqWcr0an48tPSA0h8oPum3DYFUbhNxZlTOJIJMuFo9l0lA2KYZ6PmRKm2zXSDsTOCcl9shvZDhZfx5ztrUVsIOQPPOd3ns/tAfoOJaTsXCsmGoCOBia/oRhMdx+DTWn4OOy2g+c2Ywim4Zzi3lBscePkfYtoasHrtsdpH1E8lqD7ucmMKyu662AbwzsM7QjpsO5J2y9uSCNcSEuGc14G4jVVOC6fVEODktKvwZb7wCaXsjj21J4yL0J/PSnP7Vp06bZueeea6tWrbLevXvb2WefbdOnR2fehRdeaKtXr7azzjrLUqmUjRkzxh577LGsctAdx3EcJxOtKeTe7A/0Ll262KxZs2zWrFmNLlNQUGCXXXaZXXbZZc399Y7jOI7TQGt6oPvkLI7jOI6TB+T45CyHW5ROo9NlZkpTyoak0q8UcqUY5KOhB5UeA61zZYLPLNQZOSno0bADAQ4a+r5/DG0tOcrvjE1sGCzA/Wb/iurX/VmuqVGy0dNjAlwtEqIKs9HQFe4LZaBNPZegoZc0/g1JU8pyt1mhU3+hr8FB7LgUCw/Qb66AM5v9xvmgAnEqdPFNR23uC/dVt2gH+HaCXaidyDlQKewXy/S/ez0RuLpAQy+WNqcuZQlf3abfwsdN0nvAYfBNgG2nSrsMPgraMibgV+hQlq7V64/3pBfwtzpVLGeGTsHWVDVNCnUNvfnJ8Qe64ziO4zTO/yrFNXUduYCH3B3HcRwnD/A3dMdxHCdvaU2D4nL8gT7QohqaPeXzpuie2eSdswSqJH1ClKb2rVvImVU5M6hqUHvBV9oTH2jM5YDQNbB/aI9cErU562IsdBPEnDKNUZC89E6hhs7QVTa6eXBRcUWcgI/9EtCEkq2bvCxKykqyeXOGyZjDHhDL3Vddvwd8mcYSJKz4g8ZdSTfGTDdN3QLWYyjkwANNVI/N0cs5I06ImpWhhl44J1xyBxGXh/DAoYvGyRgGauYs96x+Xv8FLKuhNWaR9L0I09Nq6jlL03Kb9JJi+Wmif8vxNpxlQ/PotW7GGgvrcbQUrUlD95C74ziO4+QBOf6G7jiO4ziN4yH3nKG3RUFGjUs1JcSeaUY1Bckdb0q+DmI0HTEh2UCJ+vVF5J7BTQ1Ls+RlLGat4c4K+G4OzenRlPSxWbBiqVNBegxjgOwziXdnKP7XbBdKs8XEsjn+2ZxnJaG5zQYX2iB6iLmbDLGpPNORBxGbEH7QjDJVwjRpxbB137KpE8nuSyMHqkBXxn74NCzLbJ1FbjgUy1aGZidNyWO9WcSe24oCty1y8iiNaAj7Sfh+gfvDARI7Z8oYFIJgsjumpiZde5mW1e5m9+4G+2cqf0muX806s/OTZnhrJjzk7jiO4zhOTpHjb+iO4ziO0zgecnccx3GcPMAf6DlDV4v03GwUONX8stFM6UPxRC27GGa/2N+gg6n29XroipVEVEmSaR6DIOPv/7uoXTASC38nLFZZfP8bDe2rUEOWKS62nxol9AI5FhDYmnJhBMMFmlUsSirv21xlgzHuQAZDcBhEViVwQdDdHHARmxE1adxJ0n6zj2DrsSkJXWUfhrZqtdRieYj13OGYD6ZdlqoonYLzY9id5XxthwvhuPtDW2egZb7WHTBFN6fWTTRpkPt2Ley7pM17BRPytM94OtDeVtrM9GOKW1I3cCLjh+SYj5AsVur0LUXamq6Bp5tjQzYDrqE7juM4Th6Q42/ojuM4jtM4HnJ3HMdxnDzAH+g5Q1eLMm9VQ89m6sxsloUA+CHUrduiJnNHmW75hrT/CV/SyfMK7IthHyftqovgHPJGaB8V6YWlD4da4cHX42+PVYN56BtP0vSZ9FEPCmwKrtkMoUgkmzEV2fwt+kwEy0za8SZTApt500GnNUVDx7KabA7dvgCibxeZV5jHn7qn/im1YpZLbS/XXzdeYPzjPtoP+4e+sdDQVSe/LXT9AuKx5oRz+3go9pU2NXQMxzFcxQGcwnXfRtpmYRlWs3BK2i7wcTrV2dJmGVuOZ9DyszrqiNPlOk0nxx/ojuM4jtM4ramwjD/QHcdxnLzFQ+45Q5FFYcOW2hVNGns+dF0dmgv+EbUfx1qYBqYpG5iHKxYa6yBtTKYUpI+Ymf1G2p3ws/Lsb2LhZyWcePi40DfiodAu1YBcprC02E24EhLTuZhXk1hKNdO50VzlXZP+FsU+ZUoq7kpSCD6pBGfMzxOr3SB8oGdWU2aVg96xTSNts5ieoOFlZpMlSTAp+JgqpQxbig8Ygh+m+1MR+jruHtqv/rWh+RjyTX+L1eq1uS18x8E+V0srY8fPhtKnIgD7iGH0X+h6h8CJctSBPILyveXvhHaV5M8ORN4aZ4/kcXVaDk9bcxzHcfKW+mb6tylcf/31VlFRYUVFRTZy5EibP39+o8vefPPNtt9++1m3bt2sW7duVllZmbj8hvAHuuM4jpO31DXTv2yZM2eOVVVV2YwZM+yVV16xPfbYw8aOHWurVq3a4PJPP/20nXTSSfbUU0/ZvHnzrG/fvvaVr3zFVqxYsdHf6Q90x3Ecx9kIampqgn9r1zY+Vv+aa66xiRMn2oQJE2zQoEF244032jbbbGO33nrrBpe/++677dxzz7WhQ4fawIED7Ze//KXV19fb3LlzN3r7clxD72RROlA2aTZJcFkRsGouCV1IW7lP2ouxFpZOLJH2WPh+Arvt3lH7DkRgIOMH6THU8Xf4e2gfPFEXRhnb0omhHa8bKiT0b4Y6ptmEsoJVUWhux4k5t7ZTGzpzRdRsi5/VhVl0CjX04E/ZJTY4eZsCsuk/rEc3CvlOtUtC+1Fp8z2Eb0VaqrRPhmVVm6+FBl3IvKqAHrB3Cc0FkYbO64vSvI6awCysdi7zy1RUfzR0FWH79dTHRLD2D9jV0v/lvGY4vkEFeA7k4fbKeg/DfK+HPYVl9cDKgampN/sxBwG1APXW9EFt/7uu+vbtG3w+Y8YMu/TSS2PL19bW2sKFC23q1KkNn7Vp08YqKytt3rx5G/Wdn332mX3xxRfWvXts0uxG2drueo7jOI7TbDRn2try5cutuDj6pdyhQ4cNLv/hhx9aXV2dlZWVBZ+XlZXZW2+9tVHf+d3vftd69+5tlZX8Odg4/kB3HMdx8pY6a7q2/L83/OLi4uCB3lJcddVVds8999jTTz9tRUUbXzXLH+iO4ziO04z07NnT2rZtaytXhprTypUrrbw8Sb40+/GPf2xXXXWV/elPf7Ldd989cVmS4w90zUPPhiRNHQmYmtk9LfQ8Cm1LlRFWl6SUrFndR3LZQ/GBrIwzolLaUnmQJSKfgb3HH6N26T0YeXkiTzoV2DYcZtqgvwklWbMq/Zqo8WciafrU5gKXmtb+RB+1R15v0jCExLx01u9kjnWwTc2Vb29hkQXUPEU6c1AKlCVPud+qXjKPuwy2diHHGRQmztvJExbJ2qmoSSmeV8Ve0v4+v+ZC2LLepUiq53Wsx5xbuxT2adLeDQL7PrArH4zaPYZjRWfCPkPag1HD4ltMcNeyxy9GzZp1Zl0puDc/zfmGvrEUFhbasGHDbO7cuXbUUUeZmTUMcJs8eXKjf3f11VfbFVdcYY8//rgNH86DkJkcf6A7juM4TuNsqdKvVVVVNn78eBs+fLjtvffeNmvWLFu9erVNmDDBzMxOP/1069Onj82cOdPMzH74wx/a9OnTbfbs2VZRUWHV1evf5Dp37mydO2/c/Bn+QHccx3GcZuaEE06wDz74wKZPn27V1dU2dOhQe+yxxxoGyi1btszatIliBzfccIPV1tbacceF8afGRtJvCH+gx0LsSDh7PppRre660HVfaAbhLob5WPpRA9gHs0TnTrAlrr4Nar9y0dekzRDma7A1lFfKKZNOTOEDDWpm+rUofqTGsMzpJpMx5L6pp3ZTysRmsd52O0ftorcDV1NC7omlX2PJXi3UR6ulnQpdvC6+SPBxv/Uawpxo1han5GqZba1TJoUoOKbcN8yLJl3YFyl4JfjLC6RdyPzSPWGLnPczuJbC1nQ4lnrlZaHXOOWO52CrJHfoy6FvHOP+6j8CZaJHYNk+Kgzqdbp5KqRviZD7/5g8eXKjIfann346sJcuXbqJ3xLhD3THcRwnb2lNs615pTjHcRzHyQP8Dd1xHMfJW5qzUtzWTo4/0D+1KMigIlrCdJ4xG7kbnx4R2lIBldrWK7BT0mbG0G6wD1YD1SXtS7Bl7lUqxUxbUw2N2TmcxjCQaimwxYpIaoJMptOmZ9REJc2kqUEzkTh9aixxqaVO7U0tMcy/kzKs24QaehEHP2RB0Ecl9PaEvalpa1mMI0Be1bYYH6Dnc6ZzIbGKcEVoBpPV8g9j6Xw6job7hj6Tol0nQ4TeDn/Z7QoxxsN5V2i+92zURhHm2HWs42auhG8gLrDLZWDCr7EsKkEH38OqsDvguA3WfUc+7Bqk3X0iRXFLpM2ytS1FnZkVNMM6cgEPuTuO4zhOHpDjb+iO4ziO0zitaVCcP9Adx3GcvKU1hdxz/IG+1iL9K6l856ewZbab2gNC10Gh+as3ozZTtakyq1TH/HCWqgy+dS84acsUg4XIqS3CdLwqWXKmXkqHQW58bFrflfxAyEJDh8gfSx/fVGJiEfXhLQH7JSm/uSJqdgo97CO1M8xGG/5trBpuCezmuvyxHj3RwtkmrQ/GapTJqw/LmDIvXf2ccja2a4OkzQ5Fanl4nHjvwFadHjWHcOrPMbAnykb+Fe94ELTnSDtxdlcLD+tAit2YmOsIKee6FIuyFHRK2tTxOcXztlKu40/w0dapo7USgmvozY9r6I7jOI6TB+T4G7rjOI7jNI5r6DnD3y1KV9OkEUyDZrND8/mZUftroevKf4a2TEhmnJaeRWO10ibnyTmN5ScHSpt1LPsjJ6tt48Gp1Y164uEXlp8NsuViqTwp2NmkOEkMeYdGPWYWhpAzhZODsFcsbW1rCLmTJBlIgo/oe/ZDpn5RggAxcxpjvZ9ENvIBwtL6Nd2xKHZmpdwpedOkrWlVNXAWc3pDrYk6CL6BsBOPE/a1vySc/uCJ0NcP82TX1kTtm0LXeyitqhWdEyeDM5wuDLlDrhsqKa9jkafGS15ni6Sa8F3YqqRQImDWpX6PnhqUVFoKD7k7juM4jpNT5PgbuuM4juM0TtqaHjJPN8eGbAb8ge44juPkLc0RLs+VkHtuP9B/dYBZx/+2tbYq9CnWbH1RUtFYDvFJ2Jq8xYNKvUI1oRR8tUgLK1RtmRq67Rian0VzF9ZgPSzfqBUa+auU298+yRlL9VMynTZShhf5e0xNSiJRU42JRZwrdGs7tbk9ovln0NCz0cUCSXUHepN6P1N/ZaEzqwgMMfY+CKcvNPJnZvExKoukPRO+sSg3+mUdCzMBC58KO3E6YPbL4VGzH/sTyV6/iDT02utD1534y6XS5rgYDhcJvpVCNCsgS77s8deGroG4wFR+52yv1NTVZi9wiIJm0um4nc/M7LfmNCdb213PcRzHcZoNf0N3HMdxnDyg3po+yj1X0tZ8lLvjOI7j5AE5/Yb+4ORIM9Q01OVYbjHspGWp4+nsn5wClRrfG9J+Hb7fwz5eU8t7UeyEEPZetGaWZKS2pdtEee0N2FqicRw7KVbYNhtEVRsQKoBt2yOnPiEZlWGu5FKRSTro1ohsL3KJsymJyTeHQM+s4NK83Df18uffQfWXi2oBKghzzIpWguXxZoq1Tv/7G/j+AvsdGUwy8TY4vwF7qJ47mfpIx7f8O3Qteii0r4qal2Etj8PW+1Cm/GzN+16NOZE7YZpTO1rayNUf8kJod10StTG7a+w+qdvI6gYjYV+i9RDkJlqzzuzrHLTUAnjI3XEcx3HyAH+gO47jOE4e0Jo09Jx+oN9i0Q4slc8ZdWJoXH9t8UCxHKJOoDQLPqaTHCNthsKfg318UGJ2MLxIC5K6kJzJaIU1DtNJUNU2mOFpF9Rr3Plf6Jle2ZR+Vf/Q0NV+fmDWSewu06/g5FAk9zab7d1UmL6VBLdBwryIWTalJGaQitiHZ2hT+iGLv5U4OoqjxmQfvTZZofVA2IukTemJIXdlJ9wQvvwiFhiq506mxEoV4ZAqeR9MueCSQuxm4THnFvC60BLUnAHyhNfwwQXS5k0J6b16j6L0mFQ0mOcrj80D0g87STspMdbZNHL6ge44juM4STTH27W/oTuO4zjOFqY1PdA9bc1xHMdx8oCcfkP/2KKEGU3loKZDTapHgo+VE4+UdrdKOP8cmiVSlpW/6DjFYFgbthxOTP8qmiSncOW+agLcLvBRZ9RqmUyz25l6Wy/LAj2tOD4g1NCz0Ys5FiIkm6KyzUXSFKOZkPl0senJ+xmSPO6gAnY2l3sTlpWTnadRDewSaR8H3/lfwgdyko7FmI9XQzPQqF+B78s82QN4HvGYqh/7PS80NSMrdv0DvVaHwsdrXveNWV8nYBusYOeoPeTtwFWLVFVdF7LhYvcSTU3jV74H+wppD9Hvt81DnTV9cpVceUPP6Qe64ziO4yTRmh7oHnJ3HMdxnDzA39Adx3GcvKU1DYrL6Qd6B4t2ICmTtAK2SuHdM3xHIJvvCSeSakskx5LbENOKgzNkezhxWERUzZQ3f5i0v98z9F0OaV41M1SF3IDot6l56NuFLsSENG+afcbwUVBgtD2csdKvW+LUTvrOBC22e6OemM3d5lSroT0gYU1mm66TZ5g+tX6DTTOL124okfZe8MWmf5VBIBx1wutAoR4cP7cTdPHEcRLoz1Roai43jxvvOzqj67kQrF+D1q1lZDnVarwwhdwEBoUaegpL6rAElt3l8KGL5Zb1EOZwZplbvbdoPns2I06agofcHcdxHMfJKbJ+oD/77LM2btw46927txUUFNgDDzwQ+NPptE2fPt169eplHTt2tMrKSnvnnXeCZT766CM75ZRTrLi42EpKSuzMM8+0Tz/1ukGO4zhO81Jv69/Sm/IvV97Qs45Lrl692vbYYw8744wz7Jhjjon5r776arvuuuvsjjvusP79+9u0adNs7Nix9sYbb1hR0foQ1SmnnGL/+te/7IknnrAvvvjCJkyYYGeddZbNnj07q23Z06LkHw01Mrw1BvYM+RmzBkeKqV3FSTFhxM12kJA7Q9goEoncJBZLREBx26jZG7NXcV+DaN3A0DcS9Wc1/SWW9MUanUdp3I8SAYNnnzfuKwnNPvI7jqFR/trsq0ZMw3getiYnbg2lX7msHMiK0AMziDyzBCePRGHQaTyvmFAUOysTSOpDnJRyXTANlFH0vo20zSxRMdgFJ0sKi2o/sY/sI36g504H+JgYqMcRVznuBzrJGC5FvaTNzGxEgrMrQu4631vsQbOWH0h918JwA0vQEbq9vB9w+7VO704IuQ8JzeBY6Hc0pcRxNjRHLfemhuw3F1nf6Q499FA79NBDN+hLp9M2a9Ysu+SSS+zII9dncN95551WVlZmDzzwgJ144on25ptv2mOPPWYLFiyw4cOHm5nZT3/6UzvssMPsxz/+sfXu3Tu23rVr19ratdGZWlPDbFbHcRzHiVNnreeB3qwa+pIlS6y6utoqK6MhFF27drWRI0favHnryw/MmzfPSkpKGh7mZmaVlZXWpk0be+klvlGsZ+bMmda1a9eGf337xn7LO47jOE6rplljkdXV60NQZWVhoK2srKzBV11dbaWlpeFGtGtn3bt3b1iGTJ061aqqqhrsjz/+2Pr16xdUGtJAGH+NsfJWjfzcWgMfR43qsrFwFqJx+j0MJ7EqUo3GymoyLF3fqCf2PTqxVA2ivNw3XRf7oSbWabpmjndgOHltI22LxQgbO4Zm8V+buqYaRkJreHB0bzkWvLnIZpZk7p305+eNeswsPMZcS+y8SjdmmMXPgGwu/6Q+xBbLRvGoJG0/Z/iKnYNy7tDF60Dt2LnNjQjOHfZRUsj988ZdFu57pvuBXlHcPvaLrpch99ghDz4IndwGtZPuK2bhLYt3A65Xd+eLDbTT6ZZ9/21Nb+iWbgJmlr7//vsb7Oeffz5tZul//vOfwXLHH398+mtf+1o6nU6nr7jiivTOO+8cW9e2226b/vnPf75R37t8+fK0re9j/+f//J//8385/G/58uWb/hBKYM2aNeny8vJm287y8vL0mjVrWmRbm4tmfUMvL18/mGvlypXWq1dU/HvlypU2dOjQhmVWrVoV/N26devso48+avj7TPTu3duWL19u6XTa+vXrZ8uXL7fi4uLm2Yk8o6amxvr27et9lID3UWa8jzYO76fM/K+Pli1bZgUFBRscN9UcFBUV2ZIlS6y2tnmqxhcWFjYM7N5aadYHev/+/a28vNzmzp3b8ACvqamxl156yc455xwzMxs1apSlUilbuHChDRs2zMzMnnzySauvr7eRI0c2tuqANm3a2HbbbdcwOK64uNgvngx4H2XG+ygz3kcbh/dTZrp27drifVRUVLTVP4Sbk6wf6J9++qm9+240T9eSJUts0aJF1r17d+vXr59NmTLFLr/8cttpp50a0tZ69+5tRx11lJmZ7brrrnbIIYfYxIkT7cYbb7QvvvjCJk+ebCeeeGKL/VJzHMdxnHwn6wf6yy+/bAcccECD/b/BauPHj7fbb7/dLrzwQlu9erWdddZZlkqlbMyYMfbYY48Fv5Luvvtumzx5sh100EHWpk0bO/bYY+26665rht1xHMdxnNZJ1g/0L3/5y4mjEgsKCuyyyy6zyy5jRd+I7t27Z11EZkN06NDBZsyYYR06sBiE8z+8jzLjfZQZ76ONw/spM95HLUdBOunp7DiO4zhOTuCTsziO4zhOHuAPdMdxHMfJA/yB7jiO4zh5gD/QHcdxHCcP8Ae64ziO4+QBOftAv/76662iosKKiops5MiRNn/+/C29SVuMmTNn2ogRI6xLly5WWlpqRx11lC1eHE6i/Pnnn9ukSZOsR48e1rlzZzv22GNt5cqVjawx/7nqqqusoKDApkyZ0vCZ99F6VqxYYaeeeqr16NHDOnbsaEOGDLGXX47m1U6n0zZ9+nTr1auXdezY0SorK+2dd97Zglu8eamrq7Np06ZZ//79rWPHjjZgwAD7wQ9+EKTztrY+evbZZ23cuHHWu3dvKygosAceeCDwb0x/fPTRR3bKKadYcXGxlZSU2JlnnmmffsqpX5xEtlwZ+U3nnnvuSRcWFqZvvfXW9N/+9rf0xIkT0yUlJemVK1du6U3bIowdOzZ92223pV9//fX0okWL0ocddli6X79+6U8//bRhmW9+85vpvn37pufOnZt++eWX0/vss09633333YJbveWYP39+uqKiIr377runzz///IbPvY/S6Y8++ii9/fbbp7/+9a+nX3rppfT777+ffvzxx9PvvvtuwzJXXXVVumvXrukHHngg/Ze//CX91a9+Nd2/f/+tfuKK5uKKK65I9+jRI/3www+nlyxZkr733nvTnTt3Tv/kJz9pWKa19dGjjz6a/t73vpe+77770mbhpF3p9Mb1xyGHHJLeY4890i+++GL6z3/+c3rHHXdMn3TSSZt5T3KbnHyg77333ulJkyY12HV1denevXunZ86cuQW3auth1apVaTNLP/PMM+l0Op1OpVLp9u3bp++9996GZd588820maXnzZu3pTZzi/DJJ5+kd9ppp/QTTzyR3n///Rse6N5H6/nud7+bHjNmTKP++vr6dHl5efpHP/pRw2epVCrdoUOH9K9//evNsYlbnMMPPzx9xhlnBJ8dc8wx6VNOOSWdTnsf8YG+Mf3xxhtvpM0svWDBgoZl/vCHP6QLCgrSK1as2GzbnuvkXMi9trbWFi5caJWVlQ2ftWnTxiorK23evHlbcMu2Hj7++GMzW1+Rz8xs4cKF9sUXXwR9NnDgQOvXr1+r67NJkybZ4YcfHvSFmffR/3jwwQdt+PDhdvzxx1tpaantueeedvPNNzf4lyxZYtXV1UE/de3a1UaOHNlq+mnfffe1uXPn2ttvv21mZn/5y1/sueees0MPPdTMvI/IxvTHvHnzrKSkxIYPH96wTGVlpbVp08Zeeumlzb7NuUqzzra2Ofjwww+trq7OysrKgs/Lysrsrbfe2kJbtfVQX19vU6ZMsdGjR9tuu+1mZmbV1dVWWFhoJSUlwbJlZWVWXV29BbZyy3DPPffYK6+8YgsWLIj5vI/W8/7779sNN9xgVVVVdvHFF9uCBQvsvPPOs8LCQhs/fnxDX2zo+mst/XTRRRdZTU2NDRw40Nq2bWt1dXV2xRVX2CmnnGJm5n0ENqY/qqurrbS0NPC3a9fOunfv3ir7bFPJuQe6k8ykSZPs9ddft+eee25Lb8pWxfLly+3888+3J554olVNp5gt9fX1Nnz4cLvyyivNzGzPPfe0119/3W688UYbP378Ft66rYPf/OY3dvfdd9vs2bNt8ODBtmjRIpsyZYr17t3b+8jZouRcyL1nz57Wtm3b2OjjlStXWnl5+Rbaqq2DyZMn28MPP2xPPfWUbbfddg2fl5eXW21traVSqWD51tRnCxcutFWrVtlee+1l7dq1s3bt2tkzzzxj1113nbVr187KyspafR+ZmfXq1csGDRoUfLbrrrvasmXLzMwa+qI1X3/f+c537KKLLrITTzzRhgwZYqeddppdcMEFNnPmTDPzPiIb0x/l5eW2atWqwL9u3Tr76KOPWmWfbSo590AvLCy0YcOG2dy5cxs+q6+vt7lz59qoUaO24JZtOdLptE2ePNnuv/9+e/LJJ61///6Bf9iwYda+ffugzxYvXmzLli1rNX120EEH2WuvvWaLFi1q+Dd8+HA75ZRTGtqtvY/MzEaPHh1LeXz77bdt++23NzOz/v37W3l5edBPNTU19tJLL7Wafvrss8+sTZvw1tm2bVurr683M+8jsjH9MWrUKEulUrZw4cKGZZ588kmrr6+3kSNHbvZtzlm29Ki8TeGee+5Jd+jQIX377ben33jjjfRZZ52VLikpSVdXV2/pTdsinHPOOemuXbumn3766fS//vWvhn+fffZZwzLf/OY30/369Us/+eST6Zdffjk9atSo9KhRo7bgVm95dJR7Ou19lE6vT+lr165d+oorrki/88476bvvvju9zTbbpO+6666GZa666qp0SUlJ+ve//336r3/9a/rII4/M65QsMn78+HSfPn0a0tbuu+++dM+ePdMXXnhhwzKtrY8++eST9Kuvvpp+9dVX02aWvuaaa9Kvvvpq+u9//3s6nd64/jjkkEPSe+65Z/qll15KP/fcc+mddtrJ09ayJCcf6Ol0Ov3Tn/403a9fv3RhYWF67733Tr/44otbepO2GGa2wX+33XZbwzJr1qxJn3vuuelu3bqlt9lmm/TRRx+d/te//rXlNnorgA9076P1PPTQQ+nddtst3aFDh/TAgQPTN910U+Cvr69PT5s2LV1WVpbu0KFD+qCDDkovXrx4C23t5qempiZ9/vnnp/v165cuKipK77DDDunvfe976bVr1zYs09r66KmnntrgPWj8+PHpdHrj+uPf//53+qSTTkp37tw5XVxcnJ4wYUL6k08+2QJ7k7v4fOiO4ziOkwfknIbuOI7jOE4cf6A7juM4Th7gD3THcRzHyQP8ge44juM4eYA/0B3HcRwnD/AHuuM4juPkAf5AdxzHcZw8wB/ojuM4jpMH+APdcRzHcfIAf6A7juM4Th7gD3THcRzHyQP+P/8WPxrYUfX9AAAAAElFTkSuQmCC", + "image/png": "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", "text/plain": [ "
" ] @@ -259,7 +271,7 @@ "\n", "problem = HeatConduction2D()\n", "design, _ = problem.random_design()\n", - "problem.render(design)\n" + "problem.render(design)" ] }, { @@ -283,7 +295,7 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 4, "metadata": {}, "outputs": [ { @@ -340,12 +352,12 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Now, let's say we want to train the model for the heat conduction problem, use weights and biases to track the training process, and save the resulting model in wandb. The resulting command would look like this:" + "Now, let's say we want to train the model for the heat conduction problem, use weights and biases to track the training process, and save the resulting model in wandb. The resulting command would look like this (will take a while to run):" ] }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 5, "metadata": {}, "outputs": [ { @@ -358,125 +370,50 @@ " gym.logger.warn(\n", "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mflorian-felten\u001b[0m (\u001b[33mengibench\u001b[0m) to \u001b[32mhttps://api.wandb.ai\u001b[0m. Use \u001b[1m`wandb login --relogin`\u001b[0m to force relogin\n", "\u001b[34m\u001b[1mwandb\u001b[0m: Tracking run with wandb version 0.19.10\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Run data is saved locally in \u001b[35m\u001b[1m/Users/ffelte/Documents/EngiLearn/wandb/run-20250709_094019-9lgcuj7t\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run data is saved locally in \u001b[35m\u001b[1m/Users/ffelte/Documents/EngiLearn/wandb/run-20250709_140832-j9xi9nsl\u001b[0m\n", "\u001b[34m\u001b[1mwandb\u001b[0m: Run \u001b[1m`wandb offline`\u001b[0m to turn off syncing.\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Syncing run \u001b[33mheatconduction2d__diffusion_2d_cond__1__1752046819\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Syncing run \u001b[33mheatconduction2d__diffusion_2d_cond__1__1752062911\u001b[0m\n", "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at \u001b[34m\u001b[4mhttps://wandb.ai/engibench/engiopt\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run at \u001b[34m\u001b[4mhttps://wandb.ai/engibench/engiopt/runs/9lgcuj7t\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run at \u001b[34m\u001b[4mhttps://wandb.ai/engibench/engiopt/runs/j9xi9nsl\u001b[0m\n", + " 0%| | 0/200 [00:00\n", - " training_ds = problem.dataset.with_format(\"torch\", device=device)[\"train\"]\n", - " ^^^^^^^^^^^^^^^\n", - " File \"/Users/ffelte/Documents/EngiBench/engibench/core.py\", line 110, in dataset\n", - " self._dataset = load_dataset(self.dataset_id)\n", - " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", - " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/datasets/load.py\", line 2132, in load_dataset\n", - " builder_instance = load_dataset_builder(\n", - " ^^^^^^^^^^^^^^^^^^^^^\n", - " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/datasets/load.py\", line 1853, in load_dataset_builder\n", - " dataset_module = dataset_module_factory(\n", - " ^^^^^^^^^^^^^^^^^^^^^^^\n", - " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/datasets/load.py\", line 1694, in dataset_module_factory\n", - " ).get_module()\n", - " ^^^^^^^^^^^^\n", - " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/datasets/load.py\", line 1041, in get_module\n", - " exported_dataset_infos = _dataset_viewer.get_exported_dataset_infos(\n", - " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", - " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/datasets/utils/_dataset_viewer.py\", line 71, in get_exported_dataset_infos\n", - " info_response = get_session().get(\n", - " ^^^^^^^^^^^^^^^^^^\n", - " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/requests/sessions.py\", line 602, in get\n", - " return self.request(\"GET\", url, **kwargs)\n", - " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", - " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/requests/sessions.py\", line 589, in request\n", - " resp = self.send(prep, **send_kwargs)\n", - " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", - " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/requests/sessions.py\", line 703, in send\n", - " r = adapter.send(request, **kwargs)\n", - " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", - " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/huggingface_hub/utils/_http.py\", line 96, in send\n", - " return super().send(request, *args, **kwargs)\n", - " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", - " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/requests/adapters.py\", line 667, in send\n", - " resp = conn.urlopen(\n", - " ^^^^^^^^^^^^^\n", - " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/urllib3/connectionpool.py\", line 789, in urlopen\n", - " response = self._make_request(\n", - " ^^^^^^^^^^^^^^^^^^^\n", - " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/urllib3/connectionpool.py\", line 536, in _make_request\n", - " response = conn.getresponse()\n", - " ^^^^^^^^^^^^^^^^^^\n", - " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/urllib3/connection.py\", line 507, in getresponse\n", - " httplib_response = super().getresponse()\n", - " ^^^^^^^^^^^^^^^^^^^^^\n", - " File \"/opt/homebrew/Cellar/python@3.12/3.12.9/Frameworks/Python.framework/Versions/3.12/lib/python3.12/http/client.py\", line 1430, in getresponse\n", - " response.begin()\n", - " File \"/opt/homebrew/Cellar/python@3.12/3.12.9/Frameworks/Python.framework/Versions/3.12/lib/python3.12/http/client.py\", line 331, in begin\n", - " version, status, reason = self._read_status()\n", - " ^^^^^^^^^^^^^^^^^^^\n", - " File \"/opt/homebrew/Cellar/python@3.12/3.12.9/Frameworks/Python.framework/Versions/3.12/lib/python3.12/http/client.py\", line 292, in _read_status\n", - " line = str(self.fp.readline(_MAXLINE + 1), \"iso-8859-1\")\n", - " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", - " File \"/opt/homebrew/Cellar/python@3.12/3.12.9/Frameworks/Python.framework/Versions/3.12/lib/python3.12/socket.py\", line 720, in readinto\n", - " return self._sock.recv_into(b)\n", - " ^^^^^^^^^^^^^^^^^^^^^^^\n", - " File \"/opt/homebrew/Cellar/python@3.12/3.12.9/Frameworks/Python.framework/Versions/3.12/lib/python3.12/ssl.py\", line 1251, in recv_into\n", - " return self.read(nbytes, buffer)\n", - " ^^^^^^^^^^^^^^^^^^^^^^^^^\n", - " File \"/opt/homebrew/Cellar/python@3.12/3.12.9/Frameworks/Python.framework/Versions/3.12/lib/python3.12/ssl.py\", line 1103, in read\n", - " return self._sslobj.read(len, buffer)\n", - " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/engiopt/diffusion_2d_cond/diffusion_2d_cond.py\", line 382, in \n", + " optimizer.step()\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/torch/optim/optimizer.py\", line 487, in wrapper\n", + " out = func(*args, **kwargs)\n", + " ^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/torch/optim/optimizer.py\", line 91, in _use_grad\n", + " ret = func(self, *args, **kwargs)\n", + " ^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/torch/optim/adamw.py\", line 220, in step\n", + " adamw(\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/torch/optim/optimizer.py\", line 154, in maybe_fallback\n", + " return func(*args, **kwargs)\n", + " ^^^^^^^^^^^^^^^^^^^^^\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/torch/optim/adamw.py\", line 782, in adamw\n", + " func(\n", + " File \"/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/torch/optim/adamw.py\", line 376, in _single_tensor_adamw\n", + " exp_avg_sq.mul_(beta2).addcmul_(grad, grad, value=1 - beta2)\n", + " ^^^^^^^^^^^^^^^^^^^^^^\n", "KeyboardInterrupt\n" ] } @@ -489,16 +426,16 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Restoring a model and using it to generate designs\n", + "## Restoring a model\n", "\n", - "Now we can: \n", + "After training a model, we can: \n", "1. Download a model from wandb\n", - "2. Use the model to generate designs" + "2. Instantiade the model and its sampler from the artifact" ] }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 6, "metadata": {}, "outputs": [ { @@ -507,8 +444,8 @@ "text": [ "\u001b[34m\u001b[1mwandb\u001b[0m: Downloading large artifact heatconduction2d_diffusion_2d_cond_model:seed_1, 200.90MB. 1 files... \n", "\u001b[34m\u001b[1mwandb\u001b[0m: 1 of 1 files downloaded. \n", - "Done. 0:0:1.4\n", - "/var/folders/fw/69hmbxm946l1xycg8w29nw0800l9gd/T/ipykernel_88800/2524680856.py:13: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n", + "Done. 0:0:0.8\n", + "/var/folders/fw/69hmbxm946l1xycg8w29nw0800l9gd/T/ipykernel_6077/97458895.py:20: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n", " ckpt = th.load(ckpt_path, map_location=device)\n" ] } @@ -525,7 +462,14 @@ "artifact = api.artifact(artifact_path, type=\"model\")\n", "artifact_dir = artifact.download()\n", "ckpt_path = os.path.join(artifact_dir, \"model.pth\")\n", - "device = th.device(\"cuda\" if th.cuda.is_available() else \"cpu\")\n", + "\n", + "if th.cuda.is_available():\n", + " device = th.device(\"cuda\")\n", + "elif th.backends.mps.is_available():\n", + " device = th.device(\"mps\")\n", + "else:\n", + " device = th.device(\"cpu\")\n", + "\n", "ckpt = th.load(ckpt_path, map_location=device)\n", "\n", "# We will also need the \"run\" object to get the model configuration which was used to train the model.\n", @@ -549,7 +493,7 @@ "model.load_state_dict(ckpt[\"model\"])\n", "model.eval()\n", "\n", - "# The noise schedule also must be restored\n", + "# The noise schedule must also be restored\n", "options = {\n", " \"cosine\": run.config[\"noise_schedule\"] == \"cosine\",\n", " \"exp_biasing\": run.config[\"noise_schedule\"] == \"exp\",\n", @@ -562,18 +506,26 @@ " scale=1.0,\n", " options=options,\n", ")\n", - "ddm_sampler = DiffusionSampler(run.config[\"num_timesteps\"], betas)\n", - "\n" + "ddm_sampler = DiffusionSampler(run.config[\"num_timesteps\"], betas)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Sampling from a model\n", + "\n", + "Now we can use the model to generate designs." ] }, { "cell_type": "code", - "execution_count": 20, + "execution_count": 7, "metadata": {}, "outputs": [ { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -583,7 +535,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -593,7 +545,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -603,17 +555,17 @@ } ], "source": [ - "# Now we can use the model to generate designs\n", + "\n", "n_samples = 3\n", "\n", "conditions = [\n", - " {\"volume\": 0.5, \"length\": 0.5},\n", + " {\"volume\": 0.3, \"length\": 0.5},\n", " {\"volume\": 0.4, \"length\": 0.7},\n", " {\"volume\": 0.2, \"length\": 0.1},\n", "]\n", "\n", "conditions_tensor = th.tensor([list(c.values()) for c in conditions], device=device)\n", - "conditions_tensor = conditions_tensor.unsqueeze(1) # Add channel dim\n", + "conditions_tensor = conditions_tensor.unsqueeze(1) # Add channel dim\n", "\n", "design_shape = problem.design_space.shape\n", "# Denoise the designs in batch\n", @@ -623,10 +575,11 @@ " gen_designs = ddm_sampler.sample_timestep(model, gen_designs, t, conditions_tensor)\n", "\n", "gen_designs = gen_designs.cpu().numpy()\n", + "gen_designs = gen_designs.squeeze() # Remove the channel dim\n", "\n", "# Render the designs\n", "for i in range(n_samples):\n", - " problem.render(gen_designs[i][0]) # 0 removes the channel dim" + " problem.render(gen_designs[i])" ] }, { @@ -636,9 +589,18 @@ "## Evaluating the model" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Optimizing generated designs\n", + "\n", + "Let's first optimize the generated designs, to see how far away they are from optimal designs" + ] + }, { "cell_type": "code", - "execution_count": 24, + "execution_count": 8, "metadata": {}, "outputs": [ { @@ -658,7 +620,7 @@ "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.931e-17 (tol = 1.000e-07) r (rel) = 4.977e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.086e-17 (tol = 1.000e-07) r (rel) = 5.134e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", @@ -670,7 +632,7 @@ " file_print_level = 5 1\n", " hessian_approximation = limited-memory 7\n", " max_iter = 100 1\n", - " output_file = /home/fenics/shared/templates/RES_OPT/solution_V=0.5_w=0.5.txt 1\n", + " output_file = /home/fenics/shared/templates/RES_OPT/solution_V=0.3_w=0.5.txt 1\n", " print_level = 6 2\n", "\n", "******************************************************************************\n", @@ -704,7 +666,7 @@ "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", - "Least square estimates max(y_c) = 0.000000e+00, max(y_d) = 9.999282e-01\n", + "Least square estimates max(y_c) = 0.000000e+00, max(y_d) = 9.999011e-01\n", "Total number of variables............................: 10201\n", " variables with only lower bounds: 0\n", " variables with lower and upper bounds: 10201\n", @@ -719,7 +681,7 @@ "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 3.947e-17 (tol = 1.000e-07) r (rel) = 3.985e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.163e-17 (tol = 1.000e-07) r (rel) = 5.212e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "\n", @@ -735,7 +697,7 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 0 4.6981653e-04 0.00e+00 1.19e-02 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", + " 0 1.8572449e-04 0.00e+00 1.00e-04 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", "\n", "**************************************************\n", "*** Beginning Iteration 0 from the following point:\n", @@ -744,10 +706,10 @@ "Current barrier parameter mu = 1.0000000000000000e+00\n", "Current fraction-to-the-boundary parameter tau = 0.0000000000000000e+00\n", "\n", - "||curr_x||_inf = 5.7071268558502197e-01\n", - "||curr_s||_inf = 3.3710614818985063e-01\n", + "||curr_x||_inf = 8.2412612438201904e-01\n", + "||curr_s||_inf = 1.1129260906092758e-01\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 9.9992815818945313e-01\n", + "||curr_y_d||_inf = 9.9990112949266385e-01\n", "||curr_z_L||_inf = 1.0000000000000000e+00\n", "||curr_z_U||_inf = 1.0000000000000000e+00\n", "||curr_v_L||_inf = 1.0000000000000000e+00\n", @@ -757,8 +719,8 @@ "***Current NLP Values for Iteration 0:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 4.6981652935133546e-04 4.6981652935133546e-04\n", - "Dual infeasibility......: 1.1890392544966710e-02 1.1890392544966710e-02\n", + "Objective...............: 1.8572448735931820e-04 1.8572448735931820e-04\n", + "Dual infeasibility......: 1.0001366584133642e-04 1.0001366584133642e-04\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", "Complementarity.........: 9.9000001999999998e-01 9.9000001999999998e-01\n", "Overall NLP error.......: 9.9000001999999998e-01 9.9000001999999998e-01\n", @@ -770,7 +732,7 @@ "*** Update Barrier Parameter for Iteration 0:\n", "**************************************************\n", "\n", - "Setting mu_max to 4.999920e+02.\n", + "Setting mu_max to 4.999810e+02.\n", "Staying in free mu mode.\n", "The current filter has 1 entries.\n", "Solving the Primal Dual System for the affine step\n", @@ -782,18 +744,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 1.302505e-02\n", - "Barrier parameter mu computed by oracle is 6.512420e-03\n", - "Barrier parameter mu after safeguards is 6.512420e-03\n", - "Barrier Parameter: 6.512420e-03\n", + "Sigma = 1.000000e-06\n", + "Barrier parameter mu computed by oracle is 4.999810e-07\n", + "Barrier parameter mu after safeguards is 4.999810e-07\n", + "Barrier Parameter: 4.999810e-07\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 0:\n", "**************************************************\n", "\n", - "residual_ratio = 2.090756e-15\n", + "residual_ratio = 4.033025e-16\n", "Factorization successful.\n", - "residual_ratio = 1.123732e-16\n", + "residual_ratio = 1.115802e-16\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 0:\n", @@ -810,11 +772,11 @@ "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 3.995e-17 (tol = 1.000e-07) r (rel) = 4.033e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.031e-17 (tol = 1.000e-07) r (rel) = 5.079e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", - " New values of barrier function = 1.9915250594961395e+02 (reference 2.1626223139200656e+02):\n", - " New values of constraint violation = 2.7755575615628914e-16 (reference 0.0000000000000000e+00):\n", + " New values of barrier function = 1.7761639178031554e-02 (reference 1.7761788605270844e-02):\n", + " New values of constraint violation = 6.5225602696727947e-16 (reference 0.0000000000000000e+00):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -829,11 +791,11 @@ "*** Update HessianMatrix for Iteration 1:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 6.437118e-03\n", + "In limited-memory update, s_new_max is 7.065045e-07\n", "Limited-Memory test for skipping:\n", - " s^Ty = 1.695689e-03 snrm = 5.050718e-01 ynrm = 2.704053e-02\n", + " s^Ty = 1.375760e-11 snrm = 3.988065e-05 ynrm = 6.903371e-07\n", " Perform the update.\n", - "sigma (for B0) is 6.647218e-03\n", + "sigma (for B0) is 8.650042e-03\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -842,33 +804,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 1 4.6869866e-04 0.00e+00 6.30e-03 -2.2 6.44e-03 - 9.98e-01 1.00e+00f 1 sigma=1.30e-02 qf=13y \n", + " 1 1.8572445e-04 0.00e+00 1.21e-06 -6.3 7.07e-07 - 9.90e-01 1.00e+00f 1 sigma=1.00e-06 qf=13y \n", "\n", "**************************************************\n", "*** Beginning Iteration 1 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 6.5124202235485916e-03\n", + "Current barrier parameter mu = 4.9998095851772871e-07\n", "Current fraction-to-the-boundary parameter tau = 9.8999999999999999e-01\n", "\n", - "||curr_x||_inf = 5.6997274809964926e-01\n", - "||curr_s||_inf = 3.3285290627583736e-01\n", + "||curr_x||_inf = 8.2412596215053446e-01\n", + "||curr_s||_inf = 1.1129228096813605e-01\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 3.8559795683230300e-02\n", - "||curr_z_L||_inf = 2.1724872561545538e-02\n", - "||curr_z_U||_inf = 1.5901675300216689e-02\n", - "||curr_v_L||_inf = 3.4344526508572670e-02\n", + "||curr_y_d||_inf = 1.0006971437128736e-02\n", + "||curr_z_L||_inf = 1.0001740723567520e-02\n", + "||curr_z_U||_inf = 1.0005053859429958e-02\n", + "||curr_v_L||_inf = 1.0007638643969696e-02\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 1:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 4.6869866234459505e-04 4.6869866234459505e-04\n", - "Dual infeasibility......: 6.3037649451827341e-03 6.3037649451827341e-03\n", + "Objective...............: 1.8572445253169502e-04 1.8572445253169502e-04\n", + "Dual infeasibility......: 1.2092250787718722e-06 1.2092250787718722e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 1.5280891981050969e-02 1.5280891981050969e-02\n", - "Overall NLP error.......: 1.5280891981050969e-02 1.5280891981050969e-02\n", + "Complementarity.........: 9.9012414945559273e-03 9.9012414945559273e-03\n", + "Overall NLP error.......: 9.9012414945559273e-03 9.9012414945559273e-03\n", "\n", "\n", "\n", @@ -877,9 +839,13 @@ "*** Update Barrier Parameter for Iteration 1:\n", "**************************************************\n", "\n", - "Staying in free mu mode.\n", - "The current filter has 2 entries.\n", - "Solving the Primal Dual System for the affine step\n", + "Switching to fixed mu mode with mu = 4.0003526782451376e-03 and tau = 9.9599964732175483e-01.\n", + "Barrier Parameter: 4.000353e-03\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 1:\n", + "**************************************************\n", + "\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -888,20 +854,9 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "Solving the Primal Dual System for the centering step\n", - "Factorization successful.\n", - "Sigma = 1.000000e-06\n", - "Barrier parameter mu computed by oracle is 7.753027e-09\n", - "Barrier parameter mu after safeguards is 7.753027e-09\n", - "Barrier Parameter: 7.753027e-09\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 1:\n", - "**************************************************\n", - "\n", - "residual_ratio = 1.284014e-14\n", + "residual_ratio = 1.030212e-15\n", "Factorization successful.\n", - "residual_ratio = 9.665972e-17\n", + "residual_ratio = 1.981177e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 1:\n", @@ -910,17 +865,17 @@ "--> Starting line search in iteration 1 <--\n", "Mu has changed in line search - resetting watchdog counters.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.710247E-18\n", - "Starting checks for alpha (primal) = 1.00e+00\n", + "minimal step size ALPHA_MIN = 3.394411E-28\n", + "Starting checks for alpha (primal) = 4.51e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 3.950e-17 (tol = 1.000e-07) r (rel) = 3.987e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.806e-17 (tol = 1.000e-07) r (rel) = 4.852e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", - " New values of barrier function = 7.0570783360603562e-04 (reference 7.0578887277047671e-04):\n", - " New values of constraint violation = 5.5511151231257827e-17 (reference 2.7755575615628914e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 4.507222e-01:\n", + " New values of barrier function = 7.0390390173741793e+01 (reference 1.4062525623354958e+02):\n", + " New values of constraint violation = 9.7144514654701197e-16 (reference 6.5225602696727947e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -933,12 +888,11 @@ "*** Update HessianMatrix for Iteration 2:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 1.656226e-04\n", + "In limited-memory update, s_new_max is 3.130631e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 2.210383e-11 snrm = 1.074683e-03 ynrm = 2.503084e-08\n", - " Perform the update.\n", - "sigma (for B0) is 1.913845e-05\n", - "Number of successive iterations with skipping: 0\n", + " s^Ty = -1.790153e-06 snrm = 1.499971e+01 ynrm = 1.679536e-05\n", + " Skip the update.\n", + "Number of successive iterations with skipping: 1\n", "\n", "\n", "**************************************************\n", @@ -946,33 +900,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 2 4.6861860e-04 0.00e+00 6.49e-05 -8.1 1.66e-04 - 9.90e-01 1.00e+00f 1 Nhj sigma=1.00e-06 qf=13y A\n", + " 2 2.4051126e-04 0.00e+00 1.54e-02 -2.4 6.95e-01 - 5.91e-01 4.51e-01f 1 FNhj y Ws\n", "\n", "**************************************************\n", "*** Beginning Iteration 2 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 7.7530267954892570e-09\n", - "Current fraction-to-the-boundary parameter tau = 9.8999999999999999e-01\n", + "Current barrier parameter mu = 4.0003526782451376e-03\n", + "Current fraction-to-the-boundary parameter tau = 9.9599964732175483e-01\n", "\n", - "||curr_x||_inf = 5.6998421098192886e-01\n", - "||curr_s||_inf = 3.3284765987999715e-01\n", + "||curr_x||_inf = 6.9631634276192933e-01\n", + "||curr_s||_inf = 4.4519841424242590e-04\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 3.9757439530749804e-04\n", - "||curr_z_L||_inf = 2.2367460987769358e-04\n", - "||curr_z_U||_inf = 1.6845304188516097e-04\n", - "||curr_v_L||_inf = 3.5414383457792881e-04\n", + "||curr_y_d||_inf = 3.9457009625083139e-02\n", + "||curr_z_L||_inf = 1.9963021675158434e-02\n", + "||curr_z_U||_inf = 1.1964978032678477e-02\n", + "||curr_v_L||_inf = 3.8416859498239253e-02\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 2:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 4.6861859532962020e-04 4.6861859532962020e-04\n", - "Dual infeasibility......: 6.4897480859802347e-05 6.4897480859802347e-05\n", + "Objective...............: 2.4051126086857644e-04 2.4051126086857644e-04\n", + "Dual infeasibility......: 1.5405266669304682e-02 1.5405266669304682e-02\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 1.5732488550385427e-04 1.5732488550385427e-04\n", - "Overall NLP error.......: 1.5732488550385427e-04 1.5732488550385427e-04\n", + "Complementarity.........: 7.1804208850246984e-03 7.1804208850246984e-03\n", + "Overall NLP error.......: 1.5405266669304682e-02 1.5405266669304682e-02\n", "\n", "\n", "\n", @@ -981,9 +935,14 @@ "*** Update Barrier Parameter for Iteration 2:\n", "**************************************************\n", "\n", - "Staying in free mu mode.\n", - "The current filter has 2 entries.\n", - "Solving the Primal Dual System for the affine step\n", + "Remaining in fixed mu mode.\n", + "Reducing mu to 2.5301567154702848e-04 in fixed mu mode. Tau becomes 9.9599964732175483e-01\n", + "Barrier Parameter: 2.530157e-04\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 2:\n", + "**************************************************\n", + "\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -992,20 +951,9 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "Solving the Primal Dual System for the centering step\n", - "Factorization successful.\n", - "Sigma = 1.302505e-02\n", - "Barrier parameter mu computed by oracle is 1.039736e-06\n", - "Barrier parameter mu after safeguards is 1.039736e-06\n", - "Barrier Parameter: 1.039736e-06\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 2:\n", - "**************************************************\n", - "\n", - "residual_ratio = 2.930508e-14\n", + "residual_ratio = 1.863863e-16\n", "Factorization successful.\n", - "residual_ratio = 3.808328e-17\n", + "residual_ratio = 3.640358e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 2:\n", @@ -1013,19 +961,18 @@ "\n", "--> Starting line search in iteration 2 <--\n", "Mu has changed in line search - resetting watchdog counters.\n", - "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.784396E-23\n", - "Starting checks for alpha (primal) = 7.27e-01\n", + "minimal step size ALPHA_MIN = 5.357732E-24\n", + "Starting checks for alpha (primal) = 5.92e-02\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.355e-17 (tol = 1.000e-07) r (rel) = 4.396e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.916e-17 (tol = 1.000e-07) r (rel) = 4.962e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 7.265680e-01:\n", - " New values of barrier function = 3.1239024315965386e-02 (reference 3.2263967256597775e-02):\n", - " New values of constraint violation = 1.1102230246251565e-16 (reference 5.5511151231257827e-17):\n", + "Checking acceptability for trial step size alpha_primal_test= 5.922937e-02:\n", + " New values of barrier function = 4.4480923244584814e+00 (reference 4.4523007218524020e+00):\n", + " New values of constraint violation = 5.1309867812876497e-16 (reference 9.7144514654701197e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -1038,11 +985,11 @@ "*** Update HessianMatrix for Iteration 3:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 1.643432e-02\n", + "In limited-memory update, s_new_max is 2.057858e-03\n", "Limited-Memory test for skipping:\n", - " s^Ty = 1.141847e-07 snrm = 2.866860e-01 ynrm = 1.844972e-06\n", + " s^Ty = 6.031093e-10 snrm = 6.938498e-02 ynrm = 1.499893e-07\n", " Perform the update.\n", - "sigma (for B0) is 1.389296e-06\n", + "sigma (for B0) is 1.252752e-07\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -1051,33 +998,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 3 4.6176366e-04 0.00e+00 2.04e-04 -6.0 2.26e-02 - 9.83e-01 7.27e-01f 1 sigma=1.30e-02 qf=13y A\n", + " 3 2.4102359e-04 0.00e+00 8.05e-04 -3.6 3.47e-02 - 1.00e+00 5.92e-02f 1 Fy A\n", "\n", "**************************************************\n", "*** Beginning Iteration 3 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 1.0397358898902715e-06\n", - "Current fraction-to-the-boundary parameter tau = 9.9984267511449609e-01\n", + "Current barrier parameter mu = 2.5301567154702848e-04\n", + "Current fraction-to-the-boundary parameter tau = 9.9599964732175483e-01\n", "\n", - "||curr_x||_inf = 5.7023191573913001e-01\n", - "||curr_s||_inf = 3.3014142602921281e-01\n", + "||curr_x||_inf = 6.9558595276290758e-01\n", + "||curr_s||_inf = 1.7709906722919813e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 1.4169907374203415e-05\n", - "||curr_z_L||_inf = 2.0401200255619860e-04\n", - "||curr_z_U||_inf = 1.1195518519408702e-05\n", - "||curr_v_L||_inf = 1.2976521882577077e-05\n", + "||curr_y_d||_inf = 1.2143630829368719e+00\n", + "||curr_z_L||_inf = 1.2658655007020073e-03\n", + "||curr_z_U||_inf = 7.9478026811410918e-04\n", + "||curr_v_L||_inf = 1.2143253677295873e+00\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 3:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 4.6176366280509224e-04 4.6176366280509224e-04\n", - "Dual infeasibility......: 2.0387405287655886e-04 2.0387405287655886e-04\n", + "Objective...............: 2.4102358960018843e-04 2.4102358960018843e-04\n", + "Dual infeasibility......: 8.0472502989564038e-04 8.0472502989564038e-04\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 4.8643644598696965e-06 4.8643644598696965e-06\n", - "Overall NLP error.......: 2.0387405287655886e-04 2.0387405287655886e-04\n", + "Complementarity.........: 5.5994619120106461e-04 5.5994619120106461e-04\n", + "Overall NLP error.......: 8.0472502989564038e-04 8.0472502989564038e-04\n", "\n", "\n", "\n", @@ -1086,9 +1033,14 @@ "*** Update Barrier Parameter for Iteration 3:\n", "**************************************************\n", "\n", - "Staying in free mu mode.\n", - "The current filter has 3 entries.\n", - "Solving the Primal Dual System for the affine step\n", + "Remaining in fixed mu mode.\n", + "Reducing mu to 4.0245852639214270e-06 in fixed mu mode. Tau becomes 9.9974698432845299e-01\n", + "Barrier Parameter: 4.024585e-06\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 3:\n", + "**************************************************\n", + "\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -1097,20 +1049,9 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "Solving the Primal Dual System for the centering step\n", - "Factorization successful.\n", - "Sigma = 2.126374e-01\n", - "Barrier parameter mu computed by oracle is 4.858674e-07\n", - "Barrier parameter mu after safeguards is 4.858674e-07\n", - "Barrier Parameter: 4.858674e-07\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 3:\n", - "**************************************************\n", - "\n", - "residual_ratio = 9.907082e-16\n", + "residual_ratio = 3.283049e-16\n", "Factorization successful.\n", - "residual_ratio = 1.204137e-16\n", + "residual_ratio = 2.875270e-19\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 3:\n", @@ -1120,17 +1061,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 4.968812E-24\n", - "Starting checks for alpha (primal) = 4.52e-01\n", + "minimal step size ALPHA_MIN = 1.027775E-21\n", + "Starting checks for alpha (primal) = 1.00e+00\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.004e-17 (tol = 1.000e-07) r (rel) = 4.042e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.758e-17 (tol = 1.000e-07) r (rel) = 4.803e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 4.517635e-01:\n", - " New values of barrier function = 1.2061175807744772e-02 (reference 1.4843941377593832e-02):\n", - " New values of constraint violation = 5.5511151231257827e-17 (reference 1.1102230246251565e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 7.0747106848393596e-02 (reference 7.0990620609180072e-02):\n", + " New values of constraint violation = 2.7761504846259694e-16 (reference 5.1309867812876497e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -1143,11 +1084,11 @@ "*** Update HessianMatrix for Iteration 4:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 1.040402e-01\n", + "In limited-memory update, s_new_max is 6.726803e-03\n", "Limited-Memory test for skipping:\n", - " s^Ty = 1.621492e-05 snrm = 3.276010e+00 ynrm = 1.644879e-05\n", + " s^Ty = 3.114592e-07 snrm = 2.997108e-01 ynrm = 1.587526e-06\n", " Perform the update.\n", - "sigma (for B0) is 1.510860e-06\n", + "sigma (for B0) is 3.467341e-06\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -1156,33 +1097,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 4 3.7713666e-04 0.00e+00 5.29e-04 -6.3 2.30e-01 - 9.92e-01 4.52e-01f 1 sigma=2.13e-01 qf=12y A\n", + " 4 2.4871288e-04 0.00e+00 1.06e-06 -5.4 6.73e-03 - 1.00e+00 1.00e+00f 1 Fy A\n", "\n", "**************************************************\n", "*** Beginning Iteration 4 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 4.8586735880636006e-07\n", - "Current fraction-to-the-boundary parameter tau = 9.9979612594712342e-01\n", + "Current barrier parameter mu = 4.0245852639214270e-06\n", + "Current fraction-to-the-boundary parameter tau = 9.9974698432845299e-01\n", "\n", - "||curr_x||_inf = 6.4920348925804061e-01\n", - "||curr_s||_inf = 2.9883849060579321e-01\n", + "||curr_x||_inf = 6.9091064219368747e-01\n", + "||curr_s||_inf = 4.9621469914718535e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 5.4574287619116684e-06\n", - "||curr_z_L||_inf = 5.2967094494315995e-04\n", - "||curr_z_U||_inf = 5.9782717018244279e-06\n", - "||curr_v_L||_inf = 4.2695877359171917e-06\n", + "||curr_y_d||_inf = 8.3932593228814703e-02\n", + "||curr_z_L||_inf = 1.4675535495221209e-05\n", + "||curr_z_U||_inf = 1.0611938704661796e-05\n", + "||curr_v_L||_inf = 8.3932608674677001e-02\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 4:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 3.7713665675027903e-04 3.7713665675027903e-04\n", - "Dual infeasibility......: 5.2918727256519977e-04 5.2918727256519977e-04\n", + "Objective...............: 2.4871287737524418e-04 2.4871287737524418e-04\n", + "Dual infeasibility......: 1.0631618153561819e-06 1.0631618153561819e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 2.1836733900361981e-06 2.1836733900361981e-06\n", - "Overall NLP error.......: 5.2918727256519977e-04 5.2918727256519977e-04\n", + "Complementarity.........: 7.5614440831194792e-06 7.5614440831194792e-06\n", + "Overall NLP error.......: 7.5614440831194792e-06 7.5614440831194792e-06\n", "\n", "\n", "\n", @@ -1191,9 +1132,14 @@ "*** Update Barrier Parameter for Iteration 4:\n", "**************************************************\n", "\n", - "Staying in free mu mode.\n", - "The current filter has 2 entries.\n", - "Solving the Primal Dual System for the affine step\n", + "Remaining in fixed mu mode.\n", + "Reducing mu to 8.0738690075361328e-09 in fixed mu mode. Tau becomes 9.9999597541473606e-01\n", + "Barrier Parameter: 8.073869e-09\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 4:\n", + "**************************************************\n", + "\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -1202,20 +1148,9 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "Solving the Primal Dual System for the centering step\n", - "Factorization successful.\n", - "Sigma = 8.431257e-02\n", - "Barrier parameter mu computed by oracle is 4.044106e-08\n", - "Barrier parameter mu after safeguards is 4.044106e-08\n", - "Barrier Parameter: 4.044106e-08\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 4:\n", - "**************************************************\n", - "\n", - "residual_ratio = 1.872524e-15\n", + "residual_ratio = 2.477764e-15\n", "Factorization successful.\n", - "residual_ratio = 3.467637e-17\n", + "residual_ratio = 5.313539e-19\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 4:\n", @@ -1225,17 +1160,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 7.737005E-23\n", + "minimal step size ALPHA_MIN = 5.132967E-21\n", "Starting checks for alpha (primal) = 1.00e+00\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.397e-17 (tol = 1.000e-07) r (rel) = 4.439e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.496e-17 (tol = 1.000e-07) r (rel) = 4.538e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", - " New values of barrier function = 1.0817665871449902e-03 (reference 1.3496550354855446e-03):\n", - " New values of constraint violation = 0.0000000000000000e+00 (reference 5.5511151231257827e-17):\n", + " New values of barrier function = 3.6706332367466481e-04 (reference 3.9014230695305662e-04):\n", + " New values of constraint violation = 1.3071412442028363e-17 (reference 2.7761504846259694e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -1248,11 +1183,11 @@ "*** Update HessianMatrix for Iteration 5:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 3.996796e-01\n", + "In limited-memory update, s_new_max is 2.049094e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 1.016755e-04 snrm = 5.762415e+00 ynrm = 2.988430e-05\n", + " s^Ty = 7.044544e-06 snrm = 7.731534e-01 ynrm = 1.227059e-05\n", " Perform the update.\n", - "sigma (for B0) is 3.062016e-06\n", + "sigma (for B0) is 1.178478e-05\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -1261,33 +1196,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 5 2.1261573e-04 0.00e+00 1.85e-04 -7.4 4.00e-01 - 6.17e-01 1.00e+00f 1 sigma=8.43e-02 qf=12y A\n", + " 5 2.2478754e-04 0.00e+00 5.15e-06 -8.1 2.05e-01 - 7.60e-01 1.00e+00f 1 Fy A\n", "\n", "**************************************************\n", "*** Beginning Iteration 5 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 4.0441060660647348e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9947081272743477e-01\n", + "Current barrier parameter mu = 8.0738690075361328e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999597541473606e-01\n", "\n", - "||curr_x||_inf = 9.8126103107849083e-01\n", - "||curr_s||_inf = 2.5108965233471248e-01\n", + "||curr_x||_inf = 8.5030114676927715e-01\n", + "||curr_s||_inf = 4.8795497546569574e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 2.5236602439941591e-06\n", - "||curr_z_L||_inf = 1.8555237791197571e-04\n", - "||curr_z_U||_inf = 5.8122187124815824e-06\n", - "||curr_v_L||_inf = 2.1383340779829057e-06\n", + "||curr_y_d||_inf = 2.2434664157648190e-02\n", + "||curr_z_L||_inf = 3.7125796608629522e-06\n", + "||curr_z_U||_inf = 6.8639255528706076e-06\n", + "||curr_v_L||_inf = 2.2434667381497742e-02\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 5:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 2.1261573362645707e-04 2.1261573362645707e-04\n", - "Dual infeasibility......: 1.8534233532916164e-04 1.8534233532916164e-04\n", + "Objective...............: 2.2478753838083446e-04 2.2478753838083446e-04\n", + "Dual infeasibility......: 5.1460032095092485e-06 5.1460032095092485e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 8.4195637365473728e-07 8.4195637365473728e-07\n", - "Overall NLP error.......: 1.8534233532916164e-04 1.8534233532916164e-04\n", + "Complementarity.........: 1.8255007743336357e-06 1.8255007743336357e-06\n", + "Overall NLP error.......: 5.1460032095092485e-06 5.1460032095092485e-06\n", "\n", "\n", "\n", @@ -1296,9 +1231,13 @@ "*** Update Barrier Parameter for Iteration 5:\n", "**************************************************\n", "\n", - "Staying in free mu mode.\n", - "The current filter has 1 entries.\n", - "Solving the Primal Dual System for the affine step\n", + "Remaining in fixed mu mode.\n", + "Barrier Parameter: 8.073869e-09\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 5:\n", + "**************************************************\n", + "\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -1307,40 +1246,28 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "Solving the Primal Dual System for the centering step\n", - "Factorization successful.\n", - "Sigma = 5.306571e-01\n", - "Barrier parameter mu computed by oracle is 1.098352e-07\n", - "Barrier parameter mu after safeguards is 1.098352e-07\n", - "Barrier Parameter: 1.098352e-07\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 5:\n", - "**************************************************\n", - "\n", - "residual_ratio = 5.194011e-15\n", + "residual_ratio = 1.164261e-14\n", "Factorization successful.\n", - "residual_ratio = 6.412359e-17\n", + "residual_ratio = 2.224279e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 5:\n", "**************************************************\n", "\n", "--> Starting line search in iteration 5 <--\n", - "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 0.000000E+00\n", - "Starting checks for alpha (primal) = 3.48e-01\n", + "minimal step size ALPHA_MIN = 1.319532E-22\n", + "Starting checks for alpha (primal) = 1.00e+00\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.198e-17 (tol = 1.000e-07) r (rel) = 4.237e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.101e-17 (tol = 1.000e-07) r (rel) = 4.139e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 3.476103e-01:\n", - " New values of barrier function = 2.2646585305238442e-03 (reference 2.5731707289686824e-03):\n", - " New values of constraint violation = 1.3877787807814457e-16 (reference 0.0000000000000000e+00):\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 3.2590083054756015e-04 (reference 3.6706332367466481e-04):\n", + " New values of constraint violation = 5.0613606730233462e-17 (reference 1.3071412442028363e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -1353,11 +1280,11 @@ "*** Update HessianMatrix for Iteration 6:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 3.008583e-01\n", + "In limited-memory update, s_new_max is 1.503443e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 1.149335e-05 snrm = 5.471624e+00 ynrm = 1.165050e-05\n", + " s^Ty = 1.535142e-05 snrm = 1.864300e+00 ynrm = 9.818142e-06\n", " Perform the update.\n", - "sigma (for B0) is 3.838965e-07\n", + "sigma (for B0) is 4.416895e-06\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -1366,33 +1293,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 6 1.6502648e-04 0.00e+00 2.40e-04 -7.0 8.66e-01 - 7.84e-01 3.48e-01f 1 sigma=5.31e-01 qf=12y A\n", + " 6 1.8088949e-04 0.00e+00 1.19e-06 -8.1 1.50e-01 - 8.31e-01 1.00e+00f 1 Fy A\n", "\n", "**************************************************\n", "*** Beginning Iteration 6 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 1.0983518841748851e-07\n", - "Current fraction-to-the-boundary parameter tau = 9.9981465766467081e-01\n", + "Current barrier parameter mu = 8.0738690075361328e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999597541473606e-01\n", "\n", - "||curr_x||_inf = 9.9994528190258514e-01\n", - "||curr_s||_inf = 2.0296687018492257e-01\n", + "||curr_x||_inf = 8.6633944240021488e-01\n", + "||curr_s||_inf = 4.7222866367873221e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 2.2864271431690781e-06\n", - "||curr_z_L||_inf = 2.3964832790028859e-04\n", - "||curr_z_U||_inf = 8.6986550843502061e-06\n", - "||curr_v_L||_inf = 1.7290386471529564e-06\n", + "||curr_y_d||_inf = 5.7600186012461531e-03\n", + "||curr_z_L||_inf = 8.3111298984257371e-07\n", + "||curr_z_U||_inf = 1.8149395552694020e-06\n", + "||curr_v_L||_inf = 5.7600196653849398e-03\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 6:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.6502648484603295e-04 1.6502648484603295e-04\n", - "Dual infeasibility......: 2.3951685525649055e-04 2.3951685525649055e-04\n", + "Objective...............: 1.8088949364957464e-04 1.8088949364957464e-04\n", + "Dual infeasibility......: 1.1916321970272016e-06 1.1916321970272016e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 3.9210848213316732e-07 3.9210848213316732e-07\n", - "Overall NLP error.......: 2.3951685525649055e-04 2.3951685525649055e-04\n", + "Complementarity.........: 3.1754896585744371e-07 3.1754896585744371e-07\n", + "Overall NLP error.......: 1.1916321970272016e-06 1.1916321970272016e-06\n", "\n", "\n", "\n", @@ -1401,7 +1328,7 @@ "*** Update Barrier Parameter for Iteration 6:\n", "**************************************************\n", "\n", - "Staying in free mu mode.\n", + "Switching back to free mu mode.\n", "The current filter has 2 entries.\n", "Solving the Primal Dual System for the affine step\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", @@ -1414,18 +1341,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 6.069074e-01\n", - "Barrier parameter mu computed by oracle is 8.231647e-08\n", - "Barrier parameter mu after safeguards is 8.231647e-08\n", - "Barrier Parameter: 8.231647e-08\n", + "Sigma = 1.996573e-01\n", + "Barrier parameter mu computed by oracle is 3.362577e-08\n", + "Barrier parameter mu after safeguards is 3.362577e-08\n", + "Barrier Parameter: 3.362577e-08\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 6:\n", "**************************************************\n", "\n", - "residual_ratio = 8.194075e-15\n", + "residual_ratio = 7.617584e-15\n", "Factorization successful.\n", - "residual_ratio = 4.881576e-16\n", + "residual_ratio = 8.568955e-19\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 6:\n", @@ -1435,17 +1362,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.202324E-22\n", - "Starting checks for alpha (primal) = 6.32e-01\n", + "minimal step size ALPHA_MIN = 7.719655E-22\n", + "Starting checks for alpha (primal) = 9.26e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.274e-17 (tol = 1.000e-07) r (rel) = 4.315e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.562e-17 (tol = 1.000e-07) r (rel) = 4.605e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 6.316483e-01:\n", - " New values of barrier function = 1.4999285968201378e-03 (reference 1.7386050838862769e-03):\n", - " New values of constraint violation = 3.3306690738754696e-16 (reference 1.3877787807814457e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 9.259950e-01:\n", + " New values of barrier function = 7.6292270100572794e-04 (reference 7.8482763483398105e-04):\n", + " New values of constraint violation = 1.2358549513619144e-16 (reference 5.0613606730233462e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -1458,11 +1385,11 @@ "*** Update HessianMatrix for Iteration 7:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 5.026466e-01\n", + "In limited-memory update, s_new_max is 1.775743e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 4.164968e-05 snrm = 1.010274e+01 ynrm = 1.104624e-05\n", + " s^Ty = 1.263361e-05 snrm = 2.220251e+00 ynrm = 7.574498e-06\n", " Perform the update.\n", - "sigma (for B0) is 4.080689e-07\n", + "sigma (for B0) is 2.562850e-06\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -1471,33 +1398,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 7 9.0938693e-05 0.00e+00 1.74e-04 -7.1 7.96e-01 - 5.11e-01 6.32e-01f 1 sigma=6.07e-01 qf=12y A\n", + " 7 1.4490905e-04 0.00e+00 7.32e-07 -7.5 1.92e-01 - 9.94e-01 9.26e-01f 1 sigma=2.00e-01 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 7 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 8.2316471722319226e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9976048314474353e-01\n", + "Current barrier parameter mu = 3.3625767094389832e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999880836780297e-01\n", "\n", - "||curr_x||_inf = 9.9987958881490280e-01\n", - "||curr_s||_inf = 1.1226715633901446e-01\n", + "||curr_x||_inf = 9.9999985072576414e-01\n", + "||curr_s||_inf = 8.5632546004813691e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 2.0399870080882624e-06\n", - "||curr_z_L||_inf = 1.7431917112882426e-04\n", - "||curr_z_U||_inf = 3.2767280368029262e-06\n", - "||curr_v_L||_inf = 1.6778132125913295e-06\n", + "||curr_y_d||_inf = 2.0785282382648398e-03\n", + "||curr_z_L||_inf = 2.9317796180187508e-07\n", + "||curr_z_U||_inf = 2.2092410046839914e-06\n", + "||curr_v_L||_inf = 2.0785137940958461e-03\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 7:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 9.0938693341593464e-05 9.0938693341593464e-05\n", - "Dual infeasibility......: 1.7421276895974660e-04 1.7421276895974660e-04\n", + "Objective...............: 1.4490904732650581e-04 1.4490904732650581e-04\n", + "Dual infeasibility......: 7.3218678773143602e-07 7.3218678773143602e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 2.1454017604689819e-07 2.1454017604689819e-07\n", - "Overall NLP error.......: 1.7421276895974660e-04 1.7421276895974660e-04\n", + "Complementarity.........: 8.2321848990866307e-08 8.2321848990866307e-08\n", + "Overall NLP error.......: 7.3218678773143602e-07 7.3218678773143602e-07\n", "\n", "\n", "\n", @@ -1519,18 +1446,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 5.908181e-01\n", - "Barrier parameter mu computed by oracle is 6.304055e-08\n", - "Barrier parameter mu after safeguards is 6.304055e-08\n", - "Barrier Parameter: 6.304055e-08\n", + "Sigma = 2.810308e-01\n", + "Barrier parameter mu computed by oracle is 9.963785e-09\n", + "Barrier parameter mu after safeguards is 9.963785e-09\n", + "Barrier Parameter: 9.963785e-09\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 7:\n", "**************************************************\n", "\n", - "residual_ratio = 1.185736e-15\n", + "residual_ratio = 7.649897e-16\n", "Factorization successful.\n", - "residual_ratio = 2.195808e-17\n", + "residual_ratio = 2.009418e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 7:\n", @@ -1540,17 +1467,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 7.005455E-22\n", - "Starting checks for alpha (primal) = 7.79e-01\n", + "minimal step size ALPHA_MIN = 1.873008E-22\n", + "Starting checks for alpha (primal) = 7.66e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 3.696e-17 (tol = 1.000e-07) r (rel) = 3.731e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.380e-17 (tol = 1.000e-07) r (rel) = 4.421e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 7.794678e-01:\n", - " New values of barrier function = 1.0775074420967959e-03 (reference 1.1699875213672679e-03):\n", - " New values of constraint violation = 7.4940054162198066e-16 (reference 3.3306690738754696e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 7.661435e-01:\n", + " New values of barrier function = 3.0029377200096221e-04 (reference 3.2803514044315652e-04):\n", + " New values of constraint violation = 2.6688144695499044e-16 (reference 1.2358549513619144e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -1563,11 +1490,11 @@ "*** Update HessianMatrix for Iteration 8:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 4.924977e-01\n", + "In limited-memory update, s_new_max is 3.018803e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 6.898312e-06 snrm = 8.168003e+00 ynrm = 4.454266e-06\n", + " s^Ty = 1.452601e-05 snrm = 3.181887e+00 ynrm = 6.264839e-06\n", " Perform the update.\n", - "sigma (for B0) is 1.033978e-07\n", + "sigma (for B0) is 1.434752e-06\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -1576,33 +1503,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 8 6.7419287e-05 0.00e+00 1.15e-05 -7.2 6.32e-01 - 9.34e-01 7.79e-01f 1 sigma=5.91e-01 qf=12y A\n", + " 8 1.0928798e-04 0.00e+00 5.60e-07 -8.0 3.94e-01 - 9.67e-01 7.66e-01f 1 sigma=2.81e-01 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 8 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 6.3040545656067363e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9982578723104021e-01\n", + "Current barrier parameter mu = 9.9637852972779224e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999926781321224e-01\n", "\n", - "||curr_x||_inf = 9.9996010556813475e-01\n", - "||curr_s||_inf = 4.2454333745357481e-02\n", + "||curr_x||_inf = 9.9999996054858042e-01\n", + "||curr_s||_inf = 9.4305136140833296e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 1.9099643631276152e-06\n", - "||curr_z_L||_inf = 1.1534024292732947e-05\n", - "||curr_z_U||_inf = 1.4932187423352213e-06\n", - "||curr_v_L||_inf = 1.8854399454963210e-06\n", + "||curr_y_d||_inf = 9.2694484903164042e-04\n", + "||curr_z_L||_inf = 1.3153394871418884e-07\n", + "||curr_z_U||_inf = 1.4159853410464294e-06\n", + "||curr_v_L||_inf = 9.2692327334428369e-04\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 8:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 6.7419287418862974e-05 6.7419287418862974e-05\n", - "Dual infeasibility......: 1.1468057740637961e-05 1.1468057740637961e-05\n", + "Objective...............: 1.0928798121901817e-04 1.0928798121901817e-04\n", + "Dual infeasibility......: 5.5975515905211401e-07 5.5975515905211401e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 1.3163180480721070e-07 1.3163180480721070e-07\n", - "Overall NLP error.......: 1.1468057740637961e-05 1.1468057740637961e-05\n", + "Complementarity.........: 2.8083811107694010e-08 2.8083811107694010e-08\n", + "Overall NLP error.......: 5.5975515905211401e-07 5.5975515905211401e-07\n", "\n", "\n", "\n", @@ -1624,18 +1551,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 7.141901e-01\n", - "Barrier parameter mu computed by oracle is 4.502846e-08\n", - "Barrier parameter mu after safeguards is 4.502846e-08\n", - "Barrier Parameter: 4.502846e-08\n", + "Sigma = 4.758978e-01\n", + "Barrier parameter mu computed by oracle is 5.700956e-09\n", + "Barrier parameter mu after safeguards is 5.700956e-09\n", + "Barrier Parameter: 5.700956e-09\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 8:\n", "**************************************************\n", "\n", - "residual_ratio = 9.426621e-16\n", + "residual_ratio = 3.950477e-16\n", "Factorization successful.\n", - "residual_ratio = 1.216338e-16\n", + "residual_ratio = 6.605249e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 8:\n", @@ -1645,17 +1572,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.768294E-21\n", - "Starting checks for alpha (primal) = 2.48e-01\n", + "minimal step size ALPHA_MIN = 2.686997E-22\n", + "Starting checks for alpha (primal) = 7.54e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 3.553e-17 (tol = 1.000e-07) r (rel) = 3.587e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.047e-17 (tol = 1.000e-07) r (rel) = 5.095e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 2.479107e-01:\n", - " New values of barrier function = 7.6363642718371054e-04 (reference 7.8890281077480544e-04):\n", - " New values of constraint violation = 2.2204460492503131e-16 (reference 7.4940054162198066e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 7.537482e-01:\n", + " New values of barrier function = 2.0047483415567866e-04 (reference 2.1857532493634615e-04):\n", + " New values of constraint violation = 2.1060966013709825e-16 (reference 2.6688144695499044e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -1668,11 +1595,11 @@ "*** Update HessianMatrix for Iteration 9:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 1.131385e-01\n", + "In limited-memory update, s_new_max is 2.300635e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 2.270529e-07 snrm = 2.688633e+00 ynrm = 9.796588e-07\n", + " s^Ty = 9.816937e-06 snrm = 3.436755e+00 ynrm = 4.021257e-06\n", " Perform the update.\n", - "sigma (for B0) is 3.140973e-08\n", + "sigma (for B0) is 8.311491e-07\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -1681,33 +1608,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 9 6.1616508e-05 0.00e+00 3.02e-07 -7.3 4.56e-01 - 9.86e-01 2.48e-01f 1 sigma=7.14e-01 qf=12y A\n", + " 9 8.4853144e-05 0.00e+00 4.46e-07 -8.2 3.05e-01 - 9.83e-01 7.54e-01f 1 sigma=4.76e-01 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 9 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 4.5028460915594864e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9998853194225934e-01\n", + "Current barrier parameter mu = 5.7009561021759740e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999944024484100e-01\n", "\n", - "||curr_x||_inf = 9.9999880488680837e-01\n", - "||curr_s||_inf = 2.0657262339258076e-02\n", + "||curr_x||_inf = 9.9999993017425948e-01\n", + "||curr_s||_inf = 9.7087544320580980e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 4.9292979228614336e-06\n", - "||curr_z_L||_inf = 2.2492120829200955e-07\n", - "||curr_z_U||_inf = 1.2912012036860158e-06\n", - "||curr_v_L||_inf = 4.9229087298590936e-06\n", + "||curr_y_d||_inf = 5.7393491258124311e-04\n", + "||curr_z_L||_inf = 7.0909329602160347e-08\n", + "||curr_z_U||_inf = 1.3095410294672740e-06\n", + "||curr_v_L||_inf = 5.7392199694711939e-04\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 9:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 6.1616507536971964e-05 6.1616507536971964e-05\n", - "Dual infeasibility......: 3.0169602048343037e-07 3.0169602048343037e-07\n", + "Objective...............: 8.4853143994421878e-05 8.4853143994421878e-05\n", + "Dual infeasibility......: 4.4566903092094602e-07 4.4566903092094602e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 1.0169386633401037e-07 1.0169386633401037e-07\n", - "Overall NLP error.......: 3.0169602048343037e-07 3.0169602048343037e-07\n", + "Complementarity.........: 1.1876709675485983e-08 1.1876709675485983e-08\n", + "Overall NLP error.......: 4.4566903092094602e-07 4.4566903092094602e-07\n", "\n", "\n", "\n", @@ -1717,7 +1644,7 @@ "**************************************************\n", "\n", "Staying in free mu mode.\n", - "The current filter has 3 entries.\n", + "The current filter has 4 entries.\n", "Solving the Primal Dual System for the affine step\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", @@ -1729,18 +1656,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 2.599197e-01\n", - "Barrier parameter mu computed by oracle is 1.166838e-08\n", - "Barrier parameter mu after safeguards is 1.166838e-08\n", - "Barrier Parameter: 1.166838e-08\n", + "Sigma = 5.152993e-01\n", + "Barrier parameter mu computed by oracle is 3.242706e-09\n", + "Barrier parameter mu after safeguards is 3.242706e-09\n", + "Barrier Parameter: 3.242706e-09\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 9:\n", "**************************************************\n", "\n", - "residual_ratio = 5.519903e-16\n", + "residual_ratio = 1.239075e-15\n", "Factorization successful.\n", - "residual_ratio = 2.529956e-16\n", + "residual_ratio = 2.049236e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 9:\n", @@ -1750,17 +1677,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 6.165784E-22\n", - "Starting checks for alpha (primal) = 3.45e-01\n", + "minimal step size ALPHA_MIN = 1.768659E-22\n", + "Starting checks for alpha (primal) = 6.24e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 3.258e-17 (tol = 1.000e-07) r (rel) = 3.288e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.028e-17 (tol = 1.000e-07) r (rel) = 5.076e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 3.446822e-01:\n", - " New values of barrier function = 2.3283950449996471e-04 (reference 2.4353334985162379e-04):\n", - " New values of constraint violation = 5.2041704279304213e-17 (reference 2.2204460492503131e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 6.242119e-01:\n", + " New values of barrier function = 1.4169120735636748e-04 (reference 1.5061880487584128e-04):\n", + " New values of constraint violation = 3.3427308230443709e-17 (reference 2.1060966013709825e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -1773,11 +1700,11 @@ "*** Update HessianMatrix for Iteration 10:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 1.557527e-01\n", + "In limited-memory update, s_new_max is 1.872844e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 6.923097e-07 snrm = 2.573810e+00 ynrm = 1.188048e-06\n", + " s^Ty = 3.746276e-06 snrm = 2.760733e+00 ynrm = 2.003784e-06\n", " Perform the update.\n", - "sigma (for B0) is 1.045075e-07\n", + "sigma (for B0) is 4.915311e-07\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -1786,33 +1713,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 10 5.4256341e-05 0.00e+00 7.69e-07 -7.9 4.52e-01 - 8.27e-01 3.45e-01f 1 sigma=2.60e-01 qf=12y A\n", + " 10 7.2333817e-05 0.00e+00 1.68e-07 -8.5 3.00e-01 - 9.89e-01 6.24e-01f 1 sigma=5.15e-01 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 10 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 1.1668380333389575e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999969830397950e-01\n", + "Current barrier parameter mu = 3.2427059766437888e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999955433096910e-01\n", "\n", - "||curr_x||_inf = 9.9507586558926175e-01\n", - "||curr_s||_inf = 3.7677831445093179e-09\n", + "||curr_x||_inf = 9.9999997907889071e-01\n", + "||curr_s||_inf = 9.1722090811651073e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 1.3138968831959693e-05\n", - "||curr_z_L||_inf = 5.2087917090374842e-08\n", - "||curr_z_U||_inf = 5.6912596151833623e-07\n", - "||curr_v_L||_inf = 1.3126783644354312e-05\n", + "||curr_y_d||_inf = 3.8649323354190133e-04\n", + "||curr_z_L||_inf = 4.2929528640782552e-08\n", + "||curr_z_U||_inf = 6.4516179650102083e-07\n", + "||curr_v_L||_inf = 3.8648319042624229e-04\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 10:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 5.4256340702876417e-05 5.4256340702876417e-05\n", - "Dual infeasibility......: 7.6857144345135908e-07 7.6857144345135908e-07\n", + "Objective...............: 7.2333816527303332e-05 7.2333816527303332e-05\n", + "Dual infeasibility......: 1.6781998495180137e-07 1.6781998495180137e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 6.2549075446186007e-08 6.2549075446186007e-08\n", - "Overall NLP error.......: 7.6857144345135908e-07 7.6857144345135908e-07\n", + "Complementarity.........: 6.2116934612765409e-09 6.2116934612765409e-09\n", + "Overall NLP error.......: 1.6781998495180137e-07 1.6781998495180137e-07\n", "\n", "\n", "\n", @@ -1834,18 +1761,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 5.017880e-01\n", - "Barrier parameter mu computed by oracle is 8.973299e-09\n", - "Barrier parameter mu after safeguards is 8.973299e-09\n", - "Barrier Parameter: 8.973299e-09\n", + "Sigma = 6.079627e-01\n", + "Barrier parameter mu computed by oracle is 2.170914e-09\n", + "Barrier parameter mu after safeguards is 2.170914e-09\n", + "Barrier Parameter: 2.170914e-09\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 10:\n", "**************************************************\n", "\n", - "residual_ratio = 8.618274e-16\n", + "residual_ratio = 4.510048e-16\n", "Factorization successful.\n", - "residual_ratio = 2.205774e-17\n", + "residual_ratio = 4.136234e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 10:\n", @@ -1855,17 +1782,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 2.610811E-23\n", - "Starting checks for alpha (primal) = 1.01e-01\n", + "minimal step size ALPHA_MIN = 2.843753E-23\n", + "Starting checks for alpha (primal) = 3.53e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 3.357e-17 (tol = 1.000e-07) r (rel) = 3.388e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.848e-17 (tol = 1.000e-07) r (rel) = 4.893e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 1.013248e-01:\n", - " New values of barrier function = 1.9043355780963686e-04 (reference 1.9159161092197172e-04):\n", - " New values of constraint violation = 1.4220666744862998e-16 (reference 5.2041704279304213e-17):\n", + "Checking acceptability for trial step size alpha_primal_test= 3.527003e-01:\n", + " New values of barrier function = 1.1500400208612345e-04 (reference 1.1876693177720237e-04):\n", + " New values of constraint violation = 1.1539976873392588e-17 (reference 3.3427308230443709e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -1878,11 +1805,11 @@ "*** Update HessianMatrix for Iteration 11:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 5.578051e-02\n", + "In limited-memory update, s_new_max is 1.326686e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 4.354765e-08 snrm = 6.053372e-01 ynrm = 2.589328e-07\n", + " s^Ty = 8.227656e-07 snrm = 1.663479e+00 ynrm = 8.581268e-07\n", " Perform the update.\n", - "sigma (for B0) is 1.188420e-07\n", + "sigma (for B0) is 2.973320e-07\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -1891,33 +1818,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 11 5.3159759e-05 0.00e+00 4.51e-07 -8.0 5.51e-01 - 9.14e-01 1.01e-01f 1 sigma=5.02e-01 qf=12y A\n", + " 11 6.7048961e-05 0.00e+00 1.43e-07 -8.7 3.76e-01 - 1.00e+00 3.53e-01f 1 sigma=6.08e-01 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 11 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 8.9732992295176388e-09\n", - "Current fraction-to-the-boundary parameter tau = 9.9999923142855651e-01\n", + "Current barrier parameter mu = 2.1709141381359714e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999983218001509e-01\n", "\n", - "||curr_x||_inf = 9.9999996712865646e-01\n", - "||curr_s||_inf = 6.9247794269786765e-05\n", + "||curr_x||_inf = 9.9999999884920221e-01\n", + "||curr_s||_inf = 8.6292423117755249e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 2.4378993444041145e-04\n", - "||curr_z_L||_inf = 4.6761919868473101e-08\n", - "||curr_z_U||_inf = 4.2123404786458668e-07\n", - "||curr_v_L||_inf = 2.4377915681495684e-04\n", + "||curr_y_d||_inf = 3.0123199378696483e-04\n", + "||curr_z_L||_inf = 4.1909700560931919e-08\n", + "||curr_z_U||_inf = 6.7417118202554377e-07\n", + "||curr_v_L||_inf = 3.0122252054007694e-04\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 11:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 5.3159758958582040e-05 5.3159758958582040e-05\n", - "Dual infeasibility......: 4.5071671022282290e-07 4.5071671022282290e-07\n", + "Objective...............: 6.7048960853371178e-05 6.7048960853371178e-05\n", + "Dual infeasibility......: 1.4324731554481346e-07 1.4324731554481346e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 5.8124118448248091e-08 5.8124118448248091e-08\n", - "Overall NLP error.......: 4.5071671022282290e-07 4.5071671022282290e-07\n", + "Complementarity.........: 5.1315041643808989e-09 5.1315041643808989e-09\n", + "Overall NLP error.......: 1.4324731554481346e-07 1.4324731554481346e-07\n", "\n", "\n", "\n", @@ -1927,7 +1854,7 @@ "**************************************************\n", "\n", "Staying in free mu mode.\n", - "The current filter has 3 entries.\n", + "The current filter has 2 entries.\n", "Solving the Primal Dual System for the affine step\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", @@ -1939,18 +1866,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 5.701295e-01\n", - "Barrier parameter mu computed by oracle is 5.727484e-09\n", - "Barrier parameter mu after safeguards is 5.727484e-09\n", - "Barrier Parameter: 5.727484e-09\n", + "Sigma = 8.624418e+00\n", + "Barrier parameter mu computed by oracle is 2.120410e-08\n", + "Barrier parameter mu after safeguards is 2.120410e-08\n", + "Barrier Parameter: 2.120410e-08\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 11:\n", "**************************************************\n", "\n", - "residual_ratio = 3.176722e-15\n", + "residual_ratio = 6.125418e-16\n", "Factorization successful.\n", - "residual_ratio = 6.999270e-18\n", + "residual_ratio = 1.191130e-16\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 11:\n", @@ -1960,17 +1887,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.561954E-23\n", - "Starting checks for alpha (primal) = 1.00e+00\n", + "minimal step size ALPHA_MIN = 1.679671E-26\n", + "Starting checks for alpha (primal) = 1.71e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 3.560e-17 (tol = 1.000e-07) r (rel) = 3.594e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.291e-17 (tol = 1.000e-07) r (rel) = 5.341e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", - " New values of barrier function = 1.3102032591752507e-04 (reference 1.4077898616129297e-04):\n", - " New values of constraint violation = 1.0331091451071250e-16 (reference 1.4220666744862998e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 1.713067e-01:\n", + " New values of barrier function = 5.1949790423902139e-04 (reference 5.3544303240504938e-04):\n", + " New values of constraint violation = 1.1614007552982614e-16 (reference 1.1539976873392588e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -1983,11 +1910,11 @@ "*** Update HessianMatrix for Iteration 12:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 3.078676e-01\n", + "In limited-memory update, s_new_max is 3.830253e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 1.102898e-05 snrm = 9.416229e+00 ynrm = 2.622160e-06\n", + " s^Ty = 1.583843e-06 snrm = 2.549492e+00 ynrm = 1.360322e-06\n", " Perform the update.\n", - "sigma (for B0) is 1.243888e-07\n", + "sigma (for B0) is 2.436715e-07\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -1996,33 +1923,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 12 3.7956764e-05 0.00e+00 7.32e-07 -8.2 3.08e-01 - 6.77e-01 1.00e+00f 1 sigma=5.70e-01 qf=12y A\n", + " 12 7.5850625e-05 0.00e+00 3.46e-07 -7.7 2.24e+00 - 5.59e-01 1.71e-01f 1 sigma=8.62e+00 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 12 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 5.7274844182180619e-09\n", - "Current fraction-to-the-boundary parameter tau = 9.9999954928328982e-01\n", + "Current barrier parameter mu = 2.1204096295430305e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999985675268444e-01\n", "\n", - "||curr_x||_inf = 9.9437953357957953e-01\n", - "||curr_s||_inf = 3.1351610257587199e-05\n", + "||curr_x||_inf = 9.9170758955133220e-01\n", + "||curr_s||_inf = 1.3045157151553733e-05\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 2.2504329294282900e-04\n", - "||curr_z_L||_inf = 3.1309790137290428e-08\n", - "||curr_z_U||_inf = 3.0561553138126096e-07\n", - "||curr_v_L||_inf = 2.2504186731575406e-04\n", + "||curr_y_d||_inf = 1.0028678654938740e-03\n", + "||curr_z_L||_inf = 1.1939250812824760e-07\n", + "||curr_z_U||_inf = 9.1269260435600649e-07\n", + "||curr_v_L||_inf = 1.0028822186672719e-03\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 12:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 3.7956764385632902e-05 3.7956764385632902e-05\n", - "Dual infeasibility......: 7.3247186218490388e-07 7.3247186218490388e-07\n", + "Objective...............: 7.5850624839271036e-05 7.5850624839271036e-05\n", + "Dual infeasibility......: 3.4588449281436793e-07 3.4588449281436793e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 2.4231447904316846e-08 2.4231447904316846e-08\n", - "Overall NLP error.......: 7.3247186218490388e-07 7.3247186218490388e-07\n", + "Complementarity.........: 1.3955793184101916e-08 1.3955793184101916e-08\n", + "Overall NLP error.......: 3.4588449281436793e-07 3.4588449281436793e-07\n", "\n", "\n", "\n", @@ -2031,9 +1958,13 @@ "*** Update Barrier Parameter for Iteration 12:\n", "**************************************************\n", "\n", - "Staying in free mu mode.\n", - "The current filter has 3 entries.\n", - "Solving the Primal Dual System for the affine step\n", + "Switching to fixed mu mode with mu = 1.0234391828744293e-08 and tau = 9.9999998976560822e-01.\n", + "Barrier Parameter: 1.023439e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 12:\n", + "**************************************************\n", + "\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -2042,20 +1973,9 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "Solving the Primal Dual System for the centering step\n", - "Factorization successful.\n", - "Sigma = 9.138677e-01\n", - "Barrier parameter mu computed by oracle is 6.293033e-09\n", - "Barrier parameter mu after safeguards is 6.293033e-09\n", - "Barrier Parameter: 6.293033e-09\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 12:\n", - "**************************************************\n", - "\n", - "residual_ratio = 1.605133e-16\n", + "residual_ratio = 1.018810e-16\n", "Factorization successful.\n", - "residual_ratio = 1.423416e-17\n", + "residual_ratio = 8.555378e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 12:\n", @@ -2065,17 +1985,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.556573E-20\n", - "Starting checks for alpha (primal) = 3.14e-02\n", + "minimal step size ALPHA_MIN = 9.422692E-25\n", + "Starting checks for alpha (primal) = 6.15e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 3.464e-17 (tol = 1.000e-07) r (rel) = 3.497e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.189e-17 (tol = 1.000e-07) r (rel) = 5.238e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 3.135683e-02:\n", - " New values of barrier function = 1.4019061379382336e-04 (reference 1.4020969715885140e-04):\n", - " New values of constraint violation = 3.8678912503420371e-16 (reference 1.0331091451071250e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 6.153251e-01:\n", + " New values of barrier function = 2.8910241919670053e-04 (reference 2.8998189588579787e-04):\n", + " New values of constraint violation = 1.6919821934583551e-16 (reference 1.1614007552982614e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -2088,11 +2008,11 @@ "*** Update HessianMatrix for Iteration 13:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 2.907881e-02\n", + "In limited-memory update, s_new_max is 2.032926e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 2.740668e-09 snrm = 1.005609e-01 ynrm = 7.583800e-08\n", + " s^Ty = 1.533249e-08 snrm = 1.134755e+00 ynrm = 3.087990e-07\n", " Perform the update.\n", - "sigma (for B0) is 2.710180e-07\n", + "sigma (for B0) is 1.190716e-08\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -2101,33 +2021,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 13 3.7840076e-05 0.00e+00 3.54e-07 -8.2 9.27e-01 - 9.52e-01 3.14e-02f 1 sigma=9.14e-01 qf=12y A\n", + " 13 7.6043466e-05 0.00e+00 1.15e-07 -8.0 3.30e-01 - 1.00e+00 6.15e-01f 1 Fy A\n", "\n", "**************************************************\n", "*** Beginning Iteration 13 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 6.2930331650267817e-09\n", - "Current fraction-to-the-boundary parameter tau = 9.9999926752813784e-01\n", + "Current barrier parameter mu = 1.0234391828744293e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999998976560822e-01\n", "\n", - "||curr_x||_inf = 9.9999998870057361e-01\n", - "||curr_s||_inf = 3.1223051299043446e-05\n", + "||curr_x||_inf = 1.0000000086814673e+00\n", + "||curr_s||_inf = 1.3346289688018320e-05\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 2.2983971872904288e-04\n", - "||curr_z_L||_inf = 3.3881184349002631e-08\n", - "||curr_z_U||_inf = 3.9700808446844428e-07\n", - "||curr_v_L||_inf = 2.2983794081283382e-04\n", + "||curr_y_d||_inf = 7.4633856811076615e-04\n", + "||curr_z_L||_inf = 1.4238079548713705e-07\n", + "||curr_z_U||_inf = 8.9620574544999946e-07\n", + "||curr_v_L||_inf = 7.4634058395468936e-04\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 13:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 3.7840076303503619e-05 3.7840076303503619e-05\n", - "Dual infeasibility......: 3.5395331122452642e-07 3.5395331122452642e-07\n", + "Objective...............: 7.6043465991924942e-05 7.6043465991924942e-05\n", + "Dual infeasibility......: 1.1512661096963838e-07 1.1512661096963838e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 3.3989971175832244e-08 3.3989971175832244e-08\n", - "Overall NLP error.......: 3.5395331122452642e-07 3.5395331122452642e-07\n", + "Complementarity.........: 2.8944963897623939e-08 2.8944963897623939e-08\n", + "Overall NLP error.......: 1.1512661096963838e-07 1.1512661096963838e-07\n", "\n", "\n", "\n", @@ -2136,9 +2056,13 @@ "*** Update Barrier Parameter for Iteration 13:\n", "**************************************************\n", "\n", - "Staying in free mu mode.\n", - "The current filter has 4 entries.\n", - "Solving the Primal Dual System for the affine step\n", + "Remaining in fixed mu mode.\n", + "Barrier Parameter: 1.023439e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 13:\n", + "**************************************************\n", + "\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -2147,40 +2071,28 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "Solving the Primal Dual System for the centering step\n", - "Factorization successful.\n", - "Sigma = 6.966207e-02\n", - "Barrier parameter mu computed by oracle is 4.420892e-10\n", - "Barrier parameter mu after safeguards is 4.420892e-10\n", - "Barrier Parameter: 4.420892e-10\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 13:\n", - "**************************************************\n", - "\n", - "residual_ratio = 9.454483e-16\n", + "residual_ratio = 1.769512e-14\n", "Factorization successful.\n", - "residual_ratio = 3.860887e-17\n", + "residual_ratio = 4.796655e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 13:\n", "**************************************************\n", "\n", "--> Starting line search in iteration 13 <--\n", - "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 3.650622E-21\n", - "Starting checks for alpha (primal) = 9.18e-01\n", + "minimal step size ALPHA_MIN = 8.459886E-26\n", + "Starting checks for alpha (primal) = 9.00e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 3.495e-17 (tol = 1.000e-07) r (rel) = 3.528e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.867e-17 (tol = 1.000e-07) r (rel) = 4.913e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 9.183855e-01:\n", - " New values of barrier function = 3.8793514754253226e-05 (reference 4.5030260798684637e-05):\n", - " New values of constraint violation = 4.3090598905899669e-16 (reference 3.8678912503420371e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 8.995352e-01:\n", + " New values of barrier function = 2.8817205501058079e-04 (reference 2.8910241919670053e-04):\n", + " New values of constraint violation = 1.7701294531720368e-16 (reference 1.6919821934583551e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -2193,11 +2105,11 @@ "*** Update HessianMatrix for Iteration 14:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 2.188907e-01\n", + "In limited-memory update, s_new_max is 4.781683e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 2.672397e-06 snrm = 5.620350e+00 ynrm = 1.137752e-06\n", + " s^Ty = 1.111931e-06 snrm = 4.565202e+00 ynrm = 1.234560e-06\n", " Perform the update.\n", - "sigma (for B0) is 8.460077e-08\n", + "sigma (for B0) is 5.335290e-08\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -2206,33 +2118,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 14 3.1076341e-05 0.00e+00 3.56e-07 -9.4 2.38e-01 - 7.07e-01 9.18e-01f 1 sigma=6.97e-02 qf=12y A\n", + " 14 7.7168866e-05 0.00e+00 4.03e-07 -8.0 5.32e-01 - 9.27e-01 9.00e-01f 1 Fy A\n", "\n", "**************************************************\n", "*** Beginning Iteration 14 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 4.4208922200447680e-10\n", - "Current fraction-to-the-boundary parameter tau = 9.9999964604668878e-01\n", + "Current barrier parameter mu = 1.0234391828744293e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999998976560822e-01\n", "\n", - "||curr_x||_inf = 9.9999999092766290e-01\n", - "||curr_s||_inf = 1.2808792683518393e-05\n", + "||curr_x||_inf = 9.8845041864171557e-01\n", + "||curr_s||_inf = 1.4221032779714803e-05\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 1.8163747627221880e-04\n", - "||curr_z_L||_inf = 1.9104449987264878e-08\n", - "||curr_z_U||_inf = 5.4610873970380496e-07\n", - "||curr_v_L||_inf = 1.8163689478518615e-04\n", + "||curr_y_d||_inf = 7.1441837810079566e-04\n", + "||curr_z_L||_inf = 8.4276522268695759e-08\n", + "||curr_z_U||_inf = 8.7313268274591788e-07\n", + "||curr_v_L||_inf = 7.1441846827555665e-04\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 14:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 3.1076340664289982e-05 3.1076340664289982e-05\n", - "Dual infeasibility......: 3.5598833436017428e-07 3.5598833436017428e-07\n", + "Objective...............: 7.7168866459873133e-05 7.7168866459873133e-05\n", + "Dual infeasibility......: 4.0281593147673594e-07 4.0281593147673594e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 1.0664751264724913e-08 1.0664751264724913e-08\n", - "Overall NLP error.......: 3.5598833436017428e-07 3.5598833436017428e-07\n", + "Complementarity.........: 1.6218304342515872e-08 1.6218304342515872e-08\n", + "Overall NLP error.......: 4.0281593147673594e-07 4.0281593147673594e-07\n", "\n", "\n", "\n", @@ -2241,9 +2153,13 @@ "*** Update Barrier Parameter for Iteration 14:\n", "**************************************************\n", "\n", - "Staying in free mu mode.\n", - "The current filter has 5 entries.\n", - "Solving the Primal Dual System for the affine step\n", + "Remaining in fixed mu mode.\n", + "Barrier Parameter: 1.023439e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 14:\n", + "**************************************************\n", + "\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -2252,40 +2168,28 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "Solving the Primal Dual System for the centering step\n", - "Factorization successful.\n", - "Sigma = 1.000000e+00\n", - "Barrier parameter mu computed by oracle is 2.164006e-09\n", - "Barrier parameter mu after safeguards is 2.164006e-09\n", - "Barrier Parameter: 2.164006e-09\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 14:\n", - "**************************************************\n", - "\n", - "residual_ratio = 1.661119e-15\n", + "residual_ratio = 1.875747e-15\n", "Factorization successful.\n", - "residual_ratio = 1.552665e-17\n", + "residual_ratio = 3.126641e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 14:\n", "**************************************************\n", "\n", "--> Starting line search in iteration 14 <--\n", - "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.267328E-22\n", - "Starting checks for alpha (primal) = 3.30e-02\n", + "minimal step size ALPHA_MIN = 1.911216E-25\n", + "Starting checks for alpha (primal) = 9.70e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 3.445e-17 (tol = 1.000e-07) r (rel) = 3.477e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.174e-17 (tol = 1.000e-07) r (rel) = 5.222e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 3.295347e-02:\n", - " New values of barrier function = 6.8743544301378399e-05 (reference 6.8851540453917292e-05):\n", - " New values of constraint violation = 3.2368008826607481e-16 (reference 4.3090598905899669e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 9.703047e-01:\n", + " New values of barrier function = 2.8772620146418591e-04 (reference 2.8817205501058079e-04):\n", + " New values of constraint violation = 1.6224238478298320e-16 (reference 1.7701294531720368e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -2298,11 +2202,11 @@ "*** Update HessianMatrix for Iteration 15:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 4.733475e-02\n", + "In limited-memory update, s_new_max is 3.931369e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 1.582016e-09 snrm = 1.971014e-01 ynrm = 5.017688e-08\n", + " s^Ty = 6.936403e-07 snrm = 1.195057e+00 ynrm = 8.699207e-07\n", " Perform the update.\n", - "sigma (for B0) is 4.072221e-08\n", + "sigma (for B0) is 4.856873e-07\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -2311,33 +2215,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 15 3.0906703e-05 0.00e+00 3.20e-07 -8.7 1.44e+00 - 6.69e-01 3.30e-02f 1 sigma=1.00e+00 qf=12y A\n", + " 15 7.6969579e-05 0.00e+00 9.18e-08 -8.0 4.05e-01 - 1.00e+00 9.70e-01f 1 Fy A\n", "\n", "**************************************************\n", "*** Beginning Iteration 15 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 2.1640056957869612e-09\n", - "Current fraction-to-the-boundary parameter tau = 9.9999964401166563e-01\n", + "Current barrier parameter mu = 1.0234391828744293e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999998976560822e-01\n", "\n", - "||curr_x||_inf = 9.9999999314937627e-01\n", - "||curr_s||_inf = 1.2827375571853822e-05\n", + "||curr_x||_inf = 1.0000000067049402e+00\n", + "||curr_s||_inf = 1.4230289881562869e-05\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 1.6771838170555599e-04\n", - "||curr_z_L||_inf = 2.0090482403092866e-08\n", - "||curr_z_U||_inf = 4.0346990677727674e-07\n", - "||curr_v_L||_inf = 1.6771662064446498e-04\n", + "||curr_y_d||_inf = 7.1868097674196397e-04\n", + "||curr_z_L||_inf = 8.5346878920991984e-08\n", + "||curr_z_U||_inf = 8.5956714608447506e-07\n", + "||curr_v_L||_inf = 7.1868121897604637e-04\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 15:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 3.0906703339077136e-05 3.0906703339077136e-05\n", - "Dual infeasibility......: 3.1980787284086894e-07 3.1980787284086894e-07\n", + "Objective...............: 7.6969579449427095e-05 7.6969579449427095e-05\n", + "Dual infeasibility......: 9.1818739185944381e-08 9.1818739185944381e-08\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 1.2265452277247657e-08 1.2265452277247657e-08\n", - "Overall NLP error.......: 3.1980787284086894e-07 3.1980787284086894e-07\n", + "Complementarity.........: 1.3964297591438233e-08 1.3964297591438233e-08\n", + "Overall NLP error.......: 9.1818739185944381e-08 9.1818739185944381e-08\n", "\n", "\n", "\n", @@ -2346,9 +2250,14 @@ "*** Update Barrier Parameter for Iteration 15:\n", "**************************************************\n", "\n", - "Staying in free mu mode.\n", - "The current filter has 4 entries.\n", - "Solving the Primal Dual System for the affine step\n", + "Remaining in fixed mu mode.\n", + "Reducing mu to 9.0909090909090910e-10 in fixed mu mode. Tau becomes 9.9999998976560822e-01\n", + "Barrier Parameter: 9.090909e-10\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 15:\n", + "**************************************************\n", + "\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -2357,20 +2266,9 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "Solving the Primal Dual System for the centering step\n", - "Factorization successful.\n", - "Sigma = 1.000000e+00\n", - "Barrier parameter mu computed by oracle is 2.241880e-09\n", - "Barrier parameter mu after safeguards is 2.241880e-09\n", - "Barrier Parameter: 2.241880e-09\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 15:\n", - "**************************************************\n", - "\n", - "residual_ratio = 1.458171e-15\n", + "residual_ratio = 4.028402e-15\n", "Factorization successful.\n", - "residual_ratio = 1.042701e-16\n", + "residual_ratio = 5.969527e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 15:\n", @@ -2380,17 +2278,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.658530E-23\n", - "Starting checks for alpha (primal) = 5.84e-01\n", + "minimal step size ALPHA_MIN = 2.066631E-23\n", + "Starting checks for alpha (primal) = 7.03e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 3.651e-17 (tol = 1.000e-07) r (rel) = 3.685e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.681e-17 (tol = 1.000e-07) r (rel) = 5.735e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 5.837277e-01:\n", - " New values of barrier function = 6.8497431208030537e-05 (reference 7.0105147191734408e-05):\n", - " New values of constraint violation = 4.4002853390092550e-16 (reference 3.2368008826607481e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 7.028143e-01:\n", + " New values of barrier function = 8.4821763779407538e-05 (reference 9.5690469983069498e-05):\n", + " New values of constraint violation = 1.1052594115542463e-16 (reference 1.6224238478298320e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -2399,74 +2297,179 @@ "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 16:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 7.126767e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 4.002777e-06 snrm = 3.824446e+00 ynrm = 2.041555e-06\n", + " Perform the update.\n", + "sigma (for B0) is 2.736682e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", "\n", "**************************************************\n", "*** Summary of Iteration: 16:\n", "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 16 2.8155857e-05 0.00e+00 2.59e-07 -8.6 5.78e-01 - 8.73e-01 5.84e-01f 1 sigma=1.00e+00 qf=12y A\n", + " 16 6.4483124e-05 0.00e+00 1.61e-07 -9.0 1.01e+00 - 1.95e-01 7.03e-01f 1 Fy A\n", "\n", "**************************************************\n", "*** Beginning Iteration 16 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 2.2418799668728096e-09\n", - "Current fraction-to-the-boundary parameter tau = 9.9999968019212715e-01\n", + "Current barrier parameter mu = 9.0909090909090910e-10\n", + "Current fraction-to-the-boundary parameter tau = 9.9999998976560822e-01\n", "\n", - "||curr_x||_inf = 9.9999999423380315e-01\n", - "||curr_s||_inf = 1.3912584531265601e-05\n", + "||curr_x||_inf = 1.0000000085928633e+00\n", + "||curr_s||_inf = 1.0565045421518312e-05\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 1.5255675550238714e-04\n", - "||curr_z_L||_inf = 1.8194101808679699e-08\n", - "||curr_z_U||_inf = 5.6874469415335646e-07\n", - "||curr_v_L||_inf = 1.5255679148921584e-04\n", + "||curr_y_d||_inf = 6.4217500291686666e-04\n", + "||curr_z_L||_inf = 7.7050588706666413e-08\n", + "||curr_z_U||_inf = 8.1160375126493416e-07\n", + "||curr_v_L||_inf = 6.4218310874448699e-04\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 16:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 2.8155857302499072e-05 2.8155857302499072e-05\n", - "Dual infeasibility......: 2.5902092059556247e-07 2.5902092059556247e-07\n", + "Objective...............: 6.4483124310314279e-05 6.4483124310314279e-05\n", + "Dual infeasibility......: 1.6079367566407334e-07 1.6079367566407334e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 2.4338671833101106e-08 2.4338671833101106e-08\n", - "Overall NLP error.......: 2.5902092059556247e-07 2.5902092059556247e-07\n", + "Complementarity.........: 1.1137172810707333e-08 1.1137172810707333e-08\n", + "Overall NLP error.......: 1.6079367566407334e-07 1.6079367566407334e-07\n", "\n", "\n", "\n", "\n", - "Number of Iterations....: 16\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 16:\n", + "**************************************************\n", "\n", - " (scaled) (unscaled)\n", - "Objective...............: 2.8155857302499072e-05 2.8155857302499072e-05\n", - "Dual infeasibility......: 2.5902092059556247e-07 2.5902092059556247e-07\n", - "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 2.4338671833101106e-08 2.4338671833101106e-08\n", - "Overall NLP error.......: 2.5902092059556247e-07 2.5902092059556247e-07\n", + "Switching back to free mu mode.\n", + "The current filter has 3 entries.\n", + "Solving the Primal Dual System for the affine step\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 1.258987e-03\n", + "Barrier parameter mu computed by oracle is 1.000000e-11\n", + "Barrier parameter mu after safeguards is 1.000000e-11\n", + "Barrier Parameter: 1.000000e-11\n", "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 16:\n", + "**************************************************\n", "\n", - "Number of objective function evaluations = 17\n", - "Number of objective gradient evaluations = 17\n", - "Number of equality constraint evaluations = 0\n", - "Number of inequality constraint evaluations = 17\n", - "Number of equality constraint Jacobian evaluations = 0\n", - "Number of inequality constraint Jacobian evaluations = 17\n", - "Number of Lagrangian Hessian evaluations = 0\n", - "Total CPU secs in IPOPT (w/o function evaluations) = 2.123\n", - "Total CPU secs in NLP function evaluations = 1.853\n", + "residual_ratio = 6.211181e-16\n", + "Factorization successful.\n", + "residual_ratio = 1.501329e-18\n", "\n", - "EXIT: Solved To Acceptable Level.\n", - "v=0.5\n", - "w=0.5\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "rm: cannot remove '/home/fenics/shared/templates/RES_OPT/TEMP*': No such file or directory\n", - "WARNING: The requested image's platform (linux/amd64) does not match the detected host platform (linux/arm64/v8) and no specific platform was requested\n" + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 16:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 16 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 4.499179E-21\n", + "Starting checks for alpha (primal) = 3.19e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.693e-17 (tol = 1.000e-07) r (rel) = 5.746e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 3.186528e-01:\n", + " New values of barrier function = 6.1269472500741982e-05 (reference 6.4706849344474309e-05):\n", + " New values of constraint violation = 1.5930995671958881e-16 (reference 1.1052594115542463e-16):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 17:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 17 6.1037878e-05 0.00e+00 1.08e-07 -11.0 6.50e-01 - 3.23e-01 3.19e-01f 1 sigma=1.26e-03 qf=13y A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 17 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 9.9999999999999994e-12\n", + "Current fraction-to-the-boundary parameter tau = 9.9999983920632429e-01\n", + "\n", + "||curr_x||_inf = 1.0000000024193492e+00\n", + "||curr_s||_inf = 9.2805963816738961e-06\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 5.1396775197256185e-04\n", + "||curr_z_L||_inf = 5.8495053494723876e-08\n", + "||curr_z_U||_inf = 7.4569014919371131e-07\n", + "||curr_v_L||_inf = 5.1397334794937920e-04\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 17:\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 6.1037878106871076e-05 6.1037878106871076e-05\n", + "Dual infeasibility......: 1.0777843626581685e-07 1.0777843626581685e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 7.4709130714281063e-09 7.4709130714281063e-09\n", + "Overall NLP error.......: 1.0777843626581685e-07 1.0777843626581685e-07\n", + "\n", + "\n", + "\n", + "\n", + "Number of Iterations....: 17\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 6.1037878106871076e-05 6.1037878106871076e-05\n", + "Dual infeasibility......: 1.0777843626581685e-07 1.0777843626581685e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 7.4709130714281063e-09 7.4709130714281063e-09\n", + "Overall NLP error.......: 1.0777843626581685e-07 1.0777843626581685e-07\n", + "\n", + "\n", + "Number of objective function evaluations = 18\n", + "Number of objective gradient evaluations = 18\n", + "Number of equality constraint evaluations = 0\n", + "Number of inequality constraint evaluations = 18\n", + "Number of equality constraint Jacobian evaluations = 0\n", + "Number of inequality constraint Jacobian evaluations = 18\n", + "Number of Lagrangian Hessian evaluations = 0\n", + "Total CPU secs in IPOPT (w/o function evaluations) = 1.921\n", + "Total CPU secs in NLP function evaluations = 1.984\n", + "\n", + "EXIT: Solved To Acceptable Level.\n", + "v=0.3\n", + "w=0.5\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "rm: cannot remove '/home/fenics/shared/templates/RES_OPT/TEMP*': No such file or directory\n", + "WARNING: The requested image's platform (linux/amd64) does not match the detected host platform (linux/arm64/v8) and no specific platform was requested\n" ] }, { @@ -2479,7 +2482,7 @@ "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 1.748e-16 (tol = 1.000e-07) r (rel) = 1.764e-12 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 2.744e-16 (tol = 1.000e-07) r (rel) = 2.769e-12 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", @@ -2525,7 +2528,7 @@ "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", - "Least square estimates max(y_c) = 0.000000e+00, max(y_d) = 1.002046e+00\n", + "Least square estimates max(y_c) = 0.000000e+00, max(y_d) = 1.000851e+00\n", "Total number of variables............................: 10201\n", " variables with only lower bounds: 0\n", " variables with lower and upper bounds: 10201\n", @@ -2540,7 +2543,7 @@ "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.589e-17 (tol = 1.000e-07) r (rel) = 4.631e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.210e-17 (tol = 1.000e-07) r (rel) = 5.258e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "\n", @@ -2556,7 +2559,7 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 0 4.8704052e-04 0.00e+00 2.22e+00 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", + " 0 6.0632598e-04 0.00e+00 3.64e-01 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", "\n", "**************************************************\n", "*** Beginning Iteration 0 from the following point:\n", @@ -2565,10 +2568,10 @@ "Current barrier parameter mu = 1.0000000000000000e+00\n", "Current fraction-to-the-boundary parameter tau = 0.0000000000000000e+00\n", "\n", - "||curr_x||_inf = 6.2118917703628540e-01\n", - "||curr_s||_inf = 2.4599847527097249e-01\n", + "||curr_x||_inf = 6.0563534498214722e-01\n", + "||curr_s||_inf = 2.5885384523204169e-01\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 1.0020464465022398e+00\n", + "||curr_y_d||_inf = 1.0008507486521290e+00\n", "||curr_z_L||_inf = 1.0000000000000000e+00\n", "||curr_z_U||_inf = 1.0000000000000000e+00\n", "||curr_v_L||_inf = 1.0000000000000000e+00\n", @@ -2578,11 +2581,11 @@ "***Current NLP Values for Iteration 0:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 4.8704052360352306e-04 4.8704052360352306e-04\n", - "Dual infeasibility......: 2.2174831958962931e+00 2.2174831958962931e+00\n", + "Objective...............: 6.0632598111746430e-04 6.0632598111746430e-04\n", + "Dual infeasibility......: 3.6396912861514852e-01 3.6396912861514852e-01\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", "Complementarity.........: 9.9000001999999998e-01 9.9000001999999998e-01\n", - "Overall NLP error.......: 2.2174831958962931e+00 2.2174831958962931e+00\n", + "Overall NLP error.......: 9.9000001999999998e-01 9.9000001999999998e-01\n", "\n", "\n", "\n", @@ -2591,7 +2594,7 @@ "*** Update Barrier Parameter for Iteration 0:\n", "**************************************************\n", "\n", - "Setting mu_max to 4.999876e+02.\n", + "Setting mu_max to 4.999882e+02.\n", "Staying in free mu mode.\n", "The current filter has 1 entries.\n", "Solving the Primal Dual System for the affine step\n", @@ -2603,18 +2606,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 3.360003e-01\n", - "Barrier parameter mu computed by oracle is 1.679960e-01\n", - "Barrier parameter mu after safeguards is 1.679960e-01\n", - "Barrier Parameter: 1.679960e-01\n", + "Sigma = 2.467321e-01\n", + "Barrier parameter mu computed by oracle is 1.233631e-01\n", + "Barrier parameter mu after safeguards is 1.233631e-01\n", + "Barrier Parameter: 1.233631e-01\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 0:\n", "**************************************************\n", "\n", - "residual_ratio = 2.086259e-14\n", + "residual_ratio = 5.175568e-15\n", "Factorization successful.\n", - "residual_ratio = 9.074391e-17\n", + "residual_ratio = 5.974682e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 0:\n", @@ -2631,11 +2634,11 @@ "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.670e-17 (tol = 1.000e-07) r (rel) = 4.713e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.832e-17 (tol = 1.000e-07) r (rel) = 4.876e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", - " New values of barrier function = 3.0260028813960107e+03 (reference 5.8473694132959017e+03):\n", - " New values of constraint violation = 3.3306690738754696e-16 (reference 0.0000000000000000e+00):\n", + " New values of barrier function = 2.4404794996445057e+03 (reference 4.4082351880547922e+03):\n", + " New values of constraint violation = 1.2212453270876722e-15 (reference 0.0000000000000000e+00):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -2650,11 +2653,11 @@ "*** Update HessianMatrix for Iteration 1:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 1.847599e-01\n", + "In limited-memory update, s_new_max is 1.232789e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 3.600166e+00 snrm = 1.343508e+01 ynrm = 3.386928e+00\n", + " s^Ty = 1.126891e+00 snrm = 1.004418e+01 ynrm = 9.364628e-01\n", " Perform the update.\n", - "sigma (for B0) is 1.994537e-02\n", + "sigma (for B0) is 1.117000e-02\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -2663,33 +2666,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 1 4.6302980e-04 0.00e+00 7.52e-01 -0.8 1.85e-01 - 3.70e-01 1.00e+00f 1 sigma=3.36e-01 qf=12y \n", + " 1 5.7809828e-04 0.00e+00 2.40e-01 -0.9 1.23e-01 - 9.98e-01 1.00e+00f 1 sigma=2.47e-01 qf=12y \n", "\n", "**************************************************\n", "*** Beginning Iteration 1 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 1.6799595239964976e-01\n", + "Current barrier parameter mu = 1.2336314121436313e-01\n", "Current fraction-to-the-boundary parameter tau = 9.8999999999999999e-01\n", "\n", - "||curr_x||_inf = 5.8821626494252255e-01\n", - "||curr_s||_inf = 1.3258421981061613e-01\n", + "||curr_x||_inf = 5.8458155162310477e-01\n", + "||curr_s||_inf = 1.7242998671615944e-01\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 1.0943399887739254e+00\n", - "||curr_z_L||_inf = 8.1412312750053450e-01\n", - "||curr_z_U||_inf = 7.6194608145019460e-01\n", - "||curr_v_L||_inf = 1.0532476431928279e+00\n", + "||curr_y_d||_inf = 8.9607702691559810e-01\n", + "||curr_z_L||_inf = 3.6632804485714987e-01\n", + "||curr_z_U||_inf = 2.6388465773350467e-01\n", + "||curr_v_L||_inf = 8.1074697870877988e-01\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 1:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 4.6302980268635125e-04 4.6302980268635125e-04\n", - "Dual infeasibility......: 7.5201101712860219e-01 7.5201101712860219e-01\n", + "Objective...............: 5.7809828387403938e-04 5.7809828387403938e-04\n", + "Dual infeasibility......: 2.4041630124405344e-01 2.4041630124405344e-01\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 6.2335954780289182e-01 6.2335954780289182e-01\n", - "Overall NLP error.......: 7.5201101712860219e-01 7.5201101712860219e-01\n", + "Complementarity.........: 2.1696344889658170e-01 2.1696344889658170e-01\n", + "Overall NLP error.......: 2.4041630124405344e-01 2.4041630124405344e-01\n", "\n", "\n", "\n", @@ -2712,17 +2715,17 @@ "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", "Sigma = 1.000000e-06\n", - "Barrier parameter mu computed by oracle is 3.803892e-07\n", - "Barrier parameter mu after safeguards is 3.803892e-07\n", - "Barrier Parameter: 3.803892e-07\n", + "Barrier parameter mu computed by oracle is 1.288476e-07\n", + "Barrier parameter mu after safeguards is 1.288476e-07\n", + "Barrier Parameter: 1.288476e-07\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 1:\n", "**************************************************\n", "\n", - "residual_ratio = 1.803925e-16\n", + "residual_ratio = 5.505580e-16\n", "Factorization successful.\n", - "residual_ratio = 1.202617e-16\n", + "residual_ratio = 9.768880e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 1:\n", @@ -2731,17 +2734,17 @@ "--> Starting line search in iteration 1 <--\n", "Mu has changed in line search - resetting watchdog counters.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 2.776572E-17\n", + "minimal step size ALPHA_MIN = 3.678597E-17\n", "Starting checks for alpha (primal) = 1.00e+00\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.416e-17 (tol = 1.000e-07) r (rel) = 4.457e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.074e-17 (tol = 1.000e-07) r (rel) = 5.120e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", - " New values of barrier function = 7.3147281184980477e-03 (reference 7.3147341159054812e-03):\n", - " New values of constraint violation = 3.3306690738754696e-16 (reference 3.3306690738754696e-16):\n", + " New values of barrier function = 3.1270583892432208e-03 (reference 3.1270749789700228e-03):\n", + " New values of constraint violation = 5.5511151231257827e-17 (reference 1.2212453270876722e-15):\n", "Checking sufficient reduction...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -2754,11 +2757,11 @@ "*** Update HessianMatrix for Iteration 2:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 4.170181e-06\n", + "In limited-memory update, s_new_max is 4.540443e-05\n", "Limited-Memory test for skipping:\n", - " s^Ty = 1.487384e-14 snrm = 3.732204e-05 ynrm = 6.853994e-10\n", + " s^Ty = 1.649985e-12 snrm = 1.231477e-04 ynrm = 1.939551e-08\n", " Perform the update.\n", - "sigma (for B0) is 1.067806e-05\n", + "sigma (for B0) is 1.087996e-04\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -2767,33 +2770,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 2 4.6302759e-04 0.00e+00 7.52e-03 -6.4 4.17e-06 - 9.90e-01 1.00e+00h 1 Nhj sigma=1.00e-06 qf=13y \n", + " 2 5.7808343e-04 0.00e+00 2.42e-03 -6.9 4.54e-05 - 9.90e-01 1.00e+00h 1 Nhj sigma=1.00e-06 qf=13y \n", "\n", "**************************************************\n", "*** Beginning Iteration 2 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 3.8038918431582506e-07\n", + "Current barrier parameter mu = 1.2884756725947034e-07\n", "Current fraction-to-the-boundary parameter tau = 9.8999999999999999e-01\n", "\n", - "||curr_x||_inf = 5.8821650715826079e-01\n", - "||curr_s||_inf = 1.3258387518455172e-01\n", + "||curr_x||_inf = 5.8458272581402138e-01\n", + "||curr_s||_inf = 1.7242934515766895e-01\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 1.0956029356307446e-02\n", - "||curr_z_L||_inf = 8.1472283403452028e-03\n", - "||curr_z_U||_inf = 7.6257446059045941e-03\n", - "||curr_v_L||_inf = 1.0544835756068016e-02\n", + "||curr_y_d||_inf = 9.0372120362592762e-03\n", + "||curr_z_L||_inf = 3.6932230748261685e-03\n", + "||curr_z_U||_inf = 2.6612188709363438e-03\n", + "||curr_v_L||_inf = 8.1769808436709113e-03\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 2:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 4.6302759046657865e-04 4.6302759046657865e-04\n", - "Dual infeasibility......: 7.5227639510524755e-03 7.5227639510524755e-03\n", + "Objective...............: 5.7808342830889677e-04 5.7808342830889677e-04\n", + "Dual infeasibility......: 2.4228214744788160e-03 2.4228214744788160e-03\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 6.2382984541305289e-03 6.2382984541305289e-03\n", - "Overall NLP error.......: 7.5227639510524755e-03 7.5227639510524755e-03\n", + "Complementarity.........: 2.1873527015684241e-03 2.1873527015684241e-03\n", + "Overall NLP error.......: 2.4228214744788160e-03 2.4228214744788160e-03\n", "\n", "\n", "\n", @@ -2802,8 +2805,8 @@ "*** Update Barrier Parameter for Iteration 2:\n", "**************************************************\n", "\n", - "Switching to fixed mu mode with mu = 3.0453769082853552e-03 and tau = 9.9695462309171468e-01.\n", - "Barrier Parameter: 3.045377e-03\n", + "Switching to fixed mu mode with mu = 1.0392368597111350e-03 and tau = 9.9896076314028881e-01.\n", + "Barrier Parameter: 1.039237e-03\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 2:\n", @@ -2817,9 +2820,9 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 1.109790e-12\n", + "residual_ratio = 2.723651e-15\n", "Factorization successful.\n", - "residual_ratio = 9.304854e-16\n", + "residual_ratio = 3.158429e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 2:\n", @@ -2828,17 +2831,17 @@ "--> Starting line search in iteration 2 <--\n", "Mu has changed in line search - resetting watchdog counters.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 7.063344E-27\n", - "Starting checks for alpha (primal) = 7.46e-01\n", + "minimal step size ALPHA_MIN = 2.673306E-27\n", + "Starting checks for alpha (primal) = 7.59e-02\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.392e-17 (tol = 1.000e-07) r (rel) = 4.432e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.125e-17 (tol = 1.000e-07) r (rel) = 5.172e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 7.463167e-01:\n", - " New values of barrier function = 4.4874493915599857e+01 (reference 5.4854836997459245e+01):\n", - " New values of constraint violation = 9.1593399531575415e-16 (reference 3.3306690738754696e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 7.585769e-02:\n", + " New values of barrier function = 1.9832350390496664e+01 (reference 2.0559668101396795e+01):\n", + " New values of constraint violation = 8.3266726846886741e-17 (reference 5.5511151231257827e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -2851,9 +2854,9 @@ "*** Update HessianMatrix for Iteration 3:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 3.450118e-01\n", + "In limited-memory update, s_new_max is 1.331403e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = -1.481102e-04 snrm = 1.559662e+01 ynrm = 3.828571e-05\n", + " s^Ty = -1.451859e-06 snrm = 1.439043e+00 ynrm = 2.731417e-06\n", " Skip the update.\n", "Number of successive iterations with skipping: 1\n", "\n", @@ -2863,33 +2866,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 3 3.1715370e-04 0.00e+00 7.23e-03 -2.5 4.62e-01 - 1.00e+00 7.46e-01f 1 Fy Ws\n", + " 3 5.6956893e-04 0.00e+00 3.64e-03 -3.0 1.76e+00 - 3.92e-01 7.59e-02f 1 Fy Ws\n", "\n", "**************************************************\n", "*** Beginning Iteration 3 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 3.0453769082853552e-03\n", - "Current fraction-to-the-boundary parameter tau = 9.9695462309171468e-01\n", + "Current barrier parameter mu = 1.0392368597111350e-03\n", + "Current fraction-to-the-boundary parameter tau = 9.9896076314028881e-01\n", "\n", - "||curr_x||_inf = 5.3977095189069146e-01\n", - "||curr_s||_inf = 4.0375790235178122e-04\n", + "||curr_x||_inf = 5.7941527882232635e-01\n", + "||curr_s||_inf = 1.6002931706506521e-01\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 3.7060376851556567e-02\n", - "||curr_z_L||_inf = 1.5394438715324700e-02\n", - "||curr_z_U||_inf = 8.1587009855454244e-03\n", - "||curr_v_L||_inf = 3.7055577302987197e-02\n", + "||curr_y_d||_inf = 1.0907352273538005e-02\n", + "||curr_z_L||_inf = 4.6100499956645852e-03\n", + "||curr_z_U||_inf = 2.4612646723549936e-03\n", + "||curr_v_L||_inf = 1.0371590958951080e-02\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 3:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 3.1715369999061113e-04 3.1715369999061113e-04\n", - "Dual infeasibility......: 7.2320863508398957e-03 7.2320863508398957e-03\n", + "Objective...............: 5.6956892860383251e-04 5.6956892860383251e-04\n", + "Dual infeasibility......: 3.6380072938206696e-03 3.6380072938206696e-03\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 8.3094709931381145e-03 8.3094709931381145e-03\n", - "Overall NLP error.......: 8.3094709931381145e-03 8.3094709931381145e-03\n", + "Complementarity.........: 1.8993069172678625e-03 1.8993069172678625e-03\n", + "Overall NLP error.......: 3.6380072938206696e-03 3.6380072938206696e-03\n", "\n", "\n", "\n", @@ -2899,7 +2902,7 @@ "**************************************************\n", "\n", "Switching back to free mu mode.\n", - "The current filter has 3 entries.\n", + "The current filter has 2 entries.\n", "Solving the Primal Dual System for the affine step\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", @@ -2911,18 +2914,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 1.000000e-06\n", - "Barrier parameter mu computed by oracle is 2.996411e-09\n", - "Barrier parameter mu after safeguards is 2.996411e-09\n", - "Barrier Parameter: 2.996411e-09\n", + "Sigma = 8.050304e-03\n", + "Barrier parameter mu computed by oracle is 9.380861e-06\n", + "Barrier parameter mu after safeguards is 9.380861e-06\n", + "Barrier Parameter: 9.380861e-06\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 3:\n", "**************************************************\n", "\n", - "residual_ratio = 1.949754e-16\n", + "residual_ratio = 1.253733e-14\n", "Factorization successful.\n", - "residual_ratio = 3.899507e-17\n", + "residual_ratio = 6.034818e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 3:\n", @@ -2931,17 +2934,17 @@ "--> Starting line search in iteration 3 <--\n", "Mu has changed in line search - resetting watchdog counters.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 4.558255E-18\n", + "minimal step size ALPHA_MIN = 4.029331E-23\n", "Starting checks for alpha (primal) = 1.00e+00\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.283e-17 (tol = 1.000e-07) r (rel) = 4.323e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.084e-17 (tol = 1.000e-07) r (rel) = 5.131e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", - " New values of barrier function = 3.6120591857759140e-04 (reference 3.6130635951824426e-04):\n", - " New values of constraint violation = 1.9130747333506726e-16 (reference 9.1593399531575415e-16):\n", + " New values of barrier function = 1.7870549619355253e-01 (reference 1.7958476082294542e-01):\n", + " New values of constraint violation = 2.7755575615628914e-17 (reference 8.3266726846886741e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -2954,12 +2957,11 @@ "*** Update HessianMatrix for Iteration 4:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 4.081172e-04\n", + "In limited-memory update, s_new_max is 4.857406e-03\n", "Limited-Memory test for skipping:\n", - " s^Ty = 5.237898e-11 snrm = 1.938419e-03 ynrm = 3.488610e-08\n", - " Perform the update.\n", - "sigma (for B0) is 1.393997e-05\n", - "Number of successive iterations with skipping: 0\n", + " s^Ty = -4.301490e-09 snrm = 1.990570e-01 ynrm = 2.238642e-06\n", + " Skip the update.\n", + "Number of successive iterations with skipping: 2\n", "\n", "\n", "**************************************************\n", @@ -2967,33 +2969,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 4 3.1705327e-04 0.00e+00 6.54e-05 -8.5 4.08e-04 - 9.91e-01 1.00e+00f 1 sigma=1.00e-06 qf=13y A\n", + " 4 5.6668739e-04 0.00e+00 2.77e-04 -5.0 4.86e-03 - 9.95e-01 1.00e+00f 1 sigma=8.05e-03 qf=13y AWs\n", "\n", "**************************************************\n", "*** Beginning Iteration 4 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 2.9964112874601644e-09\n", - "Current fraction-to-the-boundary parameter tau = 9.9169052900686183e-01\n", + "Current barrier parameter mu = 9.3808613325604514e-06\n", + "Current fraction-to-the-boundary parameter tau = 9.9636199270617931e-01\n", "\n", - "||curr_x||_inf = 5.3977579284508148e-01\n", - "||curr_s||_inf = 3.9420500698697812e-04\n", + "||curr_x||_inf = 5.7893703636065474e-01\n", + "||curr_s||_inf = 1.5826856453898955e-01\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 1.2124954685401744e-03\n", - "||curr_z_L||_inf = 1.3960018092507714e-04\n", - "||curr_z_U||_inf = 7.4145990765397962e-05\n", - "||curr_v_L||_inf = 1.2124520516770196e-03\n", + "||curr_y_d||_inf = 2.2516872520260452e-04\n", + "||curr_z_L||_inf = 2.8671434170452186e-04\n", + "||curr_z_U||_inf = 5.8751362109901641e-05\n", + "||curr_v_L||_inf = 2.2195881008565954e-04\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 4:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 3.1705326771341679e-04 3.1705326771341679e-04\n", - "Dual infeasibility......: 6.5387572558797468e-05 6.5387572558797468e-05\n", + "Objective...............: 5.6668738821619049e-04 5.6668738821619049e-04\n", + "Dual infeasibility......: 2.7734828196926051e-04 2.7734828196926051e-04\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 7.5352799736152159e-05 7.5352799736152159e-05\n", - "Overall NLP error.......: 7.5352799736152159e-05 7.5352799736152159e-05\n", + "Complementarity.........: 3.5129104478627629e-05 3.5129104478627629e-05\n", + "Overall NLP error.......: 2.7734828196926051e-04 2.7734828196926051e-04\n", "\n", "\n", "\n", @@ -3015,18 +3017,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 6.014650e-02\n", - "Barrier parameter mu computed by oracle is 1.636026e-06\n", - "Barrier parameter mu after safeguards is 1.636026e-06\n", - "Barrier Parameter: 1.636026e-06\n", + "Sigma = 1.363906e-01\n", + "Barrier parameter mu computed by oracle is 1.960047e-06\n", + "Barrier parameter mu after safeguards is 1.960047e-06\n", + "Barrier Parameter: 1.960047e-06\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 4:\n", "**************************************************\n", "\n", - "residual_ratio = 4.714859e-15\n", + "residual_ratio = 6.279930e-15\n", "Factorization successful.\n", - "residual_ratio = 3.258087e-16\n", + "residual_ratio = 2.166223e-16\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 4:\n", @@ -3036,29 +3038,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.766649E-21\n", - "Starting checks for alpha (primal) = 1.61e-01\n", - "No Jacobian form specified for nonlinear variational problem.\n", - "Differentiating residual form F to obtain Jacobian J = F'.\n", - "Solving nonlinear variational problem.\n", - " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.186e-17 (tol = 1.000e-07) r (rel) = 4.225e-13 (tol = 1.000e-09)\n", - " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 1.607539e-01:\n", - " New values of barrier function = 2.4429408667047930e-02 (reference 2.4424185715987642e-02):\n", - " New values of constraint violation = 1.6019087098473328e-16 (reference 1.9130747333506726e-16):\n", - "Checking Armijo Condition...\n", - "Failed...\n", - "Starting checks for alpha (primal) = 8.04e-02\n", + "minimal step size ALPHA_MIN = 4.082368E-24\n", + "Starting checks for alpha (primal) = 1.00e+00\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.186e-17 (tol = 1.000e-07) r (rel) = 4.225e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.468e-17 (tol = 1.000e-07) r (rel) = 4.509e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 8.037697e-02:\n", - " New values of barrier function = 2.4420157533596547e-02 (reference 2.4424185715987642e-02):\n", - " New values of constraint violation = 2.0350474777552918e-16 (reference 1.9130747333506726e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 3.4903048616193505e-02 (reference 3.7787202145653939e-02):\n", + " New values of constraint violation = 2.2204460492503131e-16 (reference 2.7755575615628914e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -3071,12 +3061,7 @@ "*** Update HessianMatrix for Iteration 5:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 2.668800e-03\n", - "Limited-Memory test for skipping:\n", - " s^Ty = 7.955388e-11 snrm = 3.118260e-02 ynrm = 2.347997e-07\n", - " Perform the update.\n", - "sigma (for B0) is 8.181570e-08\n", - "Number of successive iterations with skipping: 0\n", + "Resetting Limited-Memory Update.\n", "\n", "\n", "**************************************************\n", @@ -3084,33 +3069,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 5 3.1663452e-04 0.00e+00 3.68e-06 -5.8 3.32e-02 - 9.99e-01 8.04e-02f 2 sigma=6.01e-02 qf=12y A\n", + " 5 4.7462820e-04 0.00e+00 1.27e-04 -5.7 1.28e-01 - 9.93e-01 1.00e+00f 1 sigma=1.36e-01 qf=12y AWr\n", "\n", "**************************************************\n", "*** Beginning Iteration 5 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 1.6360259772062144e-06\n", - "Current fraction-to-the-boundary parameter tau = 9.9992464720026386e-01\n", + "Current barrier parameter mu = 1.9600472800264057e-06\n", + "Current fraction-to-the-boundary parameter tau = 9.9972265171803076e-01\n", "\n", - "||curr_x||_inf = 5.3877389584409507e-01\n", - "||curr_s||_inf = 1.9711235609572630e-04\n", + "||curr_x||_inf = 6.7970699058116035e-01\n", + "||curr_s||_inf = 1.2360980340370592e-01\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 1.1678020670160476e-02\n", - "||curr_z_L||_inf = 6.4301367485069725e-06\n", - "||curr_z_U||_inf = 7.8028122330760423e-06\n", - "||curr_v_L||_inf = 1.1678015555874600e-02\n", + "||curr_y_d||_inf = 6.2567581658318962e-05\n", + "||curr_z_L||_inf = 1.2891044391489718e-04\n", + "||curr_z_U||_inf = 2.1410950720915148e-05\n", + "||curr_v_L||_inf = 6.2171589741291498e-05\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 5:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 3.1663452449653268e-04 3.1663452449653268e-04\n", - "Dual infeasibility......: 3.6751614129219837e-06 3.6751614129219837e-06\n", + "Objective...............: 4.7462820104574017e-04 4.7462820104574017e-04\n", + "Dual infeasibility......: 1.2684663439650783e-04 1.2684663439650783e-04\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 3.6711435487567861e-06 3.6711435487567861e-06\n", - "Overall NLP error.......: 3.6751614129219837e-06 3.6751614129219837e-06\n", + "Complementarity.........: 7.6850186069327993e-06 7.6850186069327993e-06\n", + "Overall NLP error.......: 1.2684663439650783e-04 1.2684663439650783e-04\n", "\n", "\n", "\n", @@ -3125,25 +3110,23 @@ "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", - "Factorization successful.\n", - "Factorization successful.\n", "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 3.149303e-01\n", - "Barrier parameter mu computed by oracle is 5.259011e-07\n", - "Barrier parameter mu after safeguards is 5.259011e-07\n", - "Barrier Parameter: 5.259011e-07\n", + "Sigma = 4.997848e-06\n", + "Barrier parameter mu computed by oracle is 1.000000e-11\n", + "Barrier parameter mu after safeguards is 1.000000e-11\n", + "Barrier Parameter: 1.000000e-11\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 5:\n", "**************************************************\n", "\n", - "residual_ratio = 1.553300e-11\n", + "residual_ratio = 2.272951e-12\n", "Factorization successful.\n", - "residual_ratio = 2.322437e-16\n", + "residual_ratio = 1.324751e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 5:\n", @@ -3153,18 +3136,18 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.994040E-21\n", + "minimal step size ALPHA_MIN = 1.534259E-17\n", "Starting checks for alpha (primal) = 1.00e+00\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 3.224e-17 (tol = 1.000e-07) r (rel) = 3.254e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.538e-17 (tol = 1.000e-07) r (rel) = 4.580e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", - " New values of barrier function = 8.0257073397595642e-03 (reference 8.0647200524303787e-03):\n", - " New values of constraint violation = 1.2293497383270013e-16 (reference 2.0350474777552918e-16):\n", - "Checking Armijo Condition...\n", + " New values of barrier function = 4.7479661872267052e-04 (reference 4.7480385201480597e-04):\n", + " New values of constraint violation = 2.7755575615628914e-17 (reference 2.2204460492503131e-16):\n", + "Checking sufficient reduction...\n", "Succeeded...\n", "Checking filter acceptability...\n", "Succeeded...\n", @@ -3176,11 +3159,11 @@ "*** Update HessianMatrix for Iteration 6:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 4.056222e-01\n", + "In limited-memory update, s_new_max is 2.182327e-05\n", "Limited-Memory test for skipping:\n", - " s^Ty = 1.034491e-05 snrm = 1.877164e+00 ynrm = 1.846312e-05\n", + " s^Ty = 2.466370e-13 snrm = 8.506497e-05 ynrm = 4.248028e-09\n", " Perform the update.\n", - "sigma (for B0) is 2.935772e-06\n", + "sigma (for B0) is 3.408448e-05\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -3189,33 +3172,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 6 2.8884142e-04 0.00e+00 1.73e-06 -6.3 4.06e-01 - 9.96e-01 1.00e+00f 1 sigma=3.15e-01 qf=12y A\n", + " 6 4.7462097e-04 0.00e+00 2.18e-05 -11.0 2.18e-05 - 1.00e+00 1.00e+00h 1 sigma=5.00e-06 qf=13y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 6 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 5.2590109721842701e-07\n", - "Current fraction-to-the-boundary parameter tau = 9.9999632483858703e-01\n", + "Current barrier parameter mu = 9.9999999999999994e-12\n", + "Current fraction-to-the-boundary parameter tau = 9.9987315336560345e-01\n", "\n", - "||curr_x||_inf = 9.2410030777432806e-01\n", - "||curr_s||_inf = 1.1569533729284346e-04\n", + "||curr_x||_inf = 6.7972881385126871e-01\n", + "||curr_s||_inf = 1.2360938376226158e-01\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 7.5086990403476815e-03\n", - "||curr_z_L||_inf = 1.5505001896117894e-06\n", - "||curr_z_U||_inf = 7.3897621872819613e-06\n", - "||curr_v_L||_inf = 7.5086985370896699e-03\n", + "||curr_y_d||_inf = 4.2923113340762616e-07\n", + "||curr_z_L||_inf = 2.0347424935544204e-08\n", + "||curr_z_U||_inf = 4.7097993893681345e-09\n", + "||curr_v_L||_inf = 9.6418110211377375e-09\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 6:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 2.8884142183483510e-04 2.8884142183483510e-04\n", - "Dual infeasibility......: 1.7308522604732651e-06 1.7308522604732651e-06\n", + "Objective...............: 4.7462096777451974e-04 4.7462096777451974e-04\n", + "Dual infeasibility......: 2.1817850862022120e-05 2.1817850862022120e-05\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 1.1092600152643912e-06 1.1092600152643912e-06\n", - "Overall NLP error.......: 1.7308522604732651e-06 1.7308522604732651e-06\n", + "Complementarity.........: 1.5084130840534965e-09 1.5084130840534965e-09\n", + "Overall NLP error.......: 2.1817850862022120e-05 2.1817850862022120e-05\n", "\n", "\n", "\n", @@ -3227,8 +3210,8 @@ "Staying in free mu mode.\n", "The current filter has 2 entries.\n", "Solving the Primal Dual System for the affine step\n", - "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", - "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20726\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 245484\n", "Factorization successful.\n", "Factorization successful.\n", "Factorization successful.\n", @@ -3237,18 +3220,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 3.911818e-01\n", - "Barrier parameter mu computed by oracle is 2.075177e-07\n", - "Barrier parameter mu after safeguards is 2.075177e-07\n", - "Barrier Parameter: 2.075177e-07\n", + "Sigma = 1.804931e+00\n", + "Barrier parameter mu computed by oracle is 5.617646e-10\n", + "Barrier parameter mu after safeguards is 5.617646e-10\n", + "Barrier Parameter: 5.617646e-10\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 6:\n", "**************************************************\n", "\n", - "residual_ratio = 2.660681e-11\n", + "residual_ratio = 2.993105e-17\n", "Factorization successful.\n", - "residual_ratio = 1.203473e-15\n", + "residual_ratio = 9.977016e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 6:\n", @@ -3258,17 +3241,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.082418E-21\n", - "Starting checks for alpha (primal) = 1.00e+00\n", + "minimal step size ALPHA_MIN = 3.044534E-23\n", + "Starting checks for alpha (primal) = 5.51e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 2.701e-17 (tol = 1.000e-07) r (rel) = 2.726e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.652e-17 (tol = 1.000e-07) r (rel) = 4.695e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", - " New values of barrier function = 3.3095464179937330e-03 (reference 3.3417658050835710e-03):\n", - " New values of constraint violation = 1.1714804473705875e-16 (reference 1.2293497383270013e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 5.510821e-01:\n", + " New values of barrier function = 3.0578888845124653e-04 (reference 4.8448841639266702e-04):\n", + " New values of constraint violation = 2.2204460492503131e-16 (reference 2.7755575615628914e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -3281,11 +3264,11 @@ "*** Update HessianMatrix for Iteration 7:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 2.113738e-01\n", + "In limited-memory update, s_new_max is 3.832589e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 2.950344e-05 snrm = 3.095500e+00 ynrm = 1.474585e-05\n", + " s^Ty = 1.291217e-04 snrm = 3.299308e+00 ynrm = 4.959430e-05\n", " Perform the update.\n", - "sigma (for B0) is 3.079009e-06\n", + "sigma (for B0) is 1.186188e-05\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -3294,33 +3277,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 7 2.2593574e-04 0.00e+00 1.68e-06 -6.7 2.11e-01 - 9.96e-01 1.00e+00f 1 sigma=3.91e-01 qf=12y A\n", + " 7 2.9594486e-04 0.00e+00 4.13e-05 -9.3 6.95e-01 - 1.00e+00 5.51e-01f 1 sigma=1.80e+00 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 7 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 2.0751765636208095e-07\n", - "Current fraction-to-the-boundary parameter tau = 9.9999826914773948e-01\n", + "Current barrier parameter mu = 5.6176460884667023e-10\n", + "Current fraction-to-the-boundary parameter tau = 9.9997818214913803e-01\n", "\n", - "||curr_x||_inf = 9.5005893537771646e-01\n", - "||curr_s||_inf = 7.2930630089901880e-05\n", + "||curr_x||_inf = 9.9999164793212114e-01\n", + "||curr_s||_inf = 1.0557144843963795e-01\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 4.5812734757940694e-03\n", - "||curr_z_L||_inf = 9.4171124529123179e-07\n", - "||curr_z_U||_inf = 5.2706162071997211e-06\n", - "||curr_v_L||_inf = 4.5813200352051407e-03\n", + "||curr_y_d||_inf = 2.5119488663727922e-07\n", + "||curr_z_L||_inf = 5.9637636961660206e-08\n", + "||curr_z_U||_inf = 5.4688907266590252e-09\n", + "||curr_v_L||_inf = 7.0978335094288167e-09\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 7:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 2.2593574370060362e-04 2.2593574370060362e-04\n", - "Dual infeasibility......: 1.6818748693090317e-06 1.6818748693090317e-06\n", + "Objective...............: 2.9594486325079872e-04 2.9594486325079872e-04\n", + "Dual infeasibility......: 4.1319838059223542e-05 4.1319838059223542e-05\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 3.6654665539814339e-07 3.6654665539814339e-07\n", - "Overall NLP error.......: 1.6818748693090317e-06 1.6818748693090317e-06\n", + "Complementarity.........: 8.9169472495952544e-10 8.9169472495952544e-10\n", + "Overall NLP error.......: 4.1319838059223542e-05 4.1319838059223542e-05\n", "\n", "\n", "\n", @@ -3330,10 +3313,10 @@ "**************************************************\n", "\n", "Staying in free mu mode.\n", - "The current filter has 2 entries.\n", + "The current filter has 3 entries.\n", "Solving the Primal Dual System for the affine step\n", - "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", - "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20726\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 245484\n", "Factorization successful.\n", "Factorization successful.\n", "Factorization successful.\n", @@ -3342,18 +3325,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 4.383185e-01\n", - "Barrier parameter mu computed by oracle is 9.181682e-08\n", - "Barrier parameter mu after safeguards is 9.181682e-08\n", - "Barrier Parameter: 9.181682e-08\n", + "Sigma = 3.912269e+00\n", + "Barrier parameter mu computed by oracle is 2.211469e-09\n", + "Barrier parameter mu after safeguards is 2.211469e-09\n", + "Barrier Parameter: 2.211469e-09\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 7:\n", "**************************************************\n", "\n", - "residual_ratio = 2.266956e-10\n", + "residual_ratio = 5.384779e-17\n", "Factorization successful.\n", - "residual_ratio = 4.493891e-16\n", + "residual_ratio = 3.167517e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 7:\n", @@ -3363,17 +3346,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.466112E-21\n", - "Starting checks for alpha (primal) = 8.43e-01\n", + "minimal step size ALPHA_MIN = 2.132093E-22\n", + "Starting checks for alpha (primal) = 1.62e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 3.039e-17 (tol = 1.000e-07) r (rel) = 3.067e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.444e-17 (tol = 1.000e-07) r (rel) = 4.485e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 8.431664e-01:\n", - " New values of barrier function = 1.5661255970296102e-03 (reference 1.5902886476522439e-03):\n", - " New values of constraint violation = 2.1648806879104310e-16 (reference 1.1714804473705875e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 1.623140e-01:\n", + " New values of barrier function = 2.8367530917845736e-04 (reference 3.3469730900024359e-04):\n", + " New values of constraint violation = 9.7144514654701197e-17 (reference 2.2204460492503131e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -3386,11 +3369,11 @@ "*** Update HessianMatrix for Iteration 8:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 2.569186e-01\n", + "In limited-memory update, s_new_max is 3.555651e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 9.665151e-06 snrm = 2.720363e+00 ynrm = 8.300661e-06\n", + " s^Ty = 2.754160e-05 snrm = 1.245135e+00 ynrm = 4.810689e-05\n", " Perform the update.\n", - "sigma (for B0) is 1.306036e-06\n", + "sigma (for B0) is 1.776464e-05\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -3399,33 +3382,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 8 1.8201417e-04 0.00e+00 3.04e-06 -7.0 3.05e-01 - 9.93e-01 8.43e-01f 1 sigma=4.38e-01 qf=12y A\n", + " 8 2.4494997e-04 0.00e+00 9.13e-06 -8.7 2.19e+00 - 1.00e+00 1.62e-01f 1 sigma=3.91e+00 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 8 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 9.1816817031789006e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999831812513074e-01\n", + "Current barrier parameter mu = 2.2114685898303495e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9995868016194078e-01\n", "\n", - "||curr_x||_inf = 9.9999975562499555e-01\n", - "||curr_s||_inf = 5.3730159520642967e-05\n", + "||curr_x||_inf = 9.9998531749985653e-01\n", + "||curr_s||_inf = 9.8911705019217769e-02\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 2.7025554934629560e-03\n", - "||curr_z_L||_inf = 9.9375752145242135e-07\n", - "||curr_z_U||_inf = 5.9307961397274790e-06\n", - "||curr_v_L||_inf = 2.7025701585926547e-03\n", + "||curr_y_d||_inf = 2.2521759652297025e-07\n", + "||curr_z_L||_inf = 1.2557820156188865e-07\n", + "||curr_z_U||_inf = 6.4965559832340876e-06\n", + "||curr_v_L||_inf = 2.3706148404701627e-08\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 8:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.8201416983233039e-04 1.8201416983233039e-04\n", - "Dual infeasibility......: 3.0360774061120197e-06 3.0360774061120197e-06\n", + "Objective...............: 2.4494997476961729e-04 2.4494997476961729e-04\n", + "Dual infeasibility......: 9.1259992116974073e-06 9.1259992116974073e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 1.4790706677325193e-07 1.4790706677325193e-07\n", - "Overall NLP error.......: 3.0360774061120197e-06 3.0360774061120197e-06\n", + "Complementarity.........: 4.1598322979235200e-07 4.1598322979235200e-07\n", + "Overall NLP error.......: 9.1259992116974073e-06 9.1259992116974073e-06\n", "\n", "\n", "\n", @@ -3437,8 +3420,8 @@ "Staying in free mu mode.\n", "The current filter has 3 entries.\n", "Solving the Primal Dual System for the affine step\n", - "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", - "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20726\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 245484\n", "Factorization successful.\n", "Factorization successful.\n", "Factorization successful.\n", @@ -3447,18 +3430,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 5.015646e-01\n", - "Barrier parameter mu computed by oracle is 4.710586e-08\n", - "Barrier parameter mu after safeguards is 4.710586e-08\n", - "Barrier Parameter: 4.710586e-08\n", + "Sigma = 5.247048e-01\n", + "Barrier parameter mu computed by oracle is 1.168871e-09\n", + "Barrier parameter mu after safeguards is 1.168871e-09\n", + "Barrier Parameter: 1.168871e-09\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 8:\n", "**************************************************\n", "\n", - "residual_ratio = 8.822559e-16\n", + "residual_ratio = 1.205134e-16\n", "Factorization successful.\n", - "residual_ratio = 1.485100e-17\n", + "residual_ratio = 1.854053e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 8:\n", @@ -3468,17 +3451,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 4.205836E-23\n", - "Starting checks for alpha (primal) = 3.87e-01\n", + "minimal step size ALPHA_MIN = 3.189261E-22\n", + "Starting checks for alpha (primal) = 4.43e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 3.545e-17 (tol = 1.000e-07) r (rel) = 3.578e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.681e-17 (tol = 1.000e-07) r (rel) = 4.724e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 3.871788e-01:\n", - " New values of barrier function = 8.7178547165690443e-04 (reference 8.9212103618765590e-04):\n", - " New values of constraint violation = 4.1046538997585591e-16 (reference 2.1648806879104310e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 4.434101e-01:\n", + " New values of barrier function = 2.1015847292720386e-04 (reference 2.6541824168166081e-04):\n", + " New values of constraint violation = 0.0000000000000000e+00 (reference 9.7144514654701197e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -3491,11 +3474,11 @@ "*** Update HessianMatrix for Iteration 9:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 2.293077e-01\n", + "In limited-memory update, s_new_max is 1.659446e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 4.780517e-06 snrm = 2.350285e+00 ynrm = 4.877235e-06\n", + " s^Ty = 1.746430e-05 snrm = 1.964375e+00 ynrm = 1.020056e-05\n", " Perform the update.\n", - "sigma (for B0) is 8.654337e-07\n", + "sigma (for B0) is 4.525873e-06\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -3504,33 +3487,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 9 1.5157112e-04 0.00e+00 1.69e-06 -7.3 5.92e-01 - 7.61e-01 3.87e-01f 1 sigma=5.02e-01 qf=12y A\n", + " 9 1.8967944e-04 0.00e+00 8.80e-06 -8.9 3.74e-01 - 1.00e+00 4.43e-01f 1 sigma=5.25e-01 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 9 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 4.7105855020060422e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999696392259385e-01\n", + "Current barrier parameter mu = 1.1688712326242752e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999087400078834e-01\n", "\n", - "||curr_x||_inf = 9.9474954502232282e-01\n", - "||curr_s||_inf = 4.7358888572468822e-05\n", + "||curr_x||_inf = 9.9999911421011822e-01\n", + "||curr_s||_inf = 8.7229631548235875e-02\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 1.9423310763908390e-03\n", - "||curr_z_L||_inf = 1.4602105058149585e-06\n", - "||curr_z_U||_inf = 2.1675375657906144e-06\n", - "||curr_v_L||_inf = 1.9423358598698007e-03\n", + "||curr_y_d||_inf = 1.6510359940956756e-07\n", + "||curr_z_L||_inf = 1.6977757737158464e-08\n", + "||curr_z_U||_inf = 1.4770869313922699e-05\n", + "||curr_v_L||_inf = 1.8131653123302750e-08\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 9:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.5157111954755266e-04 1.5157111954755266e-04\n", - "Dual infeasibility......: 1.6867106147978903e-06 1.6867106147978903e-06\n", + "Objective...............: 1.8967943962008491e-04 1.8967943962008491e-04\n", + "Dual infeasibility......: 8.8010660757903920e-06 8.8010660757903920e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 1.0660132784860406e-07 1.0660132784860406e-07\n", - "Overall NLP error.......: 1.6867106147978903e-06 1.6867106147978903e-06\n", + "Complementarity.........: 5.5581777801334431e-07 5.5581777801334431e-07\n", + "Overall NLP error.......: 8.8010660757903920e-06 8.8010660757903920e-06\n", "\n", "\n", "\n", @@ -3540,10 +3523,10 @@ "**************************************************\n", "\n", "Staying in free mu mode.\n", - "The current filter has 4 entries.\n", + "The current filter has 1 entries.\n", "Solving the Primal Dual System for the affine step\n", - "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", - "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20726\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 245484\n", "Factorization successful.\n", "Factorization successful.\n", "Factorization successful.\n", @@ -3552,18 +3535,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 6.119179e-01\n", - "Barrier parameter mu computed by oracle is 3.638545e-08\n", - "Barrier parameter mu after safeguards is 3.638545e-08\n", - "Barrier Parameter: 3.638545e-08\n", + "Sigma = 5.712150e+00\n", + "Barrier parameter mu computed by oracle is 6.804777e-09\n", + "Barrier parameter mu after safeguards is 6.804777e-09\n", + "Barrier Parameter: 6.804777e-09\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 9:\n", "**************************************************\n", "\n", - "residual_ratio = 3.686412e-16\n", + "residual_ratio = 2.611364e-16\n", "Factorization successful.\n", - "residual_ratio = 3.100413e-18\n", + "residual_ratio = 1.684751e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 9:\n", @@ -3573,17 +3556,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.294812E-22\n", - "Starting checks for alpha (primal) = 5.55e-01\n", + "minimal step size ALPHA_MIN = 0.000000E+00\n", + "Starting checks for alpha (primal) = 1.19e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.134e-17 (tol = 1.000e-07) r (rel) = 4.172e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.512e-17 (tol = 1.000e-07) r (rel) = 4.554e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 5.552074e-01:\n", - " New values of barrier function = 6.8955753836830690e-04 (reference 7.0787819500982025e-04):\n", - " New values of constraint violation = 2.0771280745748855e-16 (reference 4.1046538997585591e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 1.185538e-01:\n", + " New values of barrier function = 2.9359625148661368e-04 (reference 3.0890151339916074e-04):\n", + " New values of constraint violation = 2.4980018054066022e-16 (reference 0.0000000000000000e+00):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -3596,11 +3579,11 @@ "*** Update HessianMatrix for Iteration 10:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 4.031716e-01\n", + "In limited-memory update, s_new_max is 9.765522e-02\n", "Limited-Memory test for skipping:\n", - " s^Ty = 7.275344e-06 snrm = 3.216928e+00 ynrm = 4.980256e-06\n", + " s^Ty = 1.432240e-06 snrm = 6.891064e-01 ynrm = 2.845964e-06\n", " Perform the update.\n", - "sigma (for B0) is 7.030250e-07\n", + "sigma (for B0) is 3.016083e-06\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -3609,33 +3592,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 10 1.2046287e-04 0.00e+00 2.91e-06 -7.4 7.26e-01 - 9.12e-01 5.55e-01f 1 sigma=6.12e-01 qf=12y A\n", + " 10 1.7450685e-04 0.00e+00 4.72e-06 -8.2 8.24e-01 - 7.85e-01 1.19e-01f 1 sigma=5.71e+00 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 10 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 3.6385446036460799e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999831328938515e-01\n", + "Current barrier parameter mu = 6.8047768781023318e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999119893392419e-01\n", "\n", - "||curr_x||_inf = 9.9348495109050994e-01\n", - "||curr_s||_inf = 3.9850597270977183e-05\n", + "||curr_x||_inf = 9.9999915052243504e-01\n", + "||curr_s||_inf = 8.2537523747637370e-02\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 1.3773791778169796e-03\n", - "||curr_z_L||_inf = 2.0646667776405924e-06\n", - "||curr_z_U||_inf = 3.5229788276053282e-06\n", - "||curr_v_L||_inf = 1.3773814927461134e-03\n", + "||curr_y_d||_inf = 2.1705017790600958e-07\n", + "||curr_z_L||_inf = 2.7565271587902688e-07\n", + "||curr_z_U||_inf = 3.3651070777012089e-06\n", + "||curr_v_L||_inf = 7.1565545332452713e-08\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 10:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.2046287079371508e-04 1.2046287079371508e-04\n", - "Dual infeasibility......: 2.9058395063124926e-06 2.9058395063124926e-06\n", + "Objective...............: 1.7450684775751929e-04 1.7450684775751929e-04\n", + "Dual infeasibility......: 4.7168964129096463e-06 4.7168964129096463e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 6.6000103088276991e-08 6.6000103088276991e-08\n", - "Overall NLP error.......: 2.9058395063124926e-06 2.9058395063124926e-06\n", + "Complementarity.........: 8.1545100546006939e-08 8.1545100546006939e-08\n", + "Overall NLP error.......: 4.7168964129096463e-06 4.7168964129096463e-06\n", "\n", "\n", "\n", @@ -3645,10 +3628,10 @@ "**************************************************\n", "\n", "Staying in free mu mode.\n", - "The current filter has 3 entries.\n", + "The current filter has 2 entries.\n", "Solving the Primal Dual System for the affine step\n", - "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", - "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20726\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 245484\n", "Factorization successful.\n", "Factorization successful.\n", "Factorization successful.\n", @@ -3657,18 +3640,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 5.096845e-01\n", - "Barrier parameter mu computed by oracle is 2.007689e-08\n", - "Barrier parameter mu after safeguards is 2.007689e-08\n", - "Barrier Parameter: 2.007689e-08\n", + "Sigma = 3.912269e+00\n", + "Barrier parameter mu computed by oracle is 2.192765e-08\n", + "Barrier parameter mu after safeguards is 2.192765e-08\n", + "Barrier Parameter: 2.192765e-08\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 10:\n", "**************************************************\n", "\n", - "residual_ratio = 3.144128e-16\n", + "residual_ratio = 1.225039e-15\n", "Factorization successful.\n", - "residual_ratio = 2.221641e-18\n", + "residual_ratio = 3.546803e-20\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 10:\n", @@ -3678,17 +3661,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 7.677864E-23\n", - "Starting checks for alpha (primal) = 2.89e-01\n", + "minimal step size ALPHA_MIN = 3.325934E-23\n", + "Starting checks for alpha (primal) = 9.25e-02\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.506e-17 (tol = 1.000e-07) r (rel) = 4.547e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.607e-17 (tol = 1.000e-07) r (rel) = 4.649e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 2.891424e-01:\n", - " New values of barrier function = 4.2720975148139069e-04 (reference 4.3447995549844142e-04):\n", - " New values of constraint violation = 2.7057620467091370e-17 (reference 2.0771280745748855e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 9.245990e-02:\n", + " New values of barrier function = 5.4327826510422558e-04 (reference 5.5825944844870121e-04):\n", + " New values of constraint violation = 1.3877787807814457e-17 (reference 2.4980018054066022e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -3701,11 +3684,11 @@ "*** Update HessianMatrix for Iteration 11:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 1.889501e-01\n", + "In limited-memory update, s_new_max is 8.169888e-02\n", "Limited-Memory test for skipping:\n", - " s^Ty = 1.031849e-06 snrm = 1.455994e+00 ynrm = 1.976844e-06\n", + " s^Ty = 1.040866e-06 snrm = 7.519364e-01 ynrm = 2.004767e-06\n", " Perform the update.\n", - "sigma (for B0) is 4.867402e-07\n", + "sigma (for B0) is 1.840909e-06\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -3714,33 +3697,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 11 1.0972743e-04 0.00e+00 1.53e-06 -7.7 6.53e-01 - 9.96e-01 2.89e-01f 1 sigma=5.10e-01 qf=12y A\n", + " 11 1.6143144e-04 0.00e+00 3.99e-06 -7.7 8.84e-01 - 8.08e-01 9.25e-02f 1 sigma=3.91e+00 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 11 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 2.0076891141407482e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999709416049365e-01\n", + "Current barrier parameter mu = 2.1927650507305542e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999528310358710e-01\n", "\n", - "||curr_x||_inf = 9.9999980896654761e-01\n", - "||curr_s||_inf = 3.6187122950754467e-05\n", + "||curr_x||_inf = 9.9999962463304148e-01\n", + "||curr_s||_inf = 7.6238407362348146e-02\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 9.4313123085838183e-04\n", - "||curr_z_L||_inf = 1.6008460191352999e-06\n", - "||curr_z_U||_inf = 3.2752470267311920e-06\n", - "||curr_v_L||_inf = 9.4311930186213508e-04\n", + "||curr_y_d||_inf = 4.3545105117044743e-07\n", + "||curr_z_L||_inf = 2.0418149796009956e-07\n", + "||curr_z_U||_inf = 6.0675644200933568e-06\n", + "||curr_v_L||_inf = 2.7601256903364457e-07\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 11:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.0972742596878089e-04 1.0972742596878089e-04\n", - "Dual infeasibility......: 1.5293804154460305e-06 1.5293804154460305e-06\n", + "Objective...............: 1.6143143653598525e-04 1.6143143653598525e-04\n", + "Dual infeasibility......: 3.9925007150679478e-06 3.9925007150679478e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 5.7418659357479277e-08 5.7418659357479277e-08\n", - "Overall NLP error.......: 1.5293804154460305e-06 1.5293804154460305e-06\n", + "Complementarity.........: 7.8857175341615126e-08 7.8857175341615126e-08\n", + "Overall NLP error.......: 3.9925007150679478e-06 3.9925007150679478e-06\n", "\n", "\n", "\n", @@ -3762,18 +3745,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 9.900000e-01\n", - "Barrier parameter mu computed by oracle is 2.112851e-08\n", - "Barrier parameter mu after safeguards is 2.112851e-08\n", - "Barrier Parameter: 2.112851e-08\n", + "Sigma = 5.214675e+00\n", + "Barrier parameter mu computed by oracle is 9.718882e-08\n", + "Barrier parameter mu after safeguards is 9.718882e-08\n", + "Barrier Parameter: 9.718882e-08\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 11:\n", "**************************************************\n", "\n", - "residual_ratio = 2.759637e-16\n", + "residual_ratio = 4.588268e-16\n", "Factorization successful.\n", - "residual_ratio = 1.099384e-17\n", + "residual_ratio = 6.951921e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 11:\n", @@ -3783,17 +3766,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 5.593925E-24\n", - "Starting checks for alpha (primal) = 5.54e-01\n", + "minimal step size ALPHA_MIN = 2.425166E-25\n", + "Starting checks for alpha (primal) = 7.16e-02\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.900e-17 (tol = 1.000e-07) r (rel) = 4.945e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.695e-17 (tol = 1.000e-07) r (rel) = 4.738e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 5.536510e-01:\n", - " New values of barrier function = 4.3172665918063944e-04 (reference 4.4383941446602901e-04):\n", - " New values of constraint violation = 5.6500485713650850e-17 (reference 2.7057620467091370e-17):\n", + "Checking acceptability for trial step size alpha_primal_test= 7.158646e-02:\n", + " New values of barrier function = 1.8031951802543412e-03 (reference 1.8538718649524723e-03):\n", + " New values of constraint violation = 1.3877787807814457e-16 (reference 1.3877787807814457e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -3806,11 +3789,11 @@ "*** Update HessianMatrix for Iteration 12:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 3.147200e-01\n", + "In limited-memory update, s_new_max is 3.076230e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 5.351565e-06 snrm = 3.830228e+00 ynrm = 3.454497e-06\n", + " s^Ty = 1.349134e-06 snrm = 1.642182e+00 ynrm = 3.031010e-06\n", " Perform the update.\n", - "sigma (for B0) is 3.647805e-07\n", + "sigma (for B0) is 5.002792e-07\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -3819,33 +3802,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 12 8.7777014e-05 0.00e+00 1.93e-06 -7.7 5.68e-01 - 9.39e-01 5.54e-01f 1 sigma=9.90e-01 qf=12y A\n", + " 12 1.4616727e-04 0.00e+00 3.90e-06 -7.0 4.30e+00 - 7.21e-01 7.16e-02f 1 sigma=5.21e+00 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 12 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 2.1128514827613649e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999847061958458e-01\n", + "Current barrier parameter mu = 9.7188818767730056e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999600749928497e-01\n", "\n", - "||curr_x||_inf = 9.9269074254289646e-01\n", - "||curr_s||_inf = 3.1125173758473780e-05\n", + "||curr_x||_inf = 9.9999943416939696e-01\n", + "||curr_s||_inf = 6.1090664794212685e-02\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 8.2925995310985465e-04\n", - "||curr_z_L||_inf = 5.5510583861884247e-07\n", - "||curr_z_U||_inf = 2.5162498793861936e-06\n", - "||curr_v_L||_inf = 8.2926183453545103e-04\n", + "||curr_y_d||_inf = 1.7614114400852782e-06\n", + "||curr_z_L||_inf = 1.5838416428457125e-06\n", + "||curr_z_U||_inf = 4.9896288688892115e-06\n", + "||curr_v_L||_inf = 1.5481951742530442e-06\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 12:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 8.7777014205619642e-05 8.7777014205619642e-05\n", - "Dual infeasibility......: 1.9310599295723908e-06 1.9310599295723908e-06\n", + "Objective...............: 1.4616726848487558e-04 1.4616726848487558e-04\n", + "Dual infeasibility......: 3.8967095020415286e-06 3.8967095020415286e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 3.6625901774115287e-08 3.6625901774115287e-08\n", - "Overall NLP error.......: 1.9310599295723908e-06 1.9310599295723908e-06\n", + "Complementarity.........: 1.6049327131365953e-06 1.6049327131365953e-06\n", + "Overall NLP error.......: 3.8967095020415286e-06 3.8967095020415286e-06\n", "\n", "\n", "\n", @@ -3867,18 +3850,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 4.075420e-01\n", - "Barrier parameter mu computed by oracle is 8.789274e-09\n", - "Barrier parameter mu after safeguards is 8.789274e-09\n", - "Barrier Parameter: 8.789274e-09\n", + "Sigma = 8.441891e-02\n", + "Barrier parameter mu computed by oracle is 6.158000e-09\n", + "Barrier parameter mu after safeguards is 6.158000e-09\n", + "Barrier Parameter: 6.158000e-09\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 12:\n", "**************************************************\n", "\n", - "residual_ratio = 3.389405e-16\n", + "residual_ratio = 1.215735e-16\n", "Factorization successful.\n", - "residual_ratio = 2.069211e-17\n", + "residual_ratio = 2.805542e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 12:\n", @@ -3888,17 +3871,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 5.809344E-23\n", - "Starting checks for alpha (primal) = 4.29e-01\n", + "minimal step size ALPHA_MIN = 5.450510E-22\n", + "Starting checks for alpha (primal) = 7.17e-02\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.925e-17 (tol = 1.000e-07) r (rel) = 4.970e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.932e-17 (tol = 1.000e-07) r (rel) = 4.978e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 4.287589e-01:\n", - " New values of barrier function = 2.2236974805033534e-04 (reference 2.3085699915389052e-04):\n", - " New values of constraint violation = 2.1632882659695929e-16 (reference 5.6500485713650850e-17):\n", + "Checking acceptability for trial step size alpha_primal_test= 7.169744e-02:\n", + " New values of barrier function = 2.4445944661744868e-04 (reference 2.5115853512086507e-04):\n", + " New values of constraint violation = 1.8735013540549517e-16 (reference 1.3877787807814457e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -3911,11 +3894,11 @@ "*** Update HessianMatrix for Iteration 13:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 1.542398e-01\n", + "In limited-memory update, s_new_max is 1.595949e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 2.315770e-06 snrm = 2.738706e+00 ynrm = 2.016783e-06\n", + " s^Ty = 3.400575e-07 snrm = 5.031567e-01 ynrm = 1.740702e-06\n", " Perform the update.\n", - "sigma (for B0) is 3.087484e-07\n", + "sigma (for B0) is 1.343216e-06\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -3924,33 +3907,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 13 7.5639338e-05 0.00e+00 2.93e-07 -8.1 3.60e-01 - 7.50e-01 4.29e-01f 1 sigma=4.08e-01 qf=12y A\n", + " 13 1.3978413e-04 0.00e+00 3.62e-06 -8.2 2.23e+00 - 2.48e-01 7.17e-02f 1 sigma=8.44e-02 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 13 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 8.7892737430615351e-09\n", - "Current fraction-to-the-boundary parameter tau = 9.9999806894007037e-01\n", + "Current barrier parameter mu = 6.1579995803349112e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999610329049793e-01\n", "\n", - "||curr_x||_inf = 9.9999998101301035e-01\n", - "||curr_s||_inf = 2.8253250421433731e-05\n", + "||curr_x||_inf = 9.9999978536756351e-01\n", + "||curr_s||_inf = 5.6981685989546761e-02\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 5.5271654158366276e-04\n", - "||curr_z_L||_inf = 1.5806351019790904e-07\n", - "||curr_z_U||_inf = 3.1585860841021594e-06\n", - "||curr_v_L||_inf = 5.5271370496960204e-04\n", + "||curr_y_d||_inf = 1.7174317408092741e-06\n", + "||curr_z_L||_inf = 1.1315752844938043e-06\n", + "||curr_z_U||_inf = 4.2503725426134735e-06\n", + "||curr_v_L||_inf = 1.5494338772653073e-06\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 13:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 7.5639338012802059e-05 7.5639338012802059e-05\n", - "Dual infeasibility......: 2.9308063523552921e-07 2.9308063523552921e-07\n", + "Objective...............: 1.3978413433019787e-04 1.3978413433019787e-04\n", + "Dual infeasibility......: 3.6183773174209675e-06 3.6183773174209675e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 2.4306152807359659e-08 2.4306152807359659e-08\n", - "Overall NLP error.......: 2.9308063523552921e-07 2.9308063523552921e-07\n", + "Complementarity.........: 1.2003700881234475e-06 1.2003700881234475e-06\n", + "Overall NLP error.......: 3.6183773174209675e-06 3.6183773174209675e-06\n", "\n", "\n", "\n", @@ -3972,18 +3955,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 1.922145e-01\n", - "Barrier parameter mu computed by oracle is 2.370645e-09\n", - "Barrier parameter mu after safeguards is 2.370645e-09\n", - "Barrier Parameter: 2.370645e-09\n", + "Sigma = 4.231797e-02\n", + "Barrier parameter mu computed by oracle is 2.366330e-09\n", + "Barrier parameter mu after safeguards is 2.366330e-09\n", + "Barrier Parameter: 2.366330e-09\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 13:\n", "**************************************************\n", "\n", - "residual_ratio = 7.921762e-17\n", + "residual_ratio = 1.065504e-16\n", "Factorization successful.\n", - "residual_ratio = 3.217258e-17\n", + "residual_ratio = 8.879202e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 13:\n", @@ -3993,17 +3976,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.004053E-21\n", - "Starting checks for alpha (primal) = 7.02e-01\n", + "minimal step size ALPHA_MIN = 9.939801E-22\n", + "Starting checks for alpha (primal) = 1.42e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 5.543e-17 (tol = 1.000e-07) r (rel) = 5.594e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.689e-17 (tol = 1.000e-07) r (rel) = 4.732e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 7.024951e-01:\n", - " New values of barrier function = 1.0173332444787992e-04 (reference 1.1521550481585263e-04):\n", - " New values of constraint violation = 2.0838365755171395e-16 (reference 2.1632882659695929e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 1.423306e-01:\n", + " New values of barrier function = 1.6887198362840910e-04 (reference 1.8000763296455044e-04):\n", + " New values of constraint violation = 1.9428902930940239e-16 (reference 1.8735013540549517e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -4016,11 +3999,11 @@ "*** Update HessianMatrix for Iteration 14:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 3.966060e-01\n", + "In limited-memory update, s_new_max is 1.112276e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 6.358785e-06 snrm = 5.288576e+00 ynrm = 2.597870e-06\n", + " s^Ty = 8.986321e-07 snrm = 8.534028e-01 ynrm = 1.713959e-06\n", " Perform the update.\n", - "sigma (for B0) is 2.273509e-07\n", + "sigma (for B0) is 1.233882e-06\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -4029,33 +4012,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 14 5.9500382e-05 0.00e+00 2.94e-07 -8.6 5.65e-01 - 4.18e-01 7.02e-01f 1 sigma=1.92e-01 qf=12y A\n", + " 14 1.2878289e-04 0.00e+00 3.25e-06 -8.6 7.81e-01 - 4.64e-01 1.42e-01f 1 sigma=4.23e-02 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 14 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 2.3706453464152239e-09\n", - "Current fraction-to-the-boundary parameter tau = 9.9999970691936479e-01\n", + "Current barrier parameter mu = 2.3663295795116899e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999638162268256e-01\n", "\n", - "||curr_x||_inf = 9.9999995557784849e-01\n", - "||curr_s||_inf = 2.1014804531052428e-05\n", + "||curr_x||_inf = 9.9999984354886429e-01\n", + "||curr_s||_inf = 5.0117248554114840e-02\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 4.4102964736513433e-04\n", - "||curr_z_L||_inf = 1.0510144320550275e-07\n", - "||curr_z_U||_inf = 2.8623304964562342e-06\n", - "||curr_v_L||_inf = 4.4103521359022515e-04\n", + "||curr_y_d||_inf = 1.5703584637908501e-06\n", + "||curr_z_L||_inf = 6.3548105205339079e-07\n", + "||curr_z_U||_inf = 2.4740214478608344e-06\n", + "||curr_v_L||_inf = 1.4582507733731510e-06\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 14:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 5.9500382156498893e-05 5.9500382156498893e-05\n", - "Dual infeasibility......: 2.9390926419814768e-07 2.9390926419814768e-07\n", + "Objective...............: 1.2878289280940380e-04 1.2878289280940380e-04\n", + "Dual infeasibility......: 3.2522524724972614e-06 3.2522524724972614e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 1.4419656232729822e-08 1.4419656232729822e-08\n", - "Overall NLP error.......: 2.9390926419814768e-07 2.9390926419814768e-07\n", + "Complementarity.........: 1.0215871969018548e-07 1.0215871969018548e-07\n", + "Overall NLP error.......: 3.2522524724972614e-06 3.2522524724972614e-06\n", "\n", "\n", "\n", @@ -4065,7 +4048,7 @@ "**************************************************\n", "\n", "Staying in free mu mode.\n", - "The current filter has 4 entries.\n", + "The current filter has 5 entries.\n", "Solving the Primal Dual System for the affine step\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", @@ -4077,18 +4060,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 1.000000e+02\n", - "Barrier parameter mu computed by oracle is 7.926865e-07\n", - "Barrier parameter mu after safeguards is 7.926865e-07\n", - "Barrier Parameter: 7.926865e-07\n", + "Sigma = 3.492338e-01\n", + "Barrier parameter mu computed by oracle is 1.075817e-08\n", + "Barrier parameter mu after safeguards is 1.075817e-08\n", + "Barrier Parameter: 1.075817e-08\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 14:\n", "**************************************************\n", "\n", - "residual_ratio = 1.323454e-14\n", + "residual_ratio = 6.836371e-16\n", "Factorization successful.\n", - "residual_ratio = 1.115951e-15\n", + "residual_ratio = 3.418186e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 14:\n", @@ -4098,17 +4081,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.020983E-27\n", - "Starting checks for alpha (primal) = 8.96e-03\n", + "minimal step size ALPHA_MIN = 1.475955E-23\n", + "Starting checks for alpha (primal) = 2.49e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 5.367e-17 (tol = 1.000e-07) r (rel) = 5.416e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.783e-17 (tol = 1.000e-07) r (rel) = 4.827e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 8.962402e-03:\n", - " New values of barrier function = 1.3459815043010175e-02 (reference 1.4181175584509374e-02):\n", - " New values of constraint violation = 1.3408192741856673e-16 (reference 2.0838365755171395e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 2.494349e-01:\n", + " New values of barrier function = 2.8793783012781349e-04 (reference 3.1104203556945210e-04):\n", + " New values of constraint violation = 3.0184188481996443e-16 (reference 1.9428902930940239e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -4121,11 +4104,11 @@ "*** Update HessianMatrix for Iteration 15:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 9.655825e-01\n", + "In limited-memory update, s_new_max is 3.038099e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 2.737012e-06 snrm = 4.103109e+00 ynrm = 2.011375e-06\n", + " s^Ty = 3.344709e-06 snrm = 2.279230e+00 ynrm = 2.807600e-06\n", " Perform the update.\n", - "sigma (for B0) is 1.625738e-07\n", + "sigma (for B0) is 6.438459e-07\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -4134,33 +4117,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 15 7.0708911e-05 0.00e+00 7.88e-07 -6.1 1.08e+02 - 1.60e-02 8.96e-03f 1 sigma=1.00e+02 qf=13y A\n", + " 15 1.0883023e-04 0.00e+00 2.47e-06 -8.0 1.22e+00 - 7.54e-01 2.49e-01f 1 sigma=3.49e-01 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 15 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 7.9268650929104817e-07\n", - "Current fraction-to-the-boundary parameter tau = 9.9999970609073585e-01\n", + "Current barrier parameter mu = 1.0758168664805035e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999674774752745e-01\n", "\n", - "||curr_x||_inf = 9.9999972620626976e-01\n", - "||curr_s||_inf = 2.7530631909999081e-05\n", + "||curr_x||_inf = 9.9999998467105289e-01\n", + "||curr_s||_inf = 2.9807991107479035e-02\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 7.9264470760534054e-04\n", - "||curr_z_L||_inf = 3.6793286125936317e-07\n", - "||curr_z_U||_inf = 3.3840413322241640e-06\n", - "||curr_v_L||_inf = 7.9266156458749314e-04\n", + "||curr_y_d||_inf = 2.3736033723722621e-06\n", + "||curr_z_L||_inf = 6.3329327364711866e-07\n", + "||curr_z_U||_inf = 3.5605240166691881e-06\n", + "||curr_v_L||_inf = 2.3073659570188645e-06\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 15:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 7.0708911268401221e-05 7.0708911268401221e-05\n", - "Dual infeasibility......: 7.8827386556891253e-07 7.8827386556891253e-07\n", + "Objective...............: 1.0883022957462301e-04 1.0883022957462301e-04\n", + "Dual infeasibility......: 2.4743478651117703e-06 2.4743478651117703e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 3.6793276420098123e-07 3.6793276420098123e-07\n", - "Overall NLP error.......: 7.8827386556891253e-07 7.8827386556891253e-07\n", + "Complementarity.........: 9.4375112849242822e-08 9.4375112849242822e-08\n", + "Overall NLP error.......: 2.4743478651117703e-06 2.4743478651117703e-06\n", "\n", "\n", "\n", @@ -4169,13 +4152,9 @@ "*** Update Barrier Parameter for Iteration 15:\n", "**************************************************\n", "\n", - "Switching to fixed mu mode with mu = 1.6260778144823957e-08 and tau = 9.9999998373922183e-01.\n", - "Barrier Parameter: 1.626078e-08\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 15:\n", - "**************************************************\n", - "\n", + "Staying in free mu mode.\n", + "The current filter has 6 entries.\n", + "Solving the Primal Dual System for the affine step\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -4184,9 +4163,20 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 1.123118e-17\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 2.302406e+00\n", + "Barrier parameter mu computed by oracle is 3.508240e-08\n", + "Barrier parameter mu after safeguards is 3.508240e-08\n", + "Barrier Parameter: 3.508240e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 15:\n", + "**************************************************\n", + "\n", + "residual_ratio = 3.028676e-16\n", "Factorization successful.\n", - "residual_ratio = 1.778592e-18\n", + "residual_ratio = 6.730391e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 15:\n", @@ -4196,17 +4186,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.925856E-24\n", - "Starting checks for alpha (primal) = 2.81e-01\n", + "minimal step size ALPHA_MIN = 6.554749E-24\n", + "Starting checks for alpha (primal) = 1.17e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 5.355e-17 (tol = 1.000e-07) r (rel) = 5.404e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.171e-17 (tol = 1.000e-07) r (rel) = 5.218e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 2.808779e-01:\n", - " New values of barrier function = 3.4492840146736439e-04 (reference 3.4536639794680328e-04):\n", - " New values of constraint violation = 1.0561145599545518e-16 (reference 1.3408192741856673e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 1.167047e-01:\n", + " New values of barrier function = 6.6187976896193285e-04 (reference 6.9290023535316641e-04):\n", + " New values of constraint violation = 4.2674197509029455e-16 (reference 3.0184188481996443e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -4219,11 +4209,11 @@ "*** Update HessianMatrix for Iteration 16:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 1.941760e-01\n", + "In limited-memory update, s_new_max is 3.210703e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 2.027273e-08 snrm = 6.169946e-01 ynrm = 3.586992e-07\n", + " s^Ty = 1.182157e-06 snrm = 3.119149e+00 ynrm = 1.878009e-06\n", " Perform the update.\n", - "sigma (for B0) is 5.325366e-08\n", + "sigma (for B0) is 1.215074e-07\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -4232,33 +4222,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 16 7.0037779e-05 0.00e+00 3.03e-07 -7.8 6.91e-01 - 6.55e-01 2.81e-01f 1 Fy A\n", + " 16 9.6549110e-05 0.00e+00 2.75e-06 -7.5 2.75e+00 - 6.41e-01 1.17e-01f 1 sigma=2.30e+00 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 16 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 1.6260778144823957e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999998373922183e-01\n", + "Current barrier parameter mu = 3.5082395246276397e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999752565213484e-01\n", "\n", - "||curr_x||_inf = 1.0000000068425474e+00\n", - "||curr_s||_inf = 2.7298053821905725e-05\n", + "||curr_x||_inf = 9.9808476905569421e-01\n", + "||curr_s||_inf = 2.2102572293494945e-03\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 6.7582362595528574e-04\n", - "||curr_z_L||_inf = 2.8401168782898505e-07\n", - "||curr_z_U||_inf = 3.0585395471179582e-06\n", - "||curr_v_L||_inf = 6.7583214759629111e-04\n", + "||curr_y_d||_inf = 1.3389972551909327e-05\n", + "||curr_z_L||_inf = 9.9672398206692459e-07\n", + "||curr_z_U||_inf = 5.0638828746768668e-06\n", + "||curr_v_L||_inf = 1.3324379447899140e-05\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 16:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 7.0037778623259896e-05 7.0037778623259896e-05\n", - "Dual infeasibility......: 3.0266326294118364e-07 3.0266326294118364e-07\n", + "Objective...............: 9.6549110124383286e-05 9.6549110124383286e-05\n", + "Dual infeasibility......: 2.7478596920982142e-06 2.7478596920982142e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 2.3554896737735246e-07 2.3554896737735246e-07\n", - "Overall NLP error.......: 3.0266326294118364e-07 3.0266326294118364e-07\n", + "Complementarity.........: 9.7686422459341640e-08 9.7686422459341640e-08\n", + "Overall NLP error.......: 2.7478596920982142e-06 2.7478596920982142e-06\n", "\n", "\n", "\n", @@ -4267,13 +4257,9 @@ "*** Update Barrier Parameter for Iteration 16:\n", "**************************************************\n", "\n", - "Remaining in fixed mu mode.\n", - "Barrier Parameter: 1.626078e-08\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 16:\n", - "**************************************************\n", - "\n", + "Staying in free mu mode.\n", + "The current filter has 7 entries.\n", + "Solving the Primal Dual System for the affine step\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -4282,28 +4268,40 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 2.179247e-15\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 1.804931e+00\n", + "Barrier parameter mu computed by oracle is 5.079196e-08\n", + "Barrier parameter mu after safeguards is 5.079196e-08\n", + "Barrier Parameter: 5.079196e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 16:\n", + "**************************************************\n", + "\n", + "residual_ratio = 7.247913e-16\n", "Factorization successful.\n", - "residual_ratio = 6.802861e-18\n", + "residual_ratio = 1.782274e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 16:\n", "**************************************************\n", "\n", "--> Starting line search in iteration 16 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 3.458089E-26\n", - "Starting checks for alpha (primal) = 2.95e-01\n", + "minimal step size ALPHA_MIN = 6.256062E-23\n", + "Starting checks for alpha (primal) = 1.95e-02\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 5.476e-17 (tol = 1.000e-07) r (rel) = 5.526e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.812e-17 (tol = 1.000e-07) r (rel) = 4.857e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 2.945346e-01:\n", - " New values of barrier function = 3.4393488656618326e-04 (reference 3.4492840146736439e-04):\n", - " New values of constraint violation = 5.8180999081003382e-17 (reference 1.0561145599545518e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 1.946207e-02:\n", + " New values of barrier function = 9.1232498543689544e-04 (reference 9.1502943778027637e-04):\n", + " New values of constraint violation = 4.2023676205538152e-16 (reference 4.2674197509029455e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -4316,9 +4314,9 @@ "*** Update HessianMatrix for Iteration 17:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 8.766042e-01\n", + "In limited-memory update, s_new_max is 4.546274e-02\n", "Limited-Memory test for skipping:\n", - " s^Ty = -2.999276e-07 snrm = 1.384476e+00 ynrm = 1.216035e-06\n", + " s^Ty = -3.574878e-10 snrm = 2.805333e-01 ynrm = 1.187297e-07\n", " Skip the update.\n", "Number of successive iterations with skipping: 1\n", "\n", @@ -4328,33 +4326,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 17 6.8969782e-05 0.00e+00 7.87e-07 -7.8 2.98e+00 - 6.08e-01 2.95e-01f 1 Fy AWs\n", + " 17 9.5888193e-05 0.00e+00 1.23e-06 -7.3 2.34e+00 - 5.27e-01 1.95e-02f 1 sigma=1.80e+00 qf=12y AWs\n", "\n", "**************************************************\n", "*** Beginning Iteration 17 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 1.6260778144823957e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999998373922183e-01\n", + "Current barrier parameter mu = 5.0791956719929144e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999725214030788e-01\n", "\n", - "||curr_x||_inf = 9.9999999574573328e-01\n", - "||curr_s||_inf = 2.7209387033486182e-05\n", + "||curr_x||_inf = 9.9788881953043662e-01\n", + "||curr_s||_inf = 3.9264957713176785e-09\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 6.3151349170585919e-04\n", - "||curr_z_L||_inf = 1.0346085029701024e-07\n", - "||curr_z_U||_inf = 2.8973925022172645e-06\n", - "||curr_v_L||_inf = 6.3151916724826806e-04\n", + "||curr_y_d||_inf = 3.7944528595895260e-04\n", + "||curr_z_L||_inf = 5.1911720873724571e-07\n", + "||curr_z_U||_inf = 3.5214221254470305e-06\n", + "||curr_v_L||_inf = 3.7940917796523173e-04\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 17:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 6.8969781888316459e-05 6.8969781888316459e-05\n", - "Dual infeasibility......: 7.8709759613971154e-07 7.8709759613971154e-07\n", + "Objective...............: 9.5888192904728474e-05 9.5888192904728474e-05\n", + "Dual infeasibility......: 1.2309777274379566e-06 1.2309777274379566e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 8.8718473852048702e-08 8.8718473852048702e-08\n", - "Overall NLP error.......: 7.8709759613971154e-07 7.8709759613971154e-07\n", + "Complementarity.........: 9.4718469015240673e-08 9.4718469015240673e-08\n", + "Overall NLP error.......: 1.2309777274379566e-06 1.2309777274379566e-06\n", "\n", "\n", "\n", @@ -4363,13 +4361,9 @@ "*** Update Barrier Parameter for Iteration 17:\n", "**************************************************\n", "\n", - "Remaining in fixed mu mode.\n", - "Barrier Parameter: 1.626078e-08\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 17:\n", - "**************************************************\n", - "\n", + "Staying in free mu mode.\n", + "The current filter has 7 entries.\n", + "Solving the Primal Dual System for the affine step\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -4378,28 +4372,40 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 4.540383e-16\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 1.327510e-02\n", + "Barrier parameter mu computed by oracle is 5.218537e-10\n", + "Barrier parameter mu after safeguards is 5.218537e-10\n", + "Barrier Parameter: 5.218537e-10\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 17:\n", + "**************************************************\n", + "\n", + "residual_ratio = 4.788278e-16\n", "Factorization successful.\n", - "residual_ratio = 2.547272e-19\n", + "residual_ratio = 7.365046e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 17:\n", "**************************************************\n", "\n", "--> Starting line search in iteration 17 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 9.972109E-26\n", - "Starting checks for alpha (primal) = 2.80e-01\n", + "minimal step size ALPHA_MIN = 6.056113E-21\n", + "Starting checks for alpha (primal) = 5.69e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 5.086e-17 (tol = 1.000e-07) r (rel) = 5.133e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.639e-17 (tol = 1.000e-07) r (rel) = 4.682e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 2.804694e-01:\n", - " New values of barrier function = 3.4339948909169399e-04 (reference 3.4393488656618326e-04):\n", - " New values of constraint violation = 1.6249818873305399e-16 (reference 5.8180999081003382e-17):\n", + "Checking acceptability for trial step size alpha_primal_test= 5.686417e-01:\n", + " New values of barrier function = 8.7972490615553708e-05 (reference 1.0427653916588002e-04):\n", + " New values of constraint violation = 2.0148986815224802e-16 (reference 4.2023676205538152e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -4414,33 +4420,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 18 6.8317630e-05 0.00e+00 3.97e-07 -7.8 1.44e+00 - 1.00e+00 2.80e-01f 1 Fy A\n", + " 18 7.9311599e-05 0.00e+00 1.21e-06 -9.3 1.46e+00 - 6.55e-01 5.69e-01f 1 sigma=1.33e-02 qf=13y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 18 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 1.6260778144823957e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999998373922183e-01\n", + "Current barrier parameter mu = 5.2185365008694458e-10\n", + "Current fraction-to-the-boundary parameter tau = 9.9999876902227258e-01\n", "\n", - "||curr_x||_inf = 1.0000000053109623e+00\n", - "||curr_s||_inf = 2.7198896670573142e-05\n", + "||curr_x||_inf = 9.9999898911190011e-01\n", + "||curr_s||_inf = 7.7498913212568736e-07\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 5.9826160939990803e-04\n", - "||curr_z_L||_inf = 7.9325355294146254e-08\n", - "||curr_z_U||_inf = 2.2418829706615098e-06\n", - "||curr_v_L||_inf = 5.9826471336189293e-04\n", + "||curr_y_d||_inf = 3.6230944817811610e-04\n", + "||curr_z_L||_inf = 8.9612693988201973e-07\n", + "||curr_z_U||_inf = 2.0765106055154655e-06\n", + "||curr_v_L||_inf = 3.6229113813024297e-04\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 18:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 6.8317630446442925e-05 6.8317630446442925e-05\n", - "Dual infeasibility......: 3.9676785277369727e-07 3.9676785277369727e-07\n", + "Objective...............: 7.9311598799361929e-05 7.9311598799361929e-05\n", + "Dual infeasibility......: 1.2133862307492080e-06 1.2133862307492080e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 4.5913924324028463e-08 4.5913924324028463e-08\n", - "Overall NLP error.......: 3.9676785277369727e-07 3.9676785277369727e-07\n", + "Complementarity.........: 4.4696644788203451e-08 4.4696644788203451e-08\n", + "Overall NLP error.......: 1.2133862307492080e-06 1.2133862307492080e-06\n", "\n", "\n", "\n", @@ -4448,22 +4454,22 @@ "Number of Iterations....: 18\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 6.8317630446442925e-05 6.8317630446442925e-05\n", - "Dual infeasibility......: 3.9676785277369727e-07 3.9676785277369727e-07\n", + "Objective...............: 7.9311598799361929e-05 7.9311598799361929e-05\n", + "Dual infeasibility......: 1.2133862307492080e-06 1.2133862307492080e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 4.5913924324028463e-08 4.5913924324028463e-08\n", - "Overall NLP error.......: 3.9676785277369727e-07 3.9676785277369727e-07\n", + "Complementarity.........: 4.4696644788203451e-08 4.4696644788203451e-08\n", + "Overall NLP error.......: 1.2133862307492080e-06 1.2133862307492080e-06\n", "\n", "\n", - "Number of objective function evaluations = 20\n", + "Number of objective function evaluations = 19\n", "Number of objective gradient evaluations = 19\n", "Number of equality constraint evaluations = 0\n", - "Number of inequality constraint evaluations = 20\n", + "Number of inequality constraint evaluations = 19\n", "Number of equality constraint Jacobian evaluations = 0\n", "Number of inequality constraint Jacobian evaluations = 19\n", "Number of Lagrangian Hessian evaluations = 0\n", - "Total CPU secs in IPOPT (w/o function evaluations) = 2.256\n", - "Total CPU secs in NLP function evaluations = 2.173\n", + "Total CPU secs in IPOPT (w/o function evaluations) = 2.118\n", + "Total CPU secs in NLP function evaluations = 2.086\n", "\n", "EXIT: Solved To Acceptable Level.\n", "v=0.4\n", @@ -4488,7 +4494,7 @@ "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.825e-17 (tol = 1.000e-07) r (rel) = 4.873e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 9.149e-17 (tol = 1.000e-07) r (rel) = 9.240e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", @@ -4534,7 +4540,7 @@ "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", - "Least square estimates max(y_c) = 0.000000e+00, max(y_d) = 9.999011e-01\n", + "Least square estimates max(y_c) = 0.000000e+00, max(y_d) = 9.999494e-01\n", "Total number of variables............................: 10201\n", " variables with only lower bounds: 0\n", " variables with lower and upper bounds: 10201\n", @@ -4549,7 +4555,7 @@ "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.900e-17 (tol = 1.000e-07) r (rel) = 4.949e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.362e-17 (tol = 1.000e-07) r (rel) = 4.406e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "\n", @@ -4565,7 +4571,7 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 0 1.4705639e-04 0.00e+00 1.00e-04 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", + " 0 2.1942259e-04 0.00e+00 6.38e-03 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", "\n", "**************************************************\n", "*** Beginning Iteration 0 from the following point:\n", @@ -4574,10 +4580,10 @@ "Current barrier parameter mu = 1.0000000000000000e+00\n", "Current fraction-to-the-boundary parameter tau = 0.0000000000000000e+00\n", "\n", - "||curr_x||_inf = 9.2386388778686523e-01\n", - "||curr_s||_inf = 2.2181296172274489e-02\n", + "||curr_x||_inf = 8.6258387565612793e-01\n", + "||curr_s||_inf = 4.5761677855054583e-02\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 9.9990110040229219e-01\n", + "||curr_y_d||_inf = 9.9994936348413743e-01\n", "||curr_z_L||_inf = 1.0000000000000000e+00\n", "||curr_z_U||_inf = 1.0000000000000000e+00\n", "||curr_v_L||_inf = 1.0000000000000000e+00\n", @@ -4587,8 +4593,8 @@ "***Current NLP Values for Iteration 0:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.4705639051276883e-04 1.4705639051276883e-04\n", - "Dual infeasibility......: 1.0002114978358012e-04 1.0002114978358012e-04\n", + "Objective...............: 2.1942258539174670e-04 2.1942258539174670e-04\n", + "Dual infeasibility......: 6.3814363685714692e-03 6.3814363685714692e-03\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", "Complementarity.........: 9.9000001999999998e-01 9.9000001999999998e-01\n", "Overall NLP error.......: 9.9000001999999998e-01 9.9000001999999998e-01\n", @@ -4600,7 +4606,7 @@ "*** Update Barrier Parameter for Iteration 0:\n", "**************************************************\n", "\n", - "Setting mu_max to 4.999766e+02.\n", + "Setting mu_max to 4.999777e+02.\n", "Staying in free mu mode.\n", "The current filter has 1 entries.\n", "Solving the Primal Dual System for the affine step\n", @@ -4612,18 +4618,18 @@ " delta_c=0.000000e+00 delta_d=0.000000e+00\n", "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "Sigma = 1.000000e-06\n", - "Barrier parameter mu computed by oracle is 4.999766e-07\n", - "Barrier parameter mu after safeguards is 4.999766e-07\n", - "Barrier Parameter: 4.999766e-07\n", + "Sigma = 8.050304e-03\n", + "Barrier parameter mu computed by oracle is 4.024973e-03\n", + "Barrier parameter mu after safeguards is 4.024973e-03\n", + "Barrier Parameter: 4.024973e-03\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 0:\n", "**************************************************\n", "\n", - "residual_ratio = 2.231605e-16\n", + "residual_ratio = 2.242425e-16\n", "Factorization successful.\n", - "residual_ratio = 1.115803e-16\n", + "residual_ratio = 1.121212e-16\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 0:\n", @@ -4640,11 +4646,11 @@ "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.621e-17 (tol = 1.000e-07) r (rel) = 4.667e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.611e-17 (tol = 1.000e-07) r (rel) = 4.657e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", - " New values of barrier function = 1.8794893658976082e-02 (reference 1.8795059259517077e-02):\n", - " New values of constraint violation = 3.1225022567582528e-16 (reference 0.0000000000000000e+00):\n", + " New values of barrier function = 1.4305435094170917e+02 (reference 1.5242228657191637e+02):\n", + " New values of constraint violation = 3.1918911957973251e-16 (reference 0.0000000000000000e+00):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -4659,11 +4665,11 @@ "*** Update HessianMatrix for Iteration 1:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 4.867590e-07\n", + "In limited-memory update, s_new_max is 3.969198e-03\n", "Limited-Memory test for skipping:\n", - " s^Ty = 5.774963e-11 snrm = 4.170723e-05 ynrm = 2.111060e-06\n", + " s^Ty = 1.791167e-03 snrm = 3.402711e-01 ynrm = 2.386620e-02\n", " Perform the update.\n", - "sigma (for B0) is 3.319912e-02\n", + "sigma (for B0) is 1.546984e-02\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -4672,33 +4678,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 1 1.4705642e-04 0.00e+00 1.30e-06 -6.3 4.87e-07 - 9.90e-01 1.00e+00f 1 sigma=1.00e-06 qf=13y \n", + " 1 2.1996526e-04 0.00e+00 3.89e-03 -2.4 3.97e-03 - 9.96e-01 1.00e+00f 1 sigma=8.05e-03 qf=13y \n", "\n", "**************************************************\n", "*** Beginning Iteration 1 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 4.9997659095840198e-07\n", + "Current barrier parameter mu = 4.0249730728791578e-03\n", "Current fraction-to-the-boundary parameter tau = 9.8999999999999999e-01\n", "\n", - "||curr_x||_inf = 9.2386350532892458e-01\n", - "||curr_s||_inf = 2.2180963008126948e-02\n", + "||curr_x||_inf = 8.5997292196921093e-01\n", + "||curr_s||_inf = 4.3001541813346658e-02\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 1.0036028936224520e-02\n", - "||curr_z_L||_inf = 1.0000988632117624e-02\n", - "||curr_z_U||_inf = 1.0001935573413556e-02\n", - "||curr_v_L||_inf = 1.0036701407285520e-02\n", + "||curr_y_d||_inf = 1.5475754103611994e-01\n", + "||curr_z_L||_inf = 1.6325458704352203e-02\n", + "||curr_z_U||_inf = 1.4691581823443434e-02\n", + "||curr_v_L||_inf = 1.5205840589663955e-01\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 1:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.4705641727446358e-04 1.4705641727446358e-04\n", - "Dual infeasibility......: 1.3010612777680092e-06 1.3010612777680092e-06\n", + "Objective...............: 2.1996525743867739e-04 2.1996525743867739e-04\n", + "Dual infeasibility......: 3.8882951455892177e-03 3.8882951455892177e-03\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 9.9004935394333589e-03 9.9004935394333589e-03\n", - "Overall NLP error.......: 9.9004935394333589e-03 9.9004935394333589e-03\n", + "Complementarity.........: 1.2311876287447893e-02 1.2311876287447893e-02\n", + "Overall NLP error.......: 1.2311876287447893e-02 1.2311876287447893e-02\n", "\n", "\n", "\n", @@ -4707,8 +4713,8 @@ "*** Update Barrier Parameter for Iteration 1:\n", "**************************************************\n", "\n", - "Switching to fixed mu mode with mu = 4.0000153326789677e-03 and tau = 9.9599998466732098e-01.\n", - "Barrier Parameter: 4.000015e-03\n", + "Switching to fixed mu mode with mu = 4.9890526165039467e-03 and tau = 9.9501094738349605e-01.\n", + "Barrier Parameter: 4.989053e-03\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 1:\n", @@ -4722,9 +4728,9 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 9.798059e-16\n", + "residual_ratio = 2.297806e-15\n", "Factorization successful.\n", - "residual_ratio = 2.041262e-17\n", + "residual_ratio = 1.160871e-16\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 1:\n", @@ -4733,17 +4739,17 @@ "--> Starting line search in iteration 1 <--\n", "Mu has changed in line search - resetting watchdog counters.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.500076E-28\n", - "Starting checks for alpha (primal) = 9.14e-02\n", + "minimal step size ALPHA_MIN = 2.128259E-28\n", + "Starting checks for alpha (primal) = 2.08e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.619e-17 (tol = 1.000e-07) r (rel) = 4.665e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.194e-17 (tol = 1.000e-07) r (rel) = 4.236e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 9.135549e-02:\n", - " New values of barrier function = 1.0832927347178341e+02 (reference 1.4919040163438481e+02):\n", - " New values of constraint violation = 5.2388648974499574e-16 (reference 3.1225022567582528e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 2.083431e-01:\n", + " New values of barrier function = 1.1845032785048490e+02 (reference 1.7731931590127948e+02):\n", + " New values of constraint violation = 2.0816681711721685e-17 (reference 3.1918911957973251e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -4756,11 +4762,11 @@ "*** Update HessianMatrix for Iteration 2:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 1.236794e-01\n", + "In limited-memory update, s_new_max is 1.494710e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 1.073723e-06 snrm = 3.109402e+00 ynrm = 2.341267e-06\n", + " s^Ty = 2.774185e-06 snrm = 5.498271e+00 ynrm = 5.797805e-06\n", " Perform the update.\n", - "sigma (for B0) is 1.110550e-07\n", + "sigma (for B0) is 9.176627e-08\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -4769,33 +4775,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 2 1.5706600e-04 0.00e+00 3.77e-02 -2.4 1.35e+00 - 3.52e-01 9.14e-02f 1 FNhj y \n", + " 2 2.3760911e-04 0.00e+00 2.51e-01 -2.3 7.17e-01 - 6.58e-01 2.08e-01f 1 FNhj y \n", "\n", "**************************************************\n", "*** Beginning Iteration 2 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 4.0000153326789677e-03\n", - "Current fraction-to-the-boundary parameter tau = 9.9599998466732098e-01\n", + "Current barrier parameter mu = 4.9890526165039467e-03\n", + "Current fraction-to-the-boundary parameter tau = 9.9501094738349605e-01\n", "\n", - "||curr_x||_inf = 8.9279525079999800e-01\n", - "||curr_s||_inf = 8.8714232126248932e-05\n", + "||curr_x||_inf = 8.0880013000330819e-01\n", + "||curr_s||_inf = 2.1452700458810675e-04\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 1.1149088522973198e-01\n", - "||curr_z_L||_inf = 4.6646778568752101e-02\n", - "||curr_z_U||_inf = 1.0915251060010137e-02\n", - "||curr_v_L||_inf = 1.0862810916454405e-01\n", + "||curr_y_d||_inf = 6.0887417125727905e-01\n", + "||curr_z_L||_inf = 2.5838055223470702e-01\n", + "||curr_z_U||_inf = 1.1512249167357101e-02\n", + "||curr_v_L||_inf = 6.0612236954512522e-01\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 2:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.5706600049314055e-04 1.5706600049314055e-04\n", - "Dual infeasibility......: 3.7717549553117360e-02 3.7717549553117360e-02\n", + "Objective...............: 2.3760910728674287e-04 2.3760910728674287e-04\n", + "Dual infeasibility......: 2.5107298073720691e-01 2.5107298073720691e-01\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 8.8935856668390118e-03 8.8935856668390118e-03\n", - "Overall NLP error.......: 3.7717549553117360e-02 3.7717549553117360e-02\n", + "Complementarity.........: 9.2863973655349236e-03 9.2863973655349236e-03\n", + "Overall NLP error.......: 2.5107298073720691e-01 2.5107298073720691e-01\n", "\n", "\n", "\n", @@ -4805,8 +4811,7 @@ "**************************************************\n", "\n", "Remaining in fixed mu mode.\n", - "Reducing mu to 2.5298366740050934e-04 in fixed mu mode. Tau becomes 9.9599998466732098e-01\n", - "Barrier Parameter: 2.529837e-04\n", + "Barrier Parameter: 4.989053e-03\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 2:\n", @@ -4820,28 +4825,27 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 2.179391e-16\n", + "residual_ratio = 1.506347e-16\n", "Factorization successful.\n", - "residual_ratio = 8.513248e-19\n", + "residual_ratio = 3.309846e-19\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 2:\n", "**************************************************\n", "\n", "--> Starting line search in iteration 2 <--\n", - "Mu has changed in line search - resetting watchdog counters.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 4.382301E-25\n", - "Starting checks for alpha (primal) = 2.01e-02\n", + "minimal step size ALPHA_MIN = 9.802541E-29\n", + "Starting checks for alpha (primal) = 3.64e-03\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.561e-17 (tol = 1.000e-07) r (rel) = 4.606e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.269e-17 (tol = 1.000e-07) r (rel) = 4.311e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 2.014679e-02:\n", - " New values of barrier function = 6.8406535339280161e+00 (reference 6.8515050922684582e+00):\n", - " New values of constraint violation = 4.5606964385602744e-16 (reference 5.2388648974499574e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 3.637951e-03:\n", + " New values of barrier function = 1.1808811368906227e+02 (reference 1.1845032785048490e+02):\n", + " New values of constraint violation = 3.6537613212761499e-17 (reference 2.0816681711721685e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -4854,11 +4858,11 @@ "*** Update HessianMatrix for Iteration 3:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 1.656571e-03\n", + "In limited-memory update, s_new_max is 2.043233e-03\n", "Limited-Memory test for skipping:\n", - " s^Ty = 1.797414e-09 snrm = 3.777210e-02 ynrm = 7.911232e-08\n", + " s^Ty = 3.282502e-08 snrm = 1.108085e-01 ynrm = 4.660362e-07\n", " Perform the update.\n", - "sigma (for B0) is 1.259813e-06\n", + "sigma (for B0) is 2.673367e-06\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -4867,33 +4871,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 3 1.5747890e-04 0.00e+00 9.07e-04 -3.6 8.22e-02 - 9.76e-01 2.01e-02f 1 Fy \n", + " 3 2.3993882e-04 0.00e+00 8.67e-02 -2.3 5.62e-01 - 6.55e-01 3.64e-03f 1 Fy \n", "\n", "**************************************************\n", "*** Beginning Iteration 3 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 2.5298366740050934e-04\n", - "Current fraction-to-the-boundary parameter tau = 9.9599998466732098e-01\n", + "Current barrier parameter mu = 4.9890526165039467e-03\n", + "Current fraction-to-the-boundary parameter tau = 9.9501094738349605e-01\n", "\n", - "||curr_x||_inf = 8.9227696779121723e-01\n", - "||curr_s||_inf = 3.4489828888516596e-07\n", + "||curr_x||_inf = 8.0686176281733890e-01\n", + "||curr_s||_inf = 1.0603364040772225e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 8.0265653362666001e+00\n", - "||curr_z_L||_inf = 2.2214740931772239e-03\n", - "||curr_z_U||_inf = 7.8398010105133926e-04\n", - "||curr_v_L||_inf = 8.0264965095918352e+00\n", + "||curr_y_d||_inf = 1.2394395684405953e+02\n", + "||curr_z_L||_inf = 1.0500214297429558e-01\n", + "||curr_z_U||_inf = 9.5398931984432729e-03\n", + "||curr_v_L||_inf = 1.2394300679602811e+02\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 3:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.5747890013272979e-04 1.5747890013272979e-04\n", - "Dual infeasibility......: 9.0709423432018673e-04 9.0709423432018673e-04\n", + "Objective...............: 2.3993881967835905e-04 2.3993881967835905e-04\n", + "Dual infeasibility......: 8.6732203803931610e-02 8.6732203803931610e-02\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 1.3429330132109587e-03 1.3429330132109587e-03\n", - "Overall NLP error.......: 1.3429330132109587e-03 1.3429330132109587e-03\n", + "Complementarity.........: 9.8962423191280829e-03 9.8962423191280829e-03\n", + "Overall NLP error.......: 8.6732203803931610e-02 8.6732203803931610e-02\n", "\n", "\n", "\n", @@ -4903,8 +4907,7 @@ "**************************************************\n", "\n", "Remaining in fixed mu mode.\n", - "Reducing mu to 4.0238216787501142e-06 in fixed mu mode. Tau becomes 9.9974701633259944e-01\n", - "Barrier Parameter: 4.023822e-06\n", + "Barrier Parameter: 4.989053e-03\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 3:\n", @@ -4918,28 +4921,27 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 6.462265e-17\n", + "residual_ratio = 1.115028e-13\n", "Factorization successful.\n", - "residual_ratio = 1.362771e-19\n", + "residual_ratio = 5.973366e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 3:\n", "**************************************************\n", "\n", "--> Starting line search in iteration 3 <--\n", - "Mu has changed in line search - resetting watchdog counters.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.506687E-22\n", - "Starting checks for alpha (primal) = 1.00e+00\n", + "minimal step size ALPHA_MIN = 2.319218E-28\n", + "Starting checks for alpha (primal) = 3.88e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.932e-17 (tol = 1.000e-07) r (rel) = 4.981e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.967e-17 (tol = 1.000e-07) r (rel) = 4.007e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", - " New values of barrier function = 1.0748875623798022e-01 (reference 1.0895871734549814e-01):\n", - " New values of constraint violation = 4.2181923135908937e-17 (reference 4.5606964385602744e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 3.880762e-01:\n", + " New values of barrier function = 9.8706772844173358e+01 (reference 1.1808811368906227e+02):\n", + " New values of constraint violation = 4.3571374806761209e-18 (reference 3.6537613212761499e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -4952,11 +4954,11 @@ "*** Update HessianMatrix for Iteration 4:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 2.257619e-02\n", + "In limited-memory update, s_new_max is 8.028363e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 4.832449e-07 snrm = 4.442187e-01 ynrm = 1.316924e-06\n", + " s^Ty = 1.918919e-03 snrm = 1.236189e+01 ynrm = 2.175904e-04\n", " Perform the update.\n", - "sigma (for B0) is 2.448914e-06\n", + "sigma (for B0) is 1.255702e-05\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -4965,33 +4967,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 4 1.6383222e-04 0.00e+00 9.51e-08 -5.4 2.26e-02 - 1.00e+00 1.00e+00f 1 Fy A\n", + " 4 1.3355989e-03 0.00e+00 9.20e-06 -2.3 2.07e+00 - 1.00e+00 3.88e-01f 1 Fy \n", "\n", "**************************************************\n", "*** Beginning Iteration 4 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 4.0238216787501142e-06\n", - "Current fraction-to-the-boundary parameter tau = 9.9974701633259944e-01\n", + "Current barrier parameter mu = 4.9890526165039467e-03\n", + "Current fraction-to-the-boundary parameter tau = 9.9501094738349605e-01\n", "\n", - "||curr_x||_inf = 8.8364387020362267e-01\n", - "||curr_s||_inf = 8.4303043730527649e-07\n", + "||curr_x||_inf = 5.5392494124768277e-01\n", + "||curr_s||_inf = 1.6289179056517917e-05\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 7.2036774678150187e-02\n", - "||curr_z_L||_inf = 2.7078609281192705e-05\n", - "||curr_z_U||_inf = 2.0240430427977121e-05\n", - "||curr_v_L||_inf = 7.2036774876892196e-02\n", + "||curr_y_d||_inf = 1.1706085644763425e+02\n", + "||curr_z_L||_inf = 3.6912553191371558e-02\n", + "||curr_z_U||_inf = 2.5216305042701351e-02\n", + "||curr_v_L||_inf = 1.1706085624371674e+02\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 4:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.6383221527421568e-04 1.6383221527421568e-04\n", - "Dual infeasibility......: 9.5059421167469660e-08 9.5059421167469660e-08\n", + "Objective...............: 1.3355988667522343e-03 1.3355988667522343e-03\n", + "Dual infeasibility......: 9.2036685386646944e-06 9.2036685386646944e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 2.2234823091003274e-05 2.2234823091003274e-05\n", - "Overall NLP error.......: 2.2234823091003274e-05 2.2234823091003274e-05\n", + "Complementarity.........: 2.5114797920333456e-02 2.5114797920333456e-02\n", + "Overall NLP error.......: 2.5114797920333456e-02 2.5114797920333456e-02\n", "\n", "\n", "\n", @@ -5001,8 +5003,8 @@ "**************************************************\n", "\n", "Remaining in fixed mu mode.\n", - "Reducing mu to 8.0715713319459585e-09 in fixed mu mode. Tau becomes 9.9999597617832126e-01\n", - "Barrier Parameter: 8.071571e-09\n", + "Reducing mu to 3.5239288103462837e-04 in fixed mu mode. Tau becomes 9.9501094738349605e-01\n", + "Barrier Parameter: 3.523929e-04\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 4:\n", @@ -5016,9 +5018,9 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 7.748155e-15\n", + "residual_ratio = 1.380742e-16\n", "Factorization successful.\n", - "residual_ratio = 1.920912e-18\n", + "residual_ratio = 5.056430e-20\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 4:\n", @@ -5026,19 +5028,18 @@ "\n", "--> Starting line search in iteration 4 <--\n", "Mu has changed in line search - resetting watchdog counters.\n", - "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.066023E-20\n", + "minimal step size ALPHA_MIN = 3.354207E-26\n", "Starting checks for alpha (primal) = 1.00e+00\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.842e-17 (tol = 1.000e-07) r (rel) = 4.890e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.652e-17 (tol = 1.000e-07) r (rel) = 3.689e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", - " New values of barrier function = 3.7719746139176606e-04 (reference 3.7912027948436668e-04):\n", - " New values of constraint violation = 1.2391033227146346e-18 (reference 4.2181923135908937e-17):\n", + " New values of barrier function = 6.9115959742747517e+00 (reference 6.9732190561774079e+00):\n", + " New values of constraint violation = 1.0638733817514012e-16 (reference 4.3571374806761209e-18):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -5051,11 +5052,11 @@ "*** Update HessianMatrix for Iteration 5:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 3.679047e-02\n", + "In limited-memory update, s_new_max is 1.948300e-02\n", "Limited-Memory test for skipping:\n", - " s^Ty = 8.493101e-08 snrm = 1.758487e-01 ynrm = 6.934237e-07\n", + " s^Ty = 6.321949e-06 snrm = 8.472977e-01 ynrm = 1.652278e-05\n", " Perform the update.\n", - "sigma (for B0) is 2.746554e-06\n", + "sigma (for B0) is 8.806006e-06\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -5064,33 +5065,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 5 1.6154231e-04 0.00e+00 1.48e-07 -8.1 3.68e-02 - 9.54e-01 1.00e+00f 1 Fy A\n", + " 5 1.4661540e-03 0.00e+00 7.92e-07 -3.5 1.95e-02 - 1.00e+00 1.00e+00f 1 Fy A\n", "\n", "**************************************************\n", "*** Beginning Iteration 5 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 8.0715713319459585e-09\n", - "Current fraction-to-the-boundary parameter tau = 9.9999597617832126e-01\n", + "Current barrier parameter mu = 3.5239288103462837e-04\n", + "Current fraction-to-the-boundary parameter tau = 9.9501094738349605e-01\n", "\n", - "||curr_x||_inf = 8.8448176139761503e-01\n", - "||curr_s||_inf = 9.2687821942481365e-07\n", + "||curr_x||_inf = 5.3499795545488793e-01\n", + "||curr_s||_inf = 1.7327398583034476e-05\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 5.5834070059706861e-03\n", - "||curr_z_L||_inf = 1.2369686400773544e-06\n", - "||curr_z_U||_inf = 1.4733644379664898e-06\n", - "||curr_v_L||_inf = 5.5834070033907585e-03\n", + "||curr_y_d||_inf = 1.4163781748571674e+01\n", + "||curr_z_L||_inf = 2.0062032473238608e-03\n", + "||curr_z_U||_inf = 6.2443924226025967e-04\n", + "||curr_v_L||_inf = 1.4163781715087993e+01\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 5:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.6154230559720372e-04 1.6154230559720372e-04\n", - "Dual infeasibility......: 1.4772270514867860e-07 1.4772270514867860e-07\n", + "Objective...............: 1.4661539609085222e-03 1.4661539609085222e-03\n", + "Dual infeasibility......: 7.9207857976975465e-07 7.9207857976975465e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 1.0163765993962152e-06 1.0163765993962152e-06\n", - "Overall NLP error.......: 1.0163765993962152e-06 1.0163765993962152e-06\n", + "Complementarity.........: 7.0389347768359540e-04 7.0389347768359540e-04\n", + "Overall NLP error.......: 7.0389347768359540e-04 7.0389347768359540e-04\n", "\n", "\n", "\n", @@ -5100,7 +5101,8 @@ "**************************************************\n", "\n", "Remaining in fixed mu mode.\n", - "Barrier Parameter: 8.071571e-09\n", + "Reducing mu to 6.6151651268281218e-06 in fixed mu mode. Tau becomes 9.9964760711896539e-01\n", + "Barrier Parameter: 6.615165e-06\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 5:\n", @@ -5114,28 +5116,29 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 2.354619e-14\n", + "residual_ratio = 2.556258e-16\n", "Factorization successful.\n", - "residual_ratio = 2.919430e-18\n", + "residual_ratio = 6.630917e-20\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 5:\n", "**************************************************\n", "\n", "--> Starting line search in iteration 5 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 2.758936E-23\n", + "minimal step size ALPHA_MIN = 3.042071E-22\n", "Starting checks for alpha (primal) = 1.00e+00\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.407e-17 (tol = 1.000e-07) r (rel) = 4.451e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.725e-17 (tol = 1.000e-07) r (rel) = 3.762e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", - " New values of barrier function = 3.5869742365835021e-04 (reference 3.7719746139176606e-04):\n", - " New values of constraint violation = 3.7142818253574697e-17 (reference 1.2391033227146346e-18):\n", + " New values of barrier function = 1.3101096689243513e-01 (reference 1.3118401226436072e-01):\n", + " New values of constraint violation = 1.3203888394978935e-16 (reference 1.0638733817514012e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -5148,11 +5151,11 @@ "*** Update HessianMatrix for Iteration 6:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 3.517985e-01\n", + "In limited-memory update, s_new_max is 4.347495e-03\n", "Limited-Memory test for skipping:\n", - " s^Ty = 7.002066e-06 snrm = 2.227109e+00 ynrm = 5.148725e-06\n", + " s^Ty = 1.301261e-07 snrm = 1.528487e-01 ynrm = 2.294377e-06\n", " Perform the update.\n", - "sigma (for B0) is 1.411702e-06\n", + "sigma (for B0) is 5.569817e-06\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -5161,33 +5164,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 6 1.3870512e-04 0.00e+00 4.90e-07 -8.1 3.52e-01 - 4.96e-01 1.00e+00f 1 Fy A\n", + " 6 1.4873534e-03 0.00e+00 1.53e-07 -5.2 4.35e-03 - 1.00e+00 1.00e+00f 1 Fy A\n", "\n", "**************************************************\n", "*** Beginning Iteration 6 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 8.0715713319459585e-09\n", - "Current fraction-to-the-boundary parameter tau = 9.9999597617832126e-01\n", + "Current barrier parameter mu = 6.6151651268281218e-06\n", + "Current fraction-to-the-boundary parameter tau = 9.9964760711896539e-01\n", "\n", - "||curr_x||_inf = 8.9170465467028626e-01\n", - "||curr_s||_inf = 2.0188616011069191e-06\n", + "||curr_x||_inf = 5.3065046069056843e-01\n", + "||curr_s||_inf = 1.7468387203296654e-05\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 3.8581185243911236e-03\n", - "||curr_z_L||_inf = 6.2329692853758058e-07\n", - "||curr_z_U||_inf = 9.6391865836145831e-07\n", - "||curr_v_L||_inf = 3.8581284486378211e-03\n", + "||curr_y_d||_inf = 2.6637405079446203e-01\n", + "||curr_z_L||_inf = 3.4926046848179427e-05\n", + "||curr_z_U||_inf = 1.3533433408582875e-05\n", + "||curr_v_L||_inf = 2.6637405046696827e-01\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 6:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.3870512353646771e-04 1.3870512353646771e-04\n", - "Dual infeasibility......: 4.8973753314882331e-07 4.8973753314882331e-07\n", + "Objective...............: 1.4873534193741112e-03 1.4873534193741112e-03\n", + "Dual infeasibility......: 1.5262773278245795e-07 1.5262773278245795e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 5.1594021882408290e-07 5.1594021882408290e-07\n", - "Overall NLP error.......: 5.1594021882408290e-07 5.1594021882408290e-07\n", + "Complementarity.........: 1.3808185585908299e-05 1.3808185585908299e-05\n", + "Overall NLP error.......: 1.3808185585908299e-05 1.3808185585908299e-05\n", "\n", "\n", "\n", @@ -5196,9 +5199,14 @@ "*** Update Barrier Parameter for Iteration 6:\n", "**************************************************\n", "\n", - "Switching back to free mu mode.\n", - "The current filter has 2 entries.\n", - "Solving the Primal Dual System for the affine step\n", + "Remaining in fixed mu mode.\n", + "Reducing mu to 1.7014180435354828e-08 in fixed mu mode. Tau becomes 9.9999338483487321e-01\n", + "Barrier Parameter: 1.701418e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 6:\n", + "**************************************************\n", + "\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -5207,20 +5215,9 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "Solving the Primal Dual System for the centering step\n", + "residual_ratio = 2.450310e-14\n", "Factorization successful.\n", - "Sigma = 3.100209e-01\n", - "Barrier parameter mu computed by oracle is 3.067659e-08\n", - "Barrier parameter mu after safeguards is 3.067659e-08\n", - "Barrier Parameter: 3.067659e-08\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 6:\n", - "**************************************************\n", - "\n", - "residual_ratio = 5.258785e-14\n", - "Factorization successful.\n", - "residual_ratio = 1.152811e-17\n", + "residual_ratio = 3.315759e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 6:\n", @@ -5230,17 +5227,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 2.219683E-22\n", - "Starting checks for alpha (primal) = 8.54e-01\n", + "minimal step size ALPHA_MIN = 1.562717E-22\n", + "Starting checks for alpha (primal) = 1.00e+00\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.152e-17 (tol = 1.000e-07) r (rel) = 4.193e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.027e-17 (tol = 1.000e-07) r (rel) = 3.057e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 8.539200e-01:\n", - " New values of barrier function = 9.2581153549170883e-04 (reference 9.7480164851010604e-04):\n", - " New values of constraint violation = 9.6876852243368838e-17 (reference 3.7142818253574697e-17):\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 1.4654763307904255e-03 (reference 1.8204876780084798e-03):\n", + " New values of constraint violation = 7.7818610930147081e-17 (reference 1.3203888394978935e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -5253,11 +5250,11 @@ "*** Update HessianMatrix for Iteration 7:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 3.127310e-01\n", + "In limited-memory update, s_new_max is 2.154055e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 7.392181e-06 snrm = 3.692675e+00 ynrm = 4.801309e-06\n", + " s^Ty = 1.338051e-04 snrm = 1.894901e+00 ynrm = 9.053622e-05\n", " Perform the update.\n", - "sigma (for B0) is 5.421138e-07\n", + "sigma (for B0) is 3.726488e-05\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -5266,33 +5263,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 7 1.5711830e-04 0.00e+00 1.14e-06 -7.5 3.66e-01 - 9.59e-01 8.54e-01f 1 sigma=3.10e-01 qf=12y A\n", + " 7 1.1281828e-03 0.00e+00 8.67e-06 -7.8 2.15e-01 - 6.66e-01 1.00e+00f 1 Fy A\n", "\n", "**************************************************\n", "*** Beginning Iteration 7 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 3.0676586125868578e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999948405978123e-01\n", + "Current barrier parameter mu = 1.7014180435354828e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999338483487321e-01\n", "\n", - "||curr_x||_inf = 9.9999993711555135e-01\n", - "||curr_s||_inf = 7.0507958477153509e-06\n", + "||curr_x||_inf = 7.0345580404420227e-01\n", + "||curr_s||_inf = 1.6338962714518393e-05\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 3.9119818195015423e-03\n", - "||curr_z_L||_inf = 4.7219400259507321e-07\n", - "||curr_z_U||_inf = 1.3301917285452153e-06\n", - "||curr_v_L||_inf = 3.9119987123555685e-03\n", + "||curr_y_d||_inf = 1.0097300016250953e-01\n", + "||curr_z_L||_inf = 1.3241411509933341e-05\n", + "||curr_z_U||_inf = 8.3312899001603065e-06\n", + "||curr_v_L||_inf = 1.0097310398607715e-01\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 7:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.5711830200493092e-04 1.5711830200493092e-04\n", - "Dual infeasibility......: 1.1361686079532655e-06 1.1361686079532655e-06\n", + "Objective...............: 1.1281827552494664e-03 1.1281827552494664e-03\n", + "Dual infeasibility......: 8.6695417539875768e-06 8.6695417539875768e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 9.8941265082249117e-08 9.8941265082249117e-08\n", - "Overall NLP error.......: 1.1361686079532655e-06 1.1361686079532655e-06\n", + "Complementarity.........: 4.6957154802993652e-06 4.6957154802993652e-06\n", + "Overall NLP error.......: 8.6695417539875768e-06 8.6695417539875768e-06\n", "\n", "\n", "\n", @@ -5301,8 +5298,8 @@ "*** Update Barrier Parameter for Iteration 7:\n", "**************************************************\n", "\n", - "Switching to fixed mu mode with mu = 2.7223188723064175e-08 and tau = 9.9999997277681130e-01.\n", - "Barrier Parameter: 2.722319e-08\n", + "Remaining in fixed mu mode.\n", + "Barrier Parameter: 1.701418e-08\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 7:\n", @@ -5316,29 +5313,28 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 6.733096e-16\n", + "residual_ratio = 8.819432e-16\n", "Factorization successful.\n", - "residual_ratio = 6.287097e-18\n", + "residual_ratio = 4.520145e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 7:\n", "**************************************************\n", "\n", "--> Starting line search in iteration 7 <--\n", - "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 6.301810E-24\n", - "Starting checks for alpha (primal) = 7.44e-01\n", + "minimal step size ALPHA_MIN = 6.744516E-23\n", + "Starting checks for alpha (primal) = 1.00e+00\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.155e-17 (tol = 1.000e-07) r (rel) = 4.196e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 2.888e-17 (tol = 1.000e-07) r (rel) = 2.917e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 7.438886e-01:\n", - " New values of barrier function = 8.1515920969987658e-04 (reference 8.3927637785765848e-04):\n", - " New values of constraint violation = 4.1275915519702056e-17 (reference 9.6876852243368838e-17):\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 1.0243513083792409e-03 (reference 1.4654763307904255e-03):\n", + " New values of constraint violation = 7.9968380550278495e-17 (reference 7.7818610930147081e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -5351,11 +5347,12 @@ "*** Update HessianMatrix for Iteration 8:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 3.859246e-01\n", + "In limited-memory update, s_new_max is 2.287016e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = -5.991395e-07 snrm = 4.413591e+00 ynrm = 3.490860e-06\n", - " Skip the update.\n", - "Number of successive iterations with skipping: 1\n", + " s^Ty = 2.616642e-04 snrm = 3.418644e+00 ynrm = 9.575779e-05\n", + " Perform the update.\n", + "sigma (for B0) is 2.238910e-05\n", + "Number of successive iterations with skipping: 0\n", "\n", "\n", "**************************************************\n", @@ -5363,33 +5360,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 8 1.5437821e-04 0.00e+00 1.01e-06 -7.6 5.19e-01 - 9.12e-01 7.44e-01f 1 Fy AWs\n", + " 8 6.7640116e-04 0.00e+00 3.43e-06 -7.8 2.29e-01 - 6.75e-01 1.00e+00f 1 Fy A\n", "\n", "**************************************************\n", "*** Beginning Iteration 8 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 2.7223188723064175e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "Current barrier parameter mu = 1.7014180435354828e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999338483487321e-01\n", "\n", - "||curr_x||_inf = 1.0000000013884498e+00\n", - "||curr_s||_inf = 8.4078287509868032e-06\n", + "||curr_x||_inf = 7.6318381280224079e-01\n", + "||curr_s||_inf = 1.4313794211271403e-05\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 2.9391150279237587e-03\n", - "||curr_z_L||_inf = 3.4194439061141347e-07\n", - "||curr_z_U||_inf = 1.1241262938801358e-06\n", - "||curr_v_L||_inf = 2.9391212138849486e-03\n", + "||curr_y_d||_inf = 4.1955954930602071e-02\n", + "||curr_z_L||_inf = 5.4613851005161841e-06\n", + "||curr_z_U||_inf = 3.8786726824001930e-06\n", + "||curr_v_L||_inf = 4.1956134144022668e-02\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 8:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.5437820644055507e-04 1.5437820644055507e-04\n", - "Dual infeasibility......: 1.0067114366532949e-06 1.0067114366532949e-06\n", + "Objective...............: 6.7640116270946242e-04 6.7640116270946242e-04\n", + "Dual infeasibility......: 3.4269238816625711e-06 3.4269238816625711e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 4.8520182360606052e-08 4.8520182360606052e-08\n", - "Overall NLP error.......: 1.0067114366532949e-06 1.0067114366532949e-06\n", + "Complementarity.........: 1.5510633745735747e-06 1.5510633745735747e-06\n", + "Overall NLP error.......: 3.4269238816625711e-06 3.4269238816625711e-06\n", "\n", "\n", "\n", @@ -5399,7 +5396,7 @@ "**************************************************\n", "\n", "Remaining in fixed mu mode.\n", - "Barrier Parameter: 2.722319e-08\n", + "Barrier Parameter: 1.701418e-08\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 8:\n", @@ -5413,9 +5410,9 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 1.134608e-15\n", + "residual_ratio = 6.428517e-17\n", "Factorization successful.\n", - "residual_ratio = 1.199844e-18\n", + "residual_ratio = 1.552143e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 8:\n", @@ -5424,17 +5421,17 @@ "--> Starting line search in iteration 8 <--\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 5.080217E-26\n", - "Starting checks for alpha (primal) = 6.25e-01\n", + "minimal step size ALPHA_MIN = 1.119770E-22\n", + "Starting checks for alpha (primal) = 1.00e+00\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.460e-17 (tol = 1.000e-07) r (rel) = 4.505e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.086e-17 (tol = 1.000e-07) r (rel) = 4.127e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 6.251177e-01:\n", - " New values of barrier function = 8.0443889257491739e-04 (reference 8.1515920969987658e-04):\n", - " New values of constraint violation = 3.9254894907553295e-17 (reference 4.1275915519702056e-17):\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 7.6709862687967948e-04 (reference 1.0243513083792409e-03):\n", + " New values of constraint violation = 5.2873490633507936e-17 (reference 7.9968380550278495e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -5447,11 +5444,11 @@ "*** Update HessianMatrix for Iteration 9:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 3.811753e-01\n", + "In limited-memory update, s_new_max is 2.264424e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 2.356081e-06 snrm = 2.798376e+00 ynrm = 3.566785e-06\n", + " s^Ty = 1.812850e-04 snrm = 3.825208e+00 ynrm = 5.801809e-05\n", " Perform the update.\n", - "sigma (for B0) is 3.008694e-07\n", + "sigma (for B0) is 1.238944e-05\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -5460,33 +5457,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 9 1.5295135e-04 0.00e+00 8.44e-07 -7.6 6.10e-01 - 6.47e-01 6.25e-01f 1 Fy A\n", + " 9 4.0408975e-04 0.00e+00 2.00e-06 -7.8 2.26e-01 - 6.77e-01 1.00e+00f 1 Fy A\n", "\n", "**************************************************\n", "*** Beginning Iteration 9 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 2.7223188723064175e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "Current barrier parameter mu = 1.7014180435354828e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999338483487321e-01\n", "\n", - "||curr_x||_inf = 1.0000000014545833e+00\n", - "||curr_s||_inf = 9.1761617469650529e-06\n", + "||curr_x||_inf = 8.3610688521683940e-01\n", + "||curr_s||_inf = 1.2198011328926200e-05\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 2.8522554071847019e-03\n", - "||curr_z_L||_inf = 3.3900761536175374e-07\n", - "||curr_z_U||_inf = 9.5186430741765041e-07\n", - "||curr_v_L||_inf = 2.8522628154606087e-03\n", + "||curr_y_d||_inf = 1.8543169942774943e-02\n", + "||curr_z_L||_inf = 2.4280079495006772e-06\n", + "||curr_z_U||_inf = 2.1094055843836140e-06\n", + "||curr_v_L||_inf = 1.8543289387852560e-02\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 9:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.5295135006259956e-04 1.5295135006259956e-04\n", - "Dual infeasibility......: 8.4449396122111269e-07 8.4449396122111269e-07\n", + "Objective...............: 4.0408974679138210e-04 4.0408974679138210e-04\n", + "Dual infeasibility......: 1.9989240045191167e-06 1.9989240045191167e-06\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 3.4444459929521884e-08 3.4444459929521884e-08\n", - "Overall NLP error.......: 8.4449396122111269e-07 8.4449396122111269e-07\n", + "Complementarity.........: 5.0899224309982139e-07 5.0899224309982139e-07\n", + "Overall NLP error.......: 1.9989240045191167e-06 1.9989240045191167e-06\n", "\n", "\n", "\n", @@ -5496,7 +5493,7 @@ "**************************************************\n", "\n", "Remaining in fixed mu mode.\n", - "Barrier Parameter: 2.722319e-08\n", + "Barrier Parameter: 1.701418e-08\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 9:\n", @@ -5510,9 +5507,9 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 5.377459e-16\n", + "residual_ratio = 5.058862e-16\n", "Factorization successful.\n", - "residual_ratio = 5.901217e-18\n", + "residual_ratio = 3.842626e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 9:\n", @@ -5521,17 +5518,17 @@ "--> Starting line search in iteration 9 <--\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 3.744540E-26\n", - "Starting checks for alpha (primal) = 6.42e-01\n", + "minimal step size ALPHA_MIN = 1.658964E-22\n", + "Starting checks for alpha (primal) = 1.00e+00\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.802e-17 (tol = 1.000e-07) r (rel) = 4.850e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 3.928e-17 (tol = 1.000e-07) r (rel) = 3.967e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 6.416571e-01:\n", - " New values of barrier function = 7.9758113511439667e-04 (reference 8.0443889257491739e-04):\n", - " New values of constraint violation = 1.2704138956098898e-16 (reference 3.9254894907553295e-17):\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 6.5082081308500414e-04 (reference 7.6709862687967948e-04):\n", + " New values of constraint violation = 1.0839988845781634e-16 (reference 5.2873490633507936e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -5544,11 +5541,11 @@ "*** Update HessianMatrix for Iteration 10:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 4.232701e-01\n", + "In limited-memory update, s_new_max is 2.203736e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 4.041249e-06 snrm = 3.861719e+00 ynrm = 4.579124e-06\n", + " s^Ty = 7.598389e-05 snrm = 3.435318e+00 ynrm = 2.799207e-05\n", " Perform the update.\n", - "sigma (for B0) is 2.709906e-07\n", + "sigma (for B0) is 6.438544e-06\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -5557,33 +5554,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 10 1.4843773e-04 0.00e+00 2.61e-06 -7.6 6.60e-01 - 9.46e-01 6.42e-01f 1 Fy A\n", + " 10 2.7242392e-04 0.00e+00 6.21e-07 -7.8 2.20e-01 - 7.38e-01 1.00e+00f 1 Fy A\n", "\n", "**************************************************\n", "*** Beginning Iteration 10 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 2.7223188723064175e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "Current barrier parameter mu = 1.7014180435354828e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999338483487321e-01\n", "\n", - "||curr_x||_inf = 1.0000000000589839e+00\n", - "||curr_s||_inf = 9.7743300618451697e-06\n", + "||curr_x||_inf = 9.4014453591154956e-01\n", + "||curr_s||_inf = 1.0673024993193405e-05\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 2.6837173175875088e-03\n", - "||curr_z_L||_inf = 3.2134023460517586e-07\n", - "||curr_z_U||_inf = 5.4028900333374227e-07\n", - "||curr_v_L||_inf = 2.6837267051078649e-03\n", + "||curr_y_d||_inf = 7.6009501801899424e-03\n", + "||curr_z_L||_inf = 9.3432526052621185e-07\n", + "||curr_z_U||_inf = 1.3140704048849144e-06\n", + "||curr_v_L||_inf = 7.6009746070818448e-03\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 10:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.4843773331278846e-04 1.4843773331278846e-04\n", - "Dual infeasibility......: 2.6085318329257540e-06 2.6085318329257540e-06\n", + "Objective...............: 2.7242392407408265e-04 2.7242392407408265e-04\n", + "Dual infeasibility......: 6.2094039813404981e-07 6.2094039813404981e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 1.4249962233056421e-07 1.4249962233056421e-07\n", - "Overall NLP error.......: 2.6085318329257540e-06 2.6085318329257540e-06\n", + "Complementarity.........: 1.5917165117875071e-07 1.5917165117875071e-07\n", + "Overall NLP error.......: 6.2094039813404981e-07 6.2094039813404981e-07\n", "\n", "\n", "\n", @@ -5593,7 +5590,7 @@ "**************************************************\n", "\n", "Remaining in fixed mu mode.\n", - "Barrier Parameter: 2.722319e-08\n", + "Barrier Parameter: 1.701418e-08\n", "\n", "**************************************************\n", "*** Solving the Primal Dual System for Iteration 10:\n", @@ -5607,9 +5604,9 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 6.255437e-16\n", + "residual_ratio = 4.709658e-16\n", "Factorization successful.\n", - "residual_ratio = 9.740116e-19\n", + "residual_ratio = 1.051788e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 10:\n", @@ -5618,17 +5615,17 @@ "--> Starting line search in iteration 10 <--\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.475848E-25\n", - "Starting checks for alpha (primal) = 4.10e-01\n", + "minimal step size ALPHA_MIN = 6.601801E-22\n", + "Starting checks for alpha (primal) = 1.00e+00\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.993e-17 (tol = 1.000e-07) r (rel) = 5.043e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.378e-17 (tol = 1.000e-07) r (rel) = 4.421e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 4.100400e-01:\n", - " New values of barrier function = 7.9513275660310020e-04 (reference 7.9758113511439667e-04):\n", - " New values of constraint violation = 6.6509027018407663e-18 (reference 1.2704138956098898e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", + " New values of barrier function = 5.9273716445835674e-04 (reference 6.5082081308500414e-04):\n", + " New values of constraint violation = 1.1466623820160365e-16 (reference 1.0839988845781634e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -5641,11 +5638,11 @@ "*** Update HessianMatrix for Iteration 11:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 4.128123e-01\n", + "In limited-memory update, s_new_max is 2.164711e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 2.408930e-06 snrm = 1.492650e+00 ynrm = 3.883168e-06\n", + " s^Ty = 3.957596e-05 snrm = 3.529055e+00 ynrm = 1.467257e-05\n", " Perform the update.\n", - "sigma (for B0) is 1.081205e-06\n", + "sigma (for B0) is 3.177712e-06\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -5654,33 +5651,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 11 1.4703404e-04 0.00e+00 1.23e-06 -7.6 1.01e+00 - 5.59e-01 4.10e-01f 1 Fy A\n", + " 11 1.9527707e-04 0.00e+00 4.45e-07 -7.8 2.16e-01 - 7.91e-01 1.00e+00f 1 Fy A\n", "\n", "**************************************************\n", "*** Beginning Iteration 11 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 2.7223188723064175e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "Current barrier parameter mu = 1.7014180435354828e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999338483487321e-01\n", "\n", - "||curr_x||_inf = 1.0000000014065347e+00\n", - "||curr_s||_inf = 9.9003405003506122e-06\n", + "||curr_x||_inf = 9.8028939099711931e-01\n", + "||curr_s||_inf = 9.2599494043480248e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 2.6917071891186463e-03\n", - "||curr_z_L||_inf = 2.9938493621947088e-07\n", - "||curr_z_U||_inf = 4.5481733906074267e-07\n", - "||curr_v_L||_inf = 2.6917186962841515e-03\n", + "||curr_y_d||_inf = 3.6455192022053213e-03\n", + "||curr_z_L||_inf = 4.8196587118703776e-07\n", + "||curr_z_U||_inf = 1.1127746763234580e-06\n", + "||curr_v_L||_inf = 3.6455212099056264e-03\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 11:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.4703403987189042e-04 1.4703403987189042e-04\n", - "Dual infeasibility......: 1.2323063397704064e-06 1.2323063397704064e-06\n", + "Objective...............: 1.9527707084811123e-04 1.9527707084811123e-04\n", + "Dual infeasibility......: 4.4465084158661657e-07 4.4465084158661657e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 7.5551841170247104e-08 7.5551841170247104e-08\n", - "Overall NLP error.......: 1.2323063397704064e-06 1.2323063397704064e-06\n", + "Complementarity.........: 5.5003288128345442e-08 5.5003288128345442e-08\n", + "Overall NLP error.......: 4.4465084158661657e-07 4.4465084158661657e-07\n", "\n", "\n", "\n", @@ -5689,13 +5686,9 @@ "*** Update Barrier Parameter for Iteration 11:\n", "**************************************************\n", "\n", - "Remaining in fixed mu mode.\n", - "Barrier Parameter: 2.722319e-08\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 11:\n", - "**************************************************\n", - "\n", + "Switching back to free mu mode.\n", + "The current filter has 2 entries.\n", + "Solving the Primal Dual System for the affine step\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -5704,28 +5697,40 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 5.749126e-16\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 5.016691e-01\n", + "Barrier parameter mu computed by oracle is 1.575817e-08\n", + "Barrier parameter mu after safeguards is 1.575817e-08\n", + "Barrier Parameter: 1.575817e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 11:\n", + "**************************************************\n", + "\n", + "residual_ratio = 1.135916e-16\n", "Factorization successful.\n", - "residual_ratio = 1.750567e-18\n", + "residual_ratio = 2.293659e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 11:\n", "**************************************************\n", "\n", "--> Starting line search in iteration 11 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 7.783225E-27\n", - "Starting checks for alpha (primal) = 6.46e-01\n", + "minimal step size ALPHA_MIN = 1.558537E-21\n", + "Starting checks for alpha (primal) = 8.10e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.634e-17 (tol = 1.000e-07) r (rel) = 4.680e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.590e-17 (tol = 1.000e-07) r (rel) = 4.636e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 6.461774e-01:\n", - " New values of barrier function = 7.9183800201424468e-04 (reference 7.9513275660310020e-04):\n", - " New values of constraint violation = 6.2565241615991640e-17 (reference 6.6509027018407663e-18):\n", + "Checking acceptability for trial step size alpha_primal_test= 8.099143e-01:\n", + " New values of barrier function = 5.4196131083041141e-04 (reference 5.6339617773549589e-04):\n", + " New values of constraint violation = 7.9231461886167254e-18 (reference 1.1466623820160365e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -5738,11 +5743,11 @@ "*** Update HessianMatrix for Iteration 12:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 3.533840e-01\n", + "In limited-memory update, s_new_max is 2.226766e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 6.292786e-07 snrm = 1.457286e+00 ynrm = 1.834871e-06\n", + " s^Ty = 1.152956e-05 snrm = 2.751114e+00 ynrm = 6.415046e-06\n", " Perform the update.\n", - "sigma (for B0) is 2.963147e-07\n", + "sigma (for B0) is 1.523335e-06\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -5751,33 +5756,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 12 1.4705477e-04 0.00e+00 7.78e-07 -7.6 5.47e-01 - 7.61e-01 6.46e-01f 1 Fy A\n", + " 12 1.5958699e-04 0.00e+00 4.45e-07 -7.8 2.75e-01 - 9.57e-01 8.10e-01f 1 sigma=5.02e-01 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 12 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 2.7223188723064175e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "Current barrier parameter mu = 1.5758172976292438e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999955534915841e-01\n", "\n", - "||curr_x||_inf = 1.0000000003797616e+00\n", - "||curr_s||_inf = 1.0216885760132184e-05\n", + "||curr_x||_inf = 9.9999997055548617e-01\n", + "||curr_s||_inf = 8.4358719887798899e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 2.6324848272244804e-03\n", - "||curr_z_L||_inf = 3.0766355124056562e-07\n", - "||curr_z_U||_inf = 6.8432984829646557e-07\n", - "||curr_v_L||_inf = 2.6324945713303931e-03\n", + "||curr_y_d||_inf = 2.1661673739142989e-03\n", + "||curr_z_L||_inf = 2.6202759839735379e-07\n", + "||curr_z_U||_inf = 9.9235312665097110e-07\n", + "||curr_v_L||_inf = 2.1661640754923489e-03\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 12:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.4705477146526895e-04 1.4705477146526895e-04\n", - "Dual infeasibility......: 7.7832444069418973e-07 7.7832444069418973e-07\n", + "Objective...............: 1.5958699244895272e-04 1.5958699244895272e-04\n", + "Dual infeasibility......: 4.4549012189946007e-07 4.4549012189946007e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 5.9639062026560725e-08 5.9639062026560725e-08\n", - "Overall NLP error.......: 7.7832444069418973e-07 7.7832444069418973e-07\n", + "Complementarity.........: 2.4593570994306414e-08 2.4593570994306414e-08\n", + "Overall NLP error.......: 4.4549012189946007e-07 4.4549012189946007e-07\n", "\n", "\n", "\n", @@ -5786,13 +5791,9 @@ "*** Update Barrier Parameter for Iteration 12:\n", "**************************************************\n", "\n", - "Remaining in fixed mu mode.\n", - "Barrier Parameter: 2.722319e-08\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 12:\n", - "**************************************************\n", - "\n", + "Staying in free mu mode.\n", + "The current filter has 2 entries.\n", + "Solving the Primal Dual System for the affine step\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -5801,28 +5802,40 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 4.623051e-16\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 6.721865e-01\n", + "Barrier parameter mu computed by oracle is 1.152566e-08\n", + "Barrier parameter mu after safeguards is 1.152566e-08\n", + "Barrier Parameter: 1.152566e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 12:\n", + "**************************************************\n", + "\n", + "residual_ratio = 6.393566e-16\n", "Factorization successful.\n", - "residual_ratio = 1.885439e-18\n", + "residual_ratio = 1.083293e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 12:\n", "**************************************************\n", "\n", "--> Starting line search in iteration 12 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 5.895139E-26\n", - "Starting checks for alpha (primal) = 4.60e-01\n", + "minimal step size ALPHA_MIN = 8.227417E-25\n", + "Starting checks for alpha (primal) = 5.24e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.619e-17 (tol = 1.000e-07) r (rel) = 4.665e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.631e-17 (tol = 1.000e-07) r (rel) = 4.677e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 4.595809e-01:\n", - " New values of barrier function = 7.8850096602701091e-04 (reference 7.9183800201424468e-04):\n", - " New values of constraint violation = 1.3349239248727773e-17 (reference 6.2565241615991640e-17):\n", + "Checking acceptability for trial step size alpha_primal_test= 5.235646e-01:\n", + " New values of barrier function = 4.2726855941740125e-04 (reference 4.3925886798422908e-04):\n", + " New values of constraint violation = 5.6405618023558368e-17 (reference 7.9231461886167254e-18):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -5835,11 +5848,11 @@ "*** Update HessianMatrix for Iteration 13:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 3.850562e-01\n", + "In limited-memory update, s_new_max is 2.822359e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 1.704929e-06 snrm = 2.086038e+00 ynrm = 2.163729e-06\n", + " s^Ty = 5.222832e-06 snrm = 2.470537e+00 ynrm = 3.651876e-06\n", " Perform the update.\n", - "sigma (for B0) is 3.917977e-07\n", + "sigma (for B0) is 8.557038e-07\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -5848,33 +5861,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 13 1.4939779e-04 0.00e+00 5.16e-07 -7.6 8.38e-01 - 8.94e-01 4.60e-01f 1 Fy A\n", + " 13 1.3948467e-04 0.00e+00 4.86e-07 -7.9 5.39e-01 - 9.74e-01 5.24e-01f 1 sigma=6.72e-01 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 13 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 2.7223188723064175e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "Current barrier parameter mu = 1.1525663674128999e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999955450987810e-01\n", "\n", - "||curr_x||_inf = 1.0000000022667506e+00\n", - "||curr_s||_inf = 1.0637360960505911e-05\n", + "||curr_x||_inf = 9.9999996479060049e-01\n", + "||curr_s||_inf = 7.8083502085814986e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 2.4481980658413128e-03\n", - "||curr_z_L||_inf = 2.9094526043824776e-07\n", - "||curr_z_U||_inf = 6.6248234625351159e-07\n", - "||curr_v_L||_inf = 2.4482004857085925e-03\n", + "||curr_y_d||_inf = 1.6847635334661124e-03\n", + "||curr_z_L||_inf = 1.7914171487451332e-07\n", + "||curr_z_U||_inf = 7.6500365446862860e-07\n", + "||curr_v_L||_inf = 1.6847664142600424e-03\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 13:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.4939779200170430e-04 1.4939779200170430e-04\n", - "Dual infeasibility......: 5.1614828927523612e-07 5.1614828927523612e-07\n", + "Objective...............: 1.3948467146953391e-04 1.3948467146953391e-04\n", + "Dual infeasibility......: 4.8578582202864654e-07 4.8578582202864654e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 8.5439547895813522e-08 8.5439547895813522e-08\n", - "Overall NLP error.......: 5.1614828927523612e-07 5.1614828927523612e-07\n", + "Complementarity.........: 1.7595471822780307e-08 1.7595471822780307e-08\n", + "Overall NLP error.......: 4.8578582202864654e-07 4.8578582202864654e-07\n", "\n", "\n", "\n", @@ -5883,13 +5896,9 @@ "*** Update Barrier Parameter for Iteration 13:\n", "**************************************************\n", "\n", - "Remaining in fixed mu mode.\n", - "Barrier Parameter: 2.722319e-08\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 13:\n", - "**************************************************\n", - "\n", + "Staying in free mu mode.\n", + "The current filter has 3 entries.\n", + "Solving the Primal Dual System for the affine step\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -5898,28 +5907,40 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 1.465565e-15\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 7.855842e-01\n", + "Barrier parameter mu computed by oracle is 9.615269e-09\n", + "Barrier parameter mu after safeguards is 9.615269e-09\n", + "Barrier Parameter: 9.615269e-09\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 13:\n", + "**************************************************\n", + "\n", + "residual_ratio = 3.610841e-16\n", "Factorization successful.\n", - "residual_ratio = 7.328570e-19\n", + "residual_ratio = 3.541318e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 13:\n", "**************************************************\n", "\n", "--> Starting line search in iteration 13 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.834115E-26\n", - "Starting checks for alpha (primal) = 1.00e+00\n", + "minimal step size ALPHA_MIN = 6.611108E-24\n", + "Starting checks for alpha (primal) = 4.55e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.924e-17 (tol = 1.000e-07) r (rel) = 4.973e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.508e-17 (tol = 1.000e-07) r (rel) = 4.553e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", - " New values of barrier function = 7.8610156016912960e-04 (reference 7.8850096602701091e-04):\n", - " New values of constraint violation = 1.1089016532274398e-16 (reference 1.3349239248727773e-17):\n", + "Checking acceptability for trial step size alpha_primal_test= 4.548408e-01:\n", + " New values of barrier function = 3.7245782496073300e-04 (reference 3.7956798930867018e-04):\n", + " New values of constraint violation = 1.8598725939336300e-16 (reference 5.6405618023558368e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -5932,11 +5953,11 @@ "*** Update HessianMatrix for Iteration 14:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 3.569959e-01\n", + "In limited-memory update, s_new_max is 4.721130e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 1.792604e-06 snrm = 2.688494e+00 ynrm = 2.847658e-06\n", + " s^Ty = 4.242770e-06 snrm = 2.565256e+00 ynrm = 2.875115e-06\n", " Perform the update.\n", - "sigma (for B0) is 2.480082e-07\n", + "sigma (for B0) is 6.447449e-07\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -5945,33 +5966,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 14 1.5362607e-04 0.00e+00 9.18e-07 -7.6 3.57e-01 - 7.89e-01 1.00e+00f 1 Fy A\n", + " 14 1.2621178e-04 0.00e+00 3.96e-07 -8.0 1.04e+00 - 9.58e-01 4.55e-01f 1 sigma=7.86e-01 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 14 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 2.7223188723064175e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "Current barrier parameter mu = 9.6152692734631776e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999951421417799e-01\n", "\n", - "||curr_x||_inf = 9.7187915272562730e-01\n", - "||curr_s||_inf = 1.1275987654546383e-05\n", + "||curr_x||_inf = 9.9999978065406925e-01\n", + "||curr_s||_inf = 7.4034442845067021e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 2.4180325983607903e-03\n", - "||curr_z_L||_inf = 2.7759047785231994e-07\n", - "||curr_z_U||_inf = 7.4115007882390933e-07\n", - "||curr_v_L||_inf = 2.4180311918402231e-03\n", + "||curr_y_d||_inf = 1.4328379383057319e-03\n", + "||curr_z_L||_inf = 1.5608002472099507e-07\n", + "||curr_z_U||_inf = 6.1740132502512337e-07\n", + "||curr_v_L||_inf = 1.4328426458293669e-03\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 14:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.5362606843249261e-04 1.5362606843249261e-04\n", - "Dual infeasibility......: 9.1812912389470822e-07 9.1812912389470822e-07\n", + "Objective...............: 1.2621178334637219e-04 1.2621178334637219e-04\n", + "Dual infeasibility......: 3.9574454545733103e-07 3.9574454545733103e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 4.2377847446537874e-08 4.2377847446537874e-08\n", - "Overall NLP error.......: 9.1812912389470822e-07 9.1812912389470822e-07\n", + "Complementarity.........: 1.7440758031090139e-08 1.7440758031090139e-08\n", + "Overall NLP error.......: 3.9574454545733103e-07 3.9574454545733103e-07\n", "\n", "\n", "\n", @@ -5980,13 +6001,9 @@ "*** Update Barrier Parameter for Iteration 14:\n", "**************************************************\n", "\n", - "Remaining in fixed mu mode.\n", - "Barrier Parameter: 2.722319e-08\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 14:\n", - "**************************************************\n", - "\n", + "Staying in free mu mode.\n", + "The current filter has 4 entries.\n", + "Solving the Primal Dual System for the affine step\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -5995,28 +6012,40 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 5.046691e-17\n", + "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "residual_ratio = 1.136027e-18\n", + "Sigma = 7.435140e-01\n", + "Barrier parameter mu computed by oracle is 7.551049e-09\n", + "Barrier parameter mu after safeguards is 7.551049e-09\n", + "Barrier Parameter: 7.551049e-09\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 14:\n", + "**************************************************\n", + "\n", + "residual_ratio = 3.488421e-16\n", + "Factorization successful.\n", + "residual_ratio = 5.261785e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 14:\n", "**************************************************\n", "\n", "--> Starting line search in iteration 14 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 1.776198E-20\n", - "Starting checks for alpha (primal) = 1.00e+00\n", + "minimal step size ALPHA_MIN = 3.214483E-23\n", + "Starting checks for alpha (primal) = 5.26e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.670e-17 (tol = 1.000e-07) r (rel) = 4.716e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.659e-17 (tol = 1.000e-07) r (rel) = 4.706e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", - " New values of barrier function = 7.8481025801009109e-04 (reference 7.8610156016912960e-04):\n", - " New values of constraint violation = 1.8214427091167024e-16 (reference 1.1089016532274398e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 5.256877e-01:\n", + " New values of barrier function = 3.1554774277217667e-04 (reference 3.1959335769371814e-04):\n", + " New values of constraint violation = 1.6286495399921511e-16 (reference 1.8598725939336300e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -6029,11 +6058,11 @@ "*** Update HessianMatrix for Iteration 15:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 3.940773e-01\n", + "In limited-memory update, s_new_max is 5.254068e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 1.671400e-06 snrm = 1.474771e+00 ynrm = 2.364038e-06\n", + " s^Ty = 2.875557e-06 snrm = 2.448011e+00 ynrm = 2.082392e-06\n", " Perform the update.\n", - "sigma (for B0) is 7.684777e-07\n", + "sigma (for B0) is 4.798385e-07\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -6042,33 +6071,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 15 1.5476349e-04 0.00e+00 5.01e-07 -7.6 3.94e-01 - 1.00e+00 1.00e+00f 1 Fy A\n", + " 15 1.1765070e-04 0.00e+00 2.67e-07 -8.1 9.99e-01 - 9.54e-01 5.26e-01f 1 sigma=7.44e-01 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 15 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 2.7223188723064175e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "Current barrier parameter mu = 7.5510489333588987e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999960425545453e-01\n", "\n", - "||curr_x||_inf = 9.5330070549010015e-01\n", - "||curr_s||_inf = 1.1406858507713101e-05\n", + "||curr_x||_inf = 9.9999980207303851e-01\n", + "||curr_s||_inf = 7.0830367970601319e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 2.3840871588389442e-03\n", - "||curr_z_L||_inf = 2.6881424074940881e-07\n", - "||curr_z_U||_inf = 5.8414507488241126e-07\n", - "||curr_v_L||_inf = 2.3840836745169887e-03\n", + "||curr_y_d||_inf = 1.1499145754809742e-03\n", + "||curr_z_L||_inf = 1.3396638055900011e-07\n", + "||curr_z_U||_inf = 5.2970059904212701e-07\n", + "||curr_v_L||_inf = 1.1499172528534000e-03\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 15:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.5476349153543129e-04 1.5476349153543129e-04\n", - "Dual infeasibility......: 5.0054583110348449e-07 5.0054583110348449e-07\n", + "Objective...............: 1.1765070036925208e-04 1.1765070036925208e-04\n", + "Dual infeasibility......: 2.6653865673269954e-07 2.6653865673269954e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 5.1864101498357542e-08 5.1864101498357542e-08\n", - "Overall NLP error.......: 5.0054583110348449e-07 5.0054583110348449e-07\n", + "Complementarity.........: 1.1173916830442450e-08 1.1173916830442450e-08\n", + "Overall NLP error.......: 2.6653865673269954e-07 2.6653865673269954e-07\n", "\n", "\n", "\n", @@ -6077,13 +6106,9 @@ "*** Update Barrier Parameter for Iteration 15:\n", "**************************************************\n", "\n", - "Remaining in fixed mu mode.\n", - "Barrier Parameter: 2.722319e-08\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 15:\n", - "**************************************************\n", - "\n", + "Staying in free mu mode.\n", + "The current filter has 4 entries.\n", + "Solving the Primal Dual System for the affine step\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -6092,28 +6117,40 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 7.509304e-17\n", + "Solving the Primal Dual System for the centering step\n", + "Factorization successful.\n", + "Sigma = 8.116906e-01\n", + "Barrier parameter mu computed by oracle is 6.445589e-09\n", + "Barrier parameter mu after safeguards is 6.445589e-09\n", + "Barrier Parameter: 6.445589e-09\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 15:\n", + "**************************************************\n", + "\n", + "residual_ratio = 7.993938e-17\n", "Factorization successful.\n", - "residual_ratio = 7.412612e-19\n", + "residual_ratio = 5.421888e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 15:\n", "**************************************************\n", "\n", "--> Starting line search in iteration 15 <--\n", + "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 3.363582E-20\n", - "Starting checks for alpha (primal) = 1.00e+00\n", + "minimal step size ALPHA_MIN = 2.268156E-23\n", + "Starting checks for alpha (primal) = 6.04e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.929e-17 (tol = 1.000e-07) r (rel) = 4.978e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.458e-17 (tol = 1.000e-07) r (rel) = 4.503e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 1.000000e+00:\n", - " New values of barrier function = 7.8341992465810837e-04 (reference 7.8481025801009109e-04):\n", - " New values of constraint violation = 5.3949222476520897e-17 (reference 1.8214427091167024e-16):\n", + "Checking acceptability for trial step size alpha_primal_test= 6.039029e-01:\n", + " New values of barrier function = 2.8417270300612387e-04 (reference 2.8657596712624201e-04):\n", + " New values of constraint violation = 6.0164750243472953e-17 (reference 1.6286495399921511e-16):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -6126,11 +6163,11 @@ "*** Update HessianMatrix for Iteration 16:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 1.981181e-01\n", + "In limited-memory update, s_new_max is 4.934272e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 1.179498e-06 snrm = 1.429086e+00 ynrm = 1.935569e-06\n", + " s^Ty = 1.791212e-06 snrm = 2.365173e+00 ynrm = 1.571980e-06\n", " Perform the update.\n", - "sigma (for B0) is 5.775381e-07\n", + "sigma (for B0) is 3.201999e-07\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -6139,33 +6176,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 16 1.5154537e-04 0.00e+00 2.11e-07 -7.6 1.98e-01 - 1.00e+00 1.00e+00f 1 Fy A\n", + " 16 1.1162605e-04 0.00e+00 2.14e-07 -8.2 8.17e-01 - 9.83e-01 6.04e-01f 1 sigma=8.12e-01 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 16 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 2.7223188723064175e-08\n", - "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "Current barrier parameter mu = 6.4455887762367371e-09\n", + "Current fraction-to-the-boundary parameter tau = 9.9999973346134330e-01\n", "\n", - "||curr_x||_inf = 9.6172656812131052e-01\n", - "||curr_s||_inf = 1.1419635363106410e-05\n", + "||curr_x||_inf = 9.9999989002993517e-01\n", + "||curr_s||_inf = 6.8155430864811444e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 2.3818039943382650e-03\n", - "||curr_z_L||_inf = 2.8810337578962471e-07\n", - "||curr_z_U||_inf = 6.2851405946942329e-07\n", - "||curr_v_L||_inf = 2.3818047330917014e-03\n", + "||curr_y_d||_inf = 9.8338734956333017e-04\n", + "||curr_z_L||_inf = 1.1945892901526540e-07\n", + "||curr_z_U||_inf = 4.7281830816883926e-07\n", + "||curr_v_L||_inf = 9.8338980961508470e-04\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 16:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.5154537407026223e-04 1.5154537407026223e-04\n", - "Dual infeasibility......: 2.1135137784591342e-07 2.1135137784591342e-07\n", + "Objective...............: 1.1162604642663981e-04 1.1162604642663981e-04\n", + "Dual infeasibility......: 2.1386151081315249e-07 2.1386151081315249e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 2.8616723661402496e-08 2.8616723661402496e-08\n", - "Overall NLP error.......: 2.1135137784591342e-07 2.1135137784591342e-07\n", + "Complementarity.........: 1.1523551347103779e-08 1.1523551347103779e-08\n", + "Overall NLP error.......: 2.1386151081315249e-07 2.1386151081315249e-07\n", "\n", "\n", "\n", @@ -6174,14 +6211,9 @@ "*** Update Barrier Parameter for Iteration 16:\n", "**************************************************\n", "\n", - "Remaining in fixed mu mode.\n", - "Reducing mu to 9.0909090909090910e-10 in fixed mu mode. Tau becomes 9.9999997277681130e-01\n", - "Barrier Parameter: 9.090909e-10\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 16:\n", - "**************************************************\n", - "\n", + "Staying in free mu mode.\n", + "The current filter has 4 entries.\n", + "Solving the Primal Dual System for the affine step\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -6190,9 +6222,20 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "residual_ratio = 1.966589e-15\n", + "Solving the Primal Dual System for the centering step\n", "Factorization successful.\n", - "residual_ratio = 8.261866e-18\n", + "Sigma = 3.912269e+00\n", + "Barrier parameter mu computed by oracle is 2.587262e-08\n", + "Barrier parameter mu after safeguards is 2.587262e-08\n", + "Barrier Parameter: 2.587262e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 16:\n", + "**************************************************\n", + "\n", + "residual_ratio = 4.525944e-16\n", + "Factorization successful.\n", + "residual_ratio = 2.157766e-17\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 16:\n", @@ -6202,17 +6245,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 4.206227E-22\n", - "Starting checks for alpha (primal) = 4.96e-01\n", + "minimal step size ALPHA_MIN = 2.607848E-25\n", + "Starting checks for alpha (primal) = 2.48e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 5.105e-17 (tol = 1.000e-07) r (rel) = 5.155e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.405e-17 (tol = 1.000e-07) r (rel) = 4.448e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 4.955224e-01:\n", - " New values of barrier function = 1.4810246367516608e-04 (reference 1.7264618689086475e-04):\n", - " New values of constraint violation = 8.8040604537611977e-18 (reference 5.3949222476520897e-17):\n", + "Checking acceptability for trial step size alpha_primal_test= 2.484775e-01:\n", + " New values of barrier function = 7.8601601938563411e-04 (reference 8.0422906148101391e-04):\n", + " New values of constraint violation = 1.1411227865409934e-17 (reference 6.0164750243472953e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -6225,11 +6268,11 @@ "*** Update HessianMatrix for Iteration 17:\n", "**************************************************\n", "\n", - "In limited-memory update, s_new_max is 2.759149e-01\n", + "In limited-memory update, s_new_max is 3.762147e-01\n", "Limited-Memory test for skipping:\n", - " s^Ty = 1.212184e-05 snrm = 2.951106e+00 ynrm = 5.285112e-06\n", + " s^Ty = 1.287324e-06 snrm = 2.435939e+00 ynrm = 2.102636e-06\n", " Perform the update.\n", - "sigma (for B0) is 1.391871e-06\n", + "sigma (for B0) is 2.169478e-07\n", "Number of successive iterations with skipping: 0\n", "\n", "\n", @@ -6238,33 +6281,33 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 17 1.2565776e-04 0.00e+00 2.55e-07 -9.0 5.57e-01 - 4.03e-01 4.96e-01f 1 Fy A\n", + " 17 1.2176570e-04 0.00e+00 2.90e-07 -7.6 1.51e+00 - 4.41e-01 2.48e-01f 1 sigma=3.91e+00 qf=12y A\n", "\n", "**************************************************\n", "*** Beginning Iteration 17 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 9.0909090909090910e-10\n", - "Current fraction-to-the-boundary parameter tau = 9.9999997277681130e-01\n", + "Current barrier parameter mu = 2.5872620824534707e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999978613848917e-01\n", "\n", - "||curr_x||_inf = 1.0000000041267956e+00\n", - "||curr_s||_inf = 9.5678170729163958e-06\n", + "||curr_x||_inf = 9.7740280907595700e-01\n", + "||curr_s||_inf = 7.8561094680576796e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 1.7680920402422696e-03\n", - "||curr_z_L||_inf = 2.0383468974066809e-07\n", - "||curr_z_U||_inf = 6.5789430451231059e-07\n", - "||curr_v_L||_inf = 1.7681117532479319e-03\n", + "||curr_y_d||_inf = 1.9547373223582206e-03\n", + "||curr_z_L||_inf = 2.1618621787071527e-07\n", + "||curr_z_U||_inf = 6.4135637932442364e-07\n", + "||curr_v_L||_inf = 1.9547572099455936e-03\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 17:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.2565776316551998e-04 1.2565776316551998e-04\n", - "Dual infeasibility......: 2.5516656650561394e-07 2.5516656650561394e-07\n", + "Objective...............: 1.2176570302348460e-04 1.2176570302348460e-04\n", + "Dual infeasibility......: 2.9043384504889128e-07 2.9043384504889128e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 1.7812930228906213e-08 1.7812930228906213e-08\n", - "Overall NLP error.......: 2.5516656650561394e-07 2.5516656650561394e-07\n", + "Complementarity.........: 1.7831998442378964e-08 1.7831998442378964e-08\n", + "Overall NLP error.......: 2.9043384504889128e-07 2.9043384504889128e-07\n", "\n", "\n", "\n", @@ -6273,9 +6316,13 @@ "*** Update Barrier Parameter for Iteration 17:\n", "**************************************************\n", "\n", - "Switching back to free mu mode.\n", - "The current filter has 2 entries.\n", - "Solving the Primal Dual System for the affine step\n", + "Switching to fixed mu mode with mu = 1.2027873062894477e-08 and tau = 9.9999998797212697e-01.\n", + "Barrier Parameter: 1.202787e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 17:\n", + "**************************************************\n", + "\n", "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", @@ -6284,20 +6331,9 @@ "Number of trial factorizations performed: 1\n", "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", " delta_c=0.000000e+00 delta_d=0.000000e+00\n", - "Solving the Primal Dual System for the centering step\n", + "residual_ratio = 8.182896e-16\n", "Factorization successful.\n", - "Sigma = 4.125976e-01\n", - "Barrier parameter mu computed by oracle is 6.804585e-09\n", - "Barrier parameter mu after safeguards is 6.804585e-09\n", - "Barrier Parameter: 6.804585e-09\n", - "\n", - "**************************************************\n", - "*** Solving the Primal Dual System for Iteration 17:\n", - "**************************************************\n", - "\n", - "residual_ratio = 1.941033e-15\n", - "Factorization successful.\n", - "residual_ratio = 2.425043e-18\n", + "residual_ratio = 8.066136e-18\n", "\n", "**************************************************\n", "*** Finding Acceptable Trial Point for Iteration 17:\n", @@ -6307,17 +6343,17 @@ "Mu has changed in line search - resetting watchdog counters.\n", "Storing current iterate as backup acceptable point.\n", "The current filter has 0 entries.\n", - "minimal step size ALPHA_MIN = 8.998556E-26\n", - "Starting checks for alpha (primal) = 9.52e-01\n", + "minimal step size ALPHA_MIN = 2.076403E-25\n", + "Starting checks for alpha (primal) = 4.47e-01\n", "No Jacobian form specified for nonlinear variational problem.\n", "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 5.410e-17 (tol = 1.000e-07) r (rel) = 5.464e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.714e-17 (tol = 1.000e-07) r (rel) = 4.761e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", - "Checking acceptability for trial step size alpha_primal_test= 9.521680e-01:\n", - " New values of barrier function = 2.9116927903389189e-04 (reference 2.9365731757830004e-04):\n", - " New values of constraint violation = 1.0633651619830486e-16 (reference 8.8040604537611977e-18):\n", + "Checking acceptability for trial step size alpha_primal_test= 4.473861e-01:\n", + " New values of barrier function = 4.2888491163681197e-04 (reference 4.3056775834580680e-04):\n", + " New values of constraint violation = 7.8828274203274207e-17 (reference 1.1411227865409934e-17):\n", "Checking Armijo Condition...\n", "Succeeded...\n", "Checking filter acceptability...\n", @@ -6326,62 +6362,159 @@ "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", "Factorization successful.\n", "\n", + "**************************************************\n", + "*** Update HessianMatrix for Iteration 18:\n", + "**************************************************\n", + "\n", + "In limited-memory update, s_new_max is 3.499704e-01\n", + "Limited-Memory test for skipping:\n", + " s^Ty = 3.028295e-07 snrm = 1.297182e+00 ynrm = 7.843733e-07\n", + " Perform the update.\n", + "sigma (for B0) is 1.799683e-07\n", + "Number of successive iterations with skipping: 0\n", + "\n", "\n", "**************************************************\n", "*** Summary of Iteration: 18:\n", "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 18 1.1906625e-04 0.00e+00 2.82e-07 -8.2 2.41e-01 - 9.98e-01 9.52e-01f 1 sigma=4.13e-01 qf=12y A\n", + " 18 1.2247828e-04 0.00e+00 1.96e-07 -7.9 7.82e-01 - 5.57e-01 4.47e-01f 1 Fy A\n", "\n", "**************************************************\n", "*** Beginning Iteration 18 from the following point:\n", "**************************************************\n", "\n", - "Current barrier parameter mu = 6.8045847875022476e-09\n", - "Current fraction-to-the-boundary parameter tau = 9.9999974483343346e-01\n", + "Current barrier parameter mu = 1.2027873062894477e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999998797212697e-01\n", "\n", - "||curr_x||_inf = 9.9999998617316055e-01\n", - "||curr_s||_inf = 9.0481780820836682e-06\n", + "||curr_x||_inf = 1.0000000083740865e+00\n", + "||curr_s||_inf = 8.1215562449621383e-06\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 8.1313752757328553e-04\n", - "||curr_z_L||_inf = 9.4352784989153782e-08\n", - "||curr_z_U||_inf = 5.2997432177651092e-07\n", - "||curr_v_L||_inf = 8.1313496326127247e-04\n", + "||curr_y_d||_inf = 1.6355194586317435e-03\n", + "||curr_z_L||_inf = 1.7926883243309775e-07\n", + "||curr_z_U||_inf = 5.4977433490897253e-07\n", + "||curr_v_L||_inf = 1.6355299342325469e-03\n", "||curr_v_U||_inf = 0.0000000000000000e+00\n", "\n", "\n", "***Current NLP Values for Iteration 18:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.1906625426859106e-04 1.1906625426859106e-04\n", - "Dual infeasibility......: 2.8191370018366373e-07 2.8191370018366373e-07\n", + "Objective...............: 1.2247828477481950e-04 1.2247828477481950e-04\n", + "Dual infeasibility......: 1.9594179028080537e-07 1.9594179028080537e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 1.2457836678372943e-08 1.2457836678372943e-08\n", - "Overall NLP error.......: 2.8191370018366373e-07 2.8191370018366373e-07\n", + "Complementarity.........: 1.5196264624183976e-08 1.5196264624183976e-08\n", + "Overall NLP error.......: 1.9594179028080537e-07 1.9594179028080537e-07\n", "\n", "\n", "\n", "\n", - "Number of Iterations....: 18\n", + "**************************************************\n", + "*** Update Barrier Parameter for Iteration 18:\n", + "**************************************************\n", + "\n", + "Remaining in fixed mu mode.\n", + "Barrier Parameter: 1.202787e-08\n", + "\n", + "**************************************************\n", + "*** Solving the Primal Dual System for Iteration 18:\n", + "**************************************************\n", + "\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Factorization successful.\n", + "Number of trial factorizations performed: 1\n", + "Perturbation parameters: delta_x=0.000000e+00 delta_s=0.000000e+00\n", + " delta_c=0.000000e+00 delta_d=0.000000e+00\n", + "residual_ratio = 2.419797e-15\n", + "Factorization successful.\n", + "residual_ratio = 1.197977e-18\n", + "\n", + "**************************************************\n", + "*** Finding Acceptable Trial Point for Iteration 18:\n", + "**************************************************\n", + "\n", + "--> Starting line search in iteration 18 <--\n", + "Storing current iterate as backup acceptable point.\n", + "The current filter has 0 entries.\n", + "minimal step size ALPHA_MIN = 2.195335E-26\n", + "Starting checks for alpha (primal) = 2.83e-01\n", + "No Jacobian form specified for nonlinear variational problem.\n", + "Differentiating residual form F to obtain Jacobian J = F'.\n", + "Solving nonlinear variational problem.\n", + " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.637e-17 (tol = 1.000e-07) r (rel) = 4.683e-13 (tol = 1.000e-09)\n", + " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", + "Checking acceptability for trial step size alpha_primal_test= 2.828678e-01:\n", + " New values of barrier function = 4.2773294117671708e-04 (reference 4.2888491163681197e-04):\n", + " New values of constraint violation = 7.0047930672036129e-17 (reference 7.8828274203274207e-17):\n", + "Checking Armijo Condition...\n", + "Succeeded...\n", + "Checking filter acceptability...\n", + "Succeeded...\n", + "Number of doubles for MUMPS to hold factorization (INFO(9)) = 20406\n", + "Number of integers for MUMPS to hold factorization (INFO(10)) = 244848\n", + "Factorization successful.\n", + "\n", + "\n", + "**************************************************\n", + "*** Summary of Iteration: 19:\n", + "**************************************************\n", + "\n", + "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", + " 19 1.2297888e-04 0.00e+00 2.14e-07 -7.9 9.43e-01 - 5.86e-01 2.83e-01f 1 Fy A\n", + "\n", + "**************************************************\n", + "*** Beginning Iteration 19 from the following point:\n", + "**************************************************\n", + "\n", + "Current barrier parameter mu = 1.2027873062894477e-08\n", + "Current fraction-to-the-boundary parameter tau = 9.9999998797212697e-01\n", + "\n", + "||curr_x||_inf = 1.0000000078808737e+00\n", + "||curr_s||_inf = 8.3286158090855116e-06\n", + "||curr_y_c||_inf = 0.0000000000000000e+00\n", + "||curr_y_d||_inf = 1.4577633268749483e-03\n", + "||curr_z_L||_inf = 1.5984297319862014e-07\n", + "||curr_z_U||_inf = 5.2689990964864298e-07\n", + "||curr_v_L||_inf = 1.4577685220642841e-03\n", + "||curr_v_U||_inf = 0.0000000000000000e+00\n", + "\n", + "\n", + "***Current NLP Values for Iteration 19:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.1906625426859106e-04 1.1906625426859106e-04\n", - "Dual infeasibility......: 2.8191370018366373e-07 2.8191370018366373e-07\n", + "Objective...............: 1.2297888298278239e-04 1.2297888298278239e-04\n", + "Dual infeasibility......: 2.1443637202631501e-07 2.1443637202631501e-07\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", - "Complementarity.........: 1.2457836678372943e-08 1.2457836678372943e-08\n", - "Overall NLP error.......: 2.8191370018366373e-07 2.8191370018366373e-07\n", + "Complementarity.........: 1.5730157321749088e-08 1.5730157321749088e-08\n", + "Overall NLP error.......: 2.1443637202631501e-07 2.1443637202631501e-07\n", "\n", "\n", - "Number of objective function evaluations = 19\n", - "Number of objective gradient evaluations = 19\n", + "\n", + "\n", + "Number of Iterations....: 19\n", + "\n", + " (scaled) (unscaled)\n", + "Objective...............: 1.2297888298278239e-04 1.2297888298278239e-04\n", + "Dual infeasibility......: 2.1443637202631501e-07 2.1443637202631501e-07\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 1.5730157321749088e-08 1.5730157321749088e-08\n", + "Overall NLP error.......: 2.1443637202631501e-07 2.1443637202631501e-07\n", + "\n", + "\n", + "Number of objective function evaluations = 20\n", + "Number of objective gradient evaluations = 20\n", "Number of equality constraint evaluations = 0\n", - "Number of inequality constraint evaluations = 19\n", + "Number of inequality constraint evaluations = 20\n", "Number of equality constraint Jacobian evaluations = 0\n", - "Number of inequality constraint Jacobian evaluations = 19\n", + "Number of inequality constraint Jacobian evaluations = 20\n", "Number of Lagrangian Hessian evaluations = 0\n", - "Total CPU secs in IPOPT (w/o function evaluations) = 1.963\n", - "Total CPU secs in NLP function evaluations = 2.144\n", + "Total CPU secs in IPOPT (w/o function evaluations) = 2.151\n", + "Total CPU secs in NLP function evaluations = 2.255\n", "\n", "EXIT: Solved To Acceptable Level.\n", "v=0.2\n", @@ -6394,46 +6527,15 @@ "text": [ "rm: cannot remove '/home/fenics/shared/templates/RES_OPT/TEMP*': No such file or directory\n" ] - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" } ], "source": [ "\n", - "# Let's first optimize the generated designs, to see how far away they are from optimal designs\n", "opt_designs = []\n", "opt_histories = []\n", "for i in range(n_samples):\n", " opt_design, opt_history = problem.optimize(\n", - " starting_point=gen_designs[i][0],\n", + " starting_point=gen_designs[i],\n", " config=conditions[i],\n", " )\n", " opt_histories.append(opt_history)\n", @@ -6444,17 +6546,19 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Eyeball testing the model" + "### Eyeball testing the model\n", + "\n", + "Render the optimal designs next to the generated designs to get a visual idea" ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 9, "metadata": {}, "outputs": [ { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -6464,7 +6568,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -6474,7 +6578,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -6484,7 +6588,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -6494,7 +6598,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -6504,7 +6608,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -6514,10 +6618,10 @@ } ], "source": [ - "# Render the optimal designs next to the generated designs to get a visual idea\n", + "\n", "for i in range(n_samples):\n", " fig_opt, ax_opt = problem.render(opt_designs[i])\n", - " fig_gen, ax_gen = problem.render(gen_designs[i][0])\n", + " fig_gen, ax_gen = problem.render(gen_designs[i])\n", " ax_opt.set_title(\"Optimal\")\n", " ax_gen.set_title(\"Generated\")\n", " plt.show()" @@ -6527,12 +6631,14 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Qualitative evaluation" + "### Qualitative evaluation\n", + "\n", + "Now we can also use metrics to get a quantitative idea of how good the generated designs are. Here we compute optimality gaps between the generated designs and optimized designs in terms of objective values (thermal compliance for heatconduction)" ] }, { "cell_type": "code", - "execution_count": 44, + "execution_count": 11, "metadata": {}, "outputs": [ { @@ -6552,7 +6658,7 @@ "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 3.392e-17 (tol = 1.000e-07) r (rel) = 3.425e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 6.329e-17 (tol = 1.000e-07) r (rel) = 6.389e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", @@ -6611,7 +6717,7 @@ "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.906e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 3.529e-17 (tol = 1.000e-07) r (rel) = 3.563e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 5.466e-17 (tol = 1.000e-07) r (rel) = 5.518e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "\n", @@ -6621,7 +6727,7 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 0 2.8915034e-05 0.00e+00 1.00e-04 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", + " 0 6.3524891e-05 0.00e+00 1.00e-04 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", "\n", "**************************************************\n", "*** Beginning Iteration 0 from the following point:\n", @@ -6633,7 +6739,7 @@ "||curr_x||_inf = 9.9000000989999992e-01\n", "||curr_s||_inf = 9.9999900000000003e-03\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 9.9990102485197330e-01\n", + "||curr_y_d||_inf = 9.9990103626823723e-01\n", "||curr_z_L||_inf = 1.0000000000000000e+00\n", "||curr_z_U||_inf = 1.0000000000000000e+00\n", "||curr_v_L||_inf = 1.0000000000000000e+00\n", @@ -6643,8 +6749,8 @@ "***Current NLP Values for Iteration 0:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 2.8915033667732908e-05 2.8915033667732908e-05\n", - "Dual infeasibility......: 9.9997412027374999e-05 9.9997412027374999e-05\n", + "Objective...............: 6.3524891326338607e-05 6.3524891326338607e-05\n", + "Dual infeasibility......: 1.0001097677203585e-04 1.0001097677203585e-04\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", "Complementarity.........: 9.9000001989999997e-01 9.9000001989999997e-01\n", "Overall NLP error.......: 9.9000001989999997e-01 9.9000001989999997e-01\n", @@ -6655,8 +6761,8 @@ "Number of Iterations....: 0\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 2.8915033667732908e-05 2.8915033667732908e-05\n", - "Dual infeasibility......: 9.9997412027374999e-05 9.9997412027374999e-05\n", + "Objective...............: 6.3524891326338607e-05 6.3524891326338607e-05\n", + "Dual infeasibility......: 1.0001097677203585e-04 1.0001097677203585e-04\n", "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", "Complementarity.........: 9.9000001989999997e-01 9.9000001989999997e-01\n", "Overall NLP error.......: 9.9000001989999997e-01 9.9000001989999997e-01\n", @@ -6669,11 +6775,11 @@ "Number of equality constraint Jacobian evaluations = 0\n", "Number of inequality constraint Jacobian evaluations = 1\n", "Number of Lagrangian Hessian evaluations = 0\n", - "Total CPU secs in IPOPT (w/o function evaluations) = 0.233\n", - "Total CPU secs in NLP function evaluations = 0.129\n", + "Total CPU secs in IPOPT (w/o function evaluations) = 0.229\n", + "Total CPU secs in NLP function evaluations = 0.136\n", "\n", "EXIT: Maximum Number of Iterations Exceeded.\n", - "Optimization complete: v=0.5, w=0.5\n" + "Optimization complete: v=0.3, w=0.5\n" ] }, { @@ -6694,7 +6800,7 @@ "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 4.834e-17 (tol = 1.000e-07) r (rel) = 4.878e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.882e-17 (tol = 1.000e-07) r (rel) = 4.927e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", @@ -6753,7 +6859,7 @@ "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.909e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 5.142e-17 (tol = 1.000e-07) r (rel) = 5.189e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.597e-17 (tol = 1.000e-07) r (rel) = 4.640e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "\n", @@ -6763,7 +6869,7 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 0 6.8136479e-05 1.94e-03 1.00e-04 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", + " 0 8.1159905e-05 0.00e+00 1.00e-04 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", "\n", "**************************************************\n", "*** Beginning Iteration 0 from the following point:\n", @@ -6775,7 +6881,7 @@ "||curr_x||_inf = 9.9000000989999992e-01\n", "||curr_s||_inf = 9.9999900000000003e-03\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 9.9990105427563769e-01\n", + "||curr_y_d||_inf = 9.9990106716702110e-01\n", "||curr_z_L||_inf = 1.0000000000000000e+00\n", "||curr_z_U||_inf = 1.0000000000000000e+00\n", "||curr_v_L||_inf = 1.0000000000000000e+00\n", @@ -6785,11 +6891,11 @@ "***Current NLP Values for Iteration 0:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 6.8136478539192145e-05 6.8136478539192145e-05\n", - "Dual infeasibility......: 9.9989669015254634e-05 9.9989669015254634e-05\n", - "Constraint violation....: 1.9428233460940195e-03 1.9428233460940195e-03\n", - "Complementarity.........: 9.9000001989999997e-01 9.9000001989999997e-01\n", - "Overall NLP error.......: 9.9000001989999997e-01 9.9000001989999997e-01\n", + "Objective...............: 8.1159905148195084e-05 8.1159905148195084e-05\n", + "Dual infeasibility......: 9.9989610066741896e-05 9.9989610066741896e-05\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 9.9000001999999998e-01 9.9000001999999998e-01\n", + "Overall NLP error.......: 9.9000001999999998e-01 9.9000001999999998e-01\n", "\n", "\n", "\n", @@ -6797,11 +6903,11 @@ "Number of Iterations....: 0\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 6.8136478539192145e-05 6.8136478539192145e-05\n", - "Dual infeasibility......: 9.9989669015254634e-05 9.9989669015254634e-05\n", - "Constraint violation....: 1.9428233460940195e-03 1.9428233460940195e-03\n", - "Complementarity.........: 9.9000001989999997e-01 9.9000001989999997e-01\n", - "Overall NLP error.......: 9.9000001989999997e-01 9.9000001989999997e-01\n", + "Objective...............: 8.1159905148195084e-05 8.1159905148195084e-05\n", + "Dual infeasibility......: 9.9989610066741896e-05 9.9989610066741896e-05\n", + "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Complementarity.........: 9.9000001999999998e-01 9.9000001999999998e-01\n", + "Overall NLP error.......: 9.9000001999999998e-01 9.9000001999999998e-01\n", "\n", "\n", "Number of objective function evaluations = 1\n", @@ -6811,8 +6917,8 @@ "Number of equality constraint Jacobian evaluations = 0\n", "Number of inequality constraint Jacobian evaluations = 1\n", "Number of Lagrangian Hessian evaluations = 0\n", - "Total CPU secs in IPOPT (w/o function evaluations) = 0.226\n", - "Total CPU secs in NLP function evaluations = 0.131\n", + "Total CPU secs in IPOPT (w/o function evaluations) = 0.229\n", + "Total CPU secs in NLP function evaluations = 0.135\n", "\n", "EXIT: Maximum Number of Iterations Exceeded.\n", "Optimization complete: v=0.4, w=0.7\n" @@ -6836,7 +6942,7 @@ "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 5.361e-17 (tol = 1.000e-07) r (rel) = 5.415e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.645e-17 (tol = 1.000e-07) r (rel) = 4.691e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", @@ -6895,7 +7001,7 @@ "Differentiating residual form F to obtain Jacobian J = F'.\n", "Solving nonlinear variational problem.\n", " Newton iteration 0: r (abs) = 9.901e-05 (tol = 1.000e-07) r (rel) = 1.000e+00 (tol = 1.000e-09)\n", - " Newton iteration 1: r (abs) = 5.597e-17 (tol = 1.000e-07) r (rel) = 5.653e-13 (tol = 1.000e-09)\n", + " Newton iteration 1: r (abs) = 4.887e-17 (tol = 1.000e-07) r (rel) = 4.935e-13 (tol = 1.000e-09)\n", " Newton solver finished in 1 iterations and 1 linear solver iterations.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "\n", @@ -6905,7 +7011,7 @@ "**************************************************\n", "\n", "iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n", - " 0 1.2998394e-04 0.00e+00 1.00e-04 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", + " 0 1.3010401e-04 1.42e-03 1.00e-04 0.0 0.00e+00 - 0.00e+00 0.00e+00 0 y\n", "\n", "**************************************************\n", "*** Beginning Iteration 0 from the following point:\n", @@ -6917,7 +7023,7 @@ "||curr_x||_inf = 9.9000000989999992e-01\n", "||curr_s||_inf = 9.9999900000000003e-03\n", "||curr_y_c||_inf = 0.0000000000000000e+00\n", - "||curr_y_d||_inf = 9.9990105260724160e-01\n", + "||curr_y_d||_inf = 9.9990105500056670e-01\n", "||curr_z_L||_inf = 1.0000000000000000e+00\n", "||curr_z_U||_inf = 1.0000000000000000e+00\n", "||curr_v_L||_inf = 1.0000000000000000e+00\n", @@ -6927,9 +7033,9 @@ "***Current NLP Values for Iteration 0:\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.2998393571133284e-04 1.2998393571133284e-04\n", - "Dual infeasibility......: 1.0001132391701528e-04 1.0001132391701528e-04\n", - "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Objective...............: 1.3010401228781608e-04 1.3010401228781608e-04\n", + "Dual infeasibility......: 9.9994541991166308e-05 9.9994541991166308e-05\n", + "Constraint violation....: 1.4202672724238034e-03 1.4202672724238034e-03\n", "Complementarity.........: 9.9000001989999997e-01 9.9000001989999997e-01\n", "Overall NLP error.......: 9.9000001989999997e-01 9.9000001989999997e-01\n", "\n", @@ -6939,9 +7045,9 @@ "Number of Iterations....: 0\n", "\n", " (scaled) (unscaled)\n", - "Objective...............: 1.2998393571133284e-04 1.2998393571133284e-04\n", - "Dual infeasibility......: 1.0001132391701528e-04 1.0001132391701528e-04\n", - "Constraint violation....: 0.0000000000000000e+00 0.0000000000000000e+00\n", + "Objective...............: 1.3010401228781608e-04 1.3010401228781608e-04\n", + "Dual infeasibility......: 9.9994541991166308e-05 9.9994541991166308e-05\n", + "Constraint violation....: 1.4202672724238034e-03 1.4202672724238034e-03\n", "Complementarity.........: 9.9000001989999997e-01 9.9000001989999997e-01\n", "Overall NLP error.......: 9.9000001989999997e-01 9.9000001989999997e-01\n", "\n", @@ -6953,14 +7059,14 @@ "Number of equality constraint Jacobian evaluations = 0\n", "Number of inequality constraint Jacobian evaluations = 1\n", "Number of Lagrangian Hessian evaluations = 0\n", - "Total CPU secs in IPOPT (w/o function evaluations) = 0.225\n", + "Total CPU secs in IPOPT (w/o function evaluations) = 0.235\n", "Total CPU secs in NLP function evaluations = 0.131\n", "\n", "EXIT: Maximum Number of Iterations Exceeded.\n", "Optimization complete: v=0.2, w=0.1\n", - "Initial optimality gaps: [array([0.0004409]), array([0.0004189]), array([1.70724543e-05])]\n", - "Cumulative optimality gaps: [0.00262544707661, 0.0026488498937400004, 0.00036407371150999987]\n", - "Final optimality gaps: [array([-7.5917667e-07]), array([1.8115146e-07]), array([-1.09176857e-05])]\n" + "\u001b[96mInitial optimality gaps: [0.00012219959867, 0.00052516607485, 8.931857771e-05]\u001b[0m\n", + "\u001b[96mCumulative optimality gaps: [0.0012739121460599997, 0.0041678009641500004, 0.006302119584199999]\u001b[0m\n", + "\u001b[96mFinal optimality gaps: [-2.487013330000005e-06, -1.848306149999991e-06, -7.125132289999996e-06]\u001b[0m\n" ] }, { @@ -6972,18 +7078,17 @@ } ], "source": [ - "# Now we can also use metrics to get a quantitative idea of how good the generated designs are\n", - "# Here we compute the average optimality gap, the cumulative optimality gap, and the final optimality gap\n", + "\n", "objective_values_opt = [problem.simulate(opt_designs[i], config=conditions[i]) for i in range(n_samples)]\n", "optimality_gaps = [metrics.optimality_gap(opt_histories[i], baseline=objective_values_opt[i]) for i in range(n_samples)]\n", "\n", - "iog = [optimality_gaps[i][0] for i in range(n_samples)]\n", + "iog = [optimality_gaps[i][0][0] for i in range(n_samples)]\n", "cog = [np.sum(optimality_gaps[i]) for i in range(n_samples)]\n", - "fog = [optimality_gaps[i][-1] for i in range(n_samples)]\n", + "fog = [optimality_gaps[i][-1][0] for i in range(n_samples)]\n", "\n", - "print(f\"Initial optimality gaps: {iog}\")\n", - "print(f\"Cumulative optimality gaps: {cog}\")\n", - "print(f\"Final optimality gaps: {fog}\")" + "print(\"\\033[96m\" + f\"Initial optimality gaps: {iog}\" + \"\\033[0m\")\n", + "print(\"\\033[96m\" + f\"Cumulative optimality gaps: {cog}\" + \"\\033[0m\")\n", + "print(\"\\033[96m\" + f\"Final optimality gaps: {fog}\" + \"\\033[0m\")" ] } ], From 7123af74b8a901cc3a2bcb2dc2b2bc04faa36a02 Mon Sep 17 00:00:00 2001 From: Florian Felten Date: Wed, 9 Jul 2025 14:18:46 +0200 Subject: [PATCH 4/5] Remove some outputs from saved notebook --- example_hard_model.ipynb | 211 ++------------------------------------- 1 file changed, 6 insertions(+), 205 deletions(-) diff --git a/example_hard_model.ipynb b/example_hard_model.ipynb index 4181f25d..9680c084 100644 --- a/example_hard_model.ipynb +++ b/example_hard_model.ipynb @@ -20,204 +20,18 @@ }, { "cell_type": "code", - "execution_count": 1, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Obtaining file:///Users/ffelte/Documents/EngiLearn\n", - " Installing build dependencies ... \u001b[?25ldone\n", - "\u001b[?25h Checking if build backend supports build_editable ... \u001b[?25ldone\n", - "\u001b[?25h Getting requirements to build editable ... \u001b[?25ldone\n", - "\u001b[?25h Preparing editable metadata (pyproject.toml) ... \u001b[?25ldone\n", - "\u001b[?25hRequirement already satisfied: engibench[all] in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.0.1a0)\n", - "Requirement already satisfied: tyro>=0.9.2 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.9.2)\n", - "Requirement already satisfied: numpy in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (1.26.2)\n", - "Requirement already satisfied: torch>=2.5.0 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (2.5.1)\n", - "Requirement already satisfied: torchvision>=0.20.1 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.20.1)\n", - "Requirement already satisfied: wandb>=0.18.7 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.19.10)\n", - "Requirement already satisfied: tqdm>=4.67.1 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (4.67.1)\n", - "Requirement already satisfied: matplotlib>=3.9.2 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (3.9.3)\n", - "Requirement already satisfied: denoising_diffusion_pytorch in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (2.1.1)\n", - "Requirement already satisfied: diffusers>=0.33.1 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.33.1)\n", - "Requirement already satisfied: pymoo>=0.6 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.6.1.3)\n", - "Requirement already satisfied: ax-platform>=0.5 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.5.0)\n", - "Requirement already satisfied: scikit-learn>=1.6.0 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (1.6.1)\n", - "Requirement already satisfied: hyppo>=0.5.0 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.5.1)\n", - "Requirement already satisfied: kaleido>=0.2.1 in ./.venv/lib/python3.12/site-packages (from engiopt==0.0.1) (0.2.1)\n", - "Requirement already satisfied: botorch==0.13.0 in ./.venv/lib/python3.12/site-packages (from ax-platform>=0.5->engiopt==0.0.1) (0.13.0)\n", - "Requirement already satisfied: jinja2 in ./.venv/lib/python3.12/site-packages (from ax-platform>=0.5->engiopt==0.0.1) (3.1.4)\n", - "Requirement already satisfied: pandas in ./.venv/lib/python3.12/site-packages (from ax-platform>=0.5->engiopt==0.0.1) (2.2.3)\n", - "Requirement already satisfied: scipy in ./.venv/lib/python3.12/site-packages (from ax-platform>=0.5->engiopt==0.0.1) (1.14.1)\n", - "Requirement already satisfied: ipywidgets in ./.venv/lib/python3.12/site-packages (from ax-platform>=0.5->engiopt==0.0.1) (8.1.6)\n", - "Requirement already satisfied: plotly>=5.12.0 in ./.venv/lib/python3.12/site-packages (from ax-platform>=0.5->engiopt==0.0.1) (6.0.1)\n", - "Requirement already satisfied: pyre-extensions in ./.venv/lib/python3.12/site-packages (from ax-platform>=0.5->engiopt==0.0.1) (0.0.32)\n", - "Requirement already satisfied: sympy in ./.venv/lib/python3.12/site-packages (from ax-platform>=0.5->engiopt==0.0.1) (1.13.1)\n", - "Requirement already satisfied: typing_extensions in ./.venv/lib/python3.12/site-packages (from botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (4.12.2)\n", - "Requirement already satisfied: gpytorch==1.14 in ./.venv/lib/python3.12/site-packages (from botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (1.14)\n", - "Requirement already satisfied: linear_operator==0.6 in ./.venv/lib/python3.12/site-packages (from botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (0.6)\n", - "Requirement already satisfied: pyro-ppl>=1.8.4 in ./.venv/lib/python3.12/site-packages (from botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (1.9.1)\n", - "Requirement already satisfied: multipledispatch in ./.venv/lib/python3.12/site-packages (from botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (1.0.0)\n", - "Requirement already satisfied: threadpoolctl in ./.venv/lib/python3.12/site-packages (from botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (3.6.0)\n", - "Requirement already satisfied: jaxtyping in ./.venv/lib/python3.12/site-packages (from gpytorch==1.14->botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (0.3.2)\n", - "Requirement already satisfied: mpmath<=1.3,>=0.19 in ./.venv/lib/python3.12/site-packages (from gpytorch==1.14->botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (1.3.0)\n", - "Requirement already satisfied: importlib-metadata in ./.venv/lib/python3.12/site-packages (from diffusers>=0.33.1->engiopt==0.0.1) (8.7.0)\n", - "Requirement already satisfied: filelock in ./.venv/lib/python3.12/site-packages (from diffusers>=0.33.1->engiopt==0.0.1) (3.16.1)\n", - "Requirement already satisfied: huggingface-hub>=0.27.0 in ./.venv/lib/python3.12/site-packages (from diffusers>=0.33.1->engiopt==0.0.1) (0.30.2)\n", - "Requirement already satisfied: regex!=2019.12.17 in ./.venv/lib/python3.12/site-packages (from diffusers>=0.33.1->engiopt==0.0.1) (2024.11.6)\n", - "Requirement already satisfied: requests in ./.venv/lib/python3.12/site-packages (from diffusers>=0.33.1->engiopt==0.0.1) (2.32.3)\n", - "Requirement already satisfied: safetensors>=0.3.1 in ./.venv/lib/python3.12/site-packages (from diffusers>=0.33.1->engiopt==0.0.1) (0.4.5)\n", - "Requirement already satisfied: Pillow in ./.venv/lib/python3.12/site-packages (from diffusers>=0.33.1->engiopt==0.0.1) (11.0.0)\n", - "Requirement already satisfied: numba>=0.46 in ./.venv/lib/python3.12/site-packages (from hyppo>=0.5.0->engiopt==0.0.1) (0.61.2)\n", - "Requirement already satisfied: autograd>=1.3 in ./.venv/lib/python3.12/site-packages (from hyppo>=0.5.0->engiopt==0.0.1) (1.7.0)\n", - "Requirement already satisfied: contourpy>=1.0.1 in ./.venv/lib/python3.12/site-packages (from matplotlib>=3.9.2->engiopt==0.0.1) (1.3.1)\n", - "Requirement already satisfied: cycler>=0.10 in ./.venv/lib/python3.12/site-packages (from matplotlib>=3.9.2->engiopt==0.0.1) (0.12.1)\n", - "Requirement already satisfied: fonttools>=4.22.0 in ./.venv/lib/python3.12/site-packages (from matplotlib>=3.9.2->engiopt==0.0.1) (4.55.0)\n", - "Requirement already satisfied: kiwisolver>=1.3.1 in ./.venv/lib/python3.12/site-packages (from matplotlib>=3.9.2->engiopt==0.0.1) (1.4.7)\n", - "Requirement already satisfied: packaging>=20.0 in ./.venv/lib/python3.12/site-packages (from matplotlib>=3.9.2->engiopt==0.0.1) (24.2)\n", - "Requirement already satisfied: pyparsing>=2.3.1 in ./.venv/lib/python3.12/site-packages (from matplotlib>=3.9.2->engiopt==0.0.1) (3.2.0)\n", - "Requirement already satisfied: python-dateutil>=2.7 in ./.venv/lib/python3.12/site-packages (from matplotlib>=3.9.2->engiopt==0.0.1) (2.9.0.post0)\n", - "Requirement already satisfied: cma==3.2.2 in ./.venv/lib/python3.12/site-packages (from pymoo>=0.6->engiopt==0.0.1) (3.2.2)\n", - "Requirement already satisfied: alive-progress in ./.venv/lib/python3.12/site-packages (from pymoo>=0.6->engiopt==0.0.1) (3.2.0)\n", - "Requirement already satisfied: dill in ./.venv/lib/python3.12/site-packages (from pymoo>=0.6->engiopt==0.0.1) (0.3.8)\n", - "Requirement already satisfied: Deprecated in ./.venv/lib/python3.12/site-packages (from pymoo>=0.6->engiopt==0.0.1) (1.2.18)\n", - "Requirement already satisfied: joblib>=1.2.0 in ./.venv/lib/python3.12/site-packages (from scikit-learn>=1.6.0->engiopt==0.0.1) (1.4.2)\n", - "Requirement already satisfied: networkx in ./.venv/lib/python3.12/site-packages (from torch>=2.5.0->engiopt==0.0.1) (3.4.2)\n", - "Requirement already satisfied: fsspec in ./.venv/lib/python3.12/site-packages (from torch>=2.5.0->engiopt==0.0.1) (2024.9.0)\n", - "Requirement already satisfied: setuptools in ./.venv/lib/python3.12/site-packages (from torch>=2.5.0->engiopt==0.0.1) (75.6.0)\n", - "Requirement already satisfied: docstring-parser>=0.16 in ./.venv/lib/python3.12/site-packages (from tyro>=0.9.2->engiopt==0.0.1) (0.16)\n", - "Requirement already satisfied: rich>=11.1.0 in ./.venv/lib/python3.12/site-packages (from tyro>=0.9.2->engiopt==0.0.1) (13.9.4)\n", - "Requirement already satisfied: shtab>=1.5.6 in ./.venv/lib/python3.12/site-packages (from tyro>=0.9.2->engiopt==0.0.1) (1.7.1)\n", - "Requirement already satisfied: typeguard>=4.0.0 in ./.venv/lib/python3.12/site-packages (from tyro>=0.9.2->engiopt==0.0.1) (4.4.1)\n", - "Requirement already satisfied: click!=8.0.0,>=7.1 in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (8.1.7)\n", - "Requirement already satisfied: docker-pycreds>=0.4.0 in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (0.4.0)\n", - "Requirement already satisfied: gitpython!=3.1.29,>=1.0.0 in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (3.1.43)\n", - "Requirement already satisfied: platformdirs in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (4.3.6)\n", - "Requirement already satisfied: protobuf!=4.21.0,!=5.28.0,<7,>=3.19.0 in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (5.28.3)\n", - "Requirement already satisfied: psutil>=5.0.0 in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (6.1.0)\n", - "Requirement already satisfied: pydantic<3 in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (2.11.3)\n", - "Requirement already satisfied: pyyaml in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (6.0.2)\n", - "Requirement already satisfied: sentry-sdk>=2.0.0 in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (2.19.0)\n", - "Requirement already satisfied: setproctitle in ./.venv/lib/python3.12/site-packages (from wandb>=0.18.7->engiopt==0.0.1) (1.3.4)\n", - "Requirement already satisfied: accelerate in ./.venv/lib/python3.12/site-packages (from denoising_diffusion_pytorch->engiopt==0.0.1) (1.2.0)\n", - "Requirement already satisfied: einops in ./.venv/lib/python3.12/site-packages (from denoising_diffusion_pytorch->engiopt==0.0.1) (0.8.0)\n", - "Requirement already satisfied: ema-pytorch>=0.4.2 in ./.venv/lib/python3.12/site-packages (from denoising_diffusion_pytorch->engiopt==0.0.1) (0.7.7)\n", - "Requirement already satisfied: pytorch-fid in ./.venv/lib/python3.12/site-packages (from denoising_diffusion_pytorch->engiopt==0.0.1) (0.3.0)\n", - "Requirement already satisfied: gymnasium>=1.0.0 in ./.venv/lib/python3.12/site-packages (from engibench[all]->engiopt==0.0.1) (1.0.0)\n", - "Requirement already satisfied: datasets>=3.1.0 in ./.venv/lib/python3.12/site-packages (from datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (3.1.0)\n", - "Requirement already satisfied: pyarrow<20.0.0,>=15.0.0 in ./.venv/lib/python3.12/site-packages (from engibench[all]->engiopt==0.0.1) (18.1.0)\n", - "Requirement already satisfied: xxhash in ./.venv/lib/python3.12/site-packages (from datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (3.5.0)\n", - "Requirement already satisfied: multiprocess<0.70.17 in ./.venv/lib/python3.12/site-packages (from datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (0.70.16)\n", - "Requirement already satisfied: aiohttp in ./.venv/lib/python3.12/site-packages (from datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (3.11.7)\n", - "Requirement already satisfied: six>=1.4.0 in ./.venv/lib/python3.12/site-packages (from docker-pycreds>=0.4.0->wandb>=0.18.7->engiopt==0.0.1) (1.16.0)\n", - "Requirement already satisfied: gitdb<5,>=4.0.1 in ./.venv/lib/python3.12/site-packages (from gitpython!=3.1.29,>=1.0.0->wandb>=0.18.7->engiopt==0.0.1) (4.0.11)\n", - "Requirement already satisfied: cloudpickle>=1.2.0 in ./.venv/lib/python3.12/site-packages (from gymnasium>=1.0.0->engibench[all]->engiopt==0.0.1) (3.1.0)\n", - "Requirement already satisfied: farama-notifications>=0.0.1 in ./.venv/lib/python3.12/site-packages (from gymnasium>=1.0.0->engibench[all]->engiopt==0.0.1) (0.0.4)\n", - "Requirement already satisfied: llvmlite<0.45,>=0.44.0dev0 in ./.venv/lib/python3.12/site-packages (from numba>=0.46->hyppo>=0.5.0->engiopt==0.0.1) (0.44.0)\n", - "Requirement already satisfied: pytz>=2020.1 in ./.venv/lib/python3.12/site-packages (from pandas->ax-platform>=0.5->engiopt==0.0.1) (2024.2)\n", - "Requirement already satisfied: tzdata>=2022.7 in ./.venv/lib/python3.12/site-packages (from pandas->ax-platform>=0.5->engiopt==0.0.1) (2024.2)\n", - "Requirement already satisfied: narwhals>=1.15.1 in ./.venv/lib/python3.12/site-packages (from plotly>=5.12.0->ax-platform>=0.5->engiopt==0.0.1) (1.37.0)\n", - "Requirement already satisfied: annotated-types>=0.6.0 in ./.venv/lib/python3.12/site-packages (from pydantic<3->wandb>=0.18.7->engiopt==0.0.1) (0.7.0)\n", - "Requirement already satisfied: pydantic-core==2.33.1 in ./.venv/lib/python3.12/site-packages (from pydantic<3->wandb>=0.18.7->engiopt==0.0.1) (2.33.1)\n", - "Requirement already satisfied: typing-inspection>=0.4.0 in ./.venv/lib/python3.12/site-packages (from pydantic<3->wandb>=0.18.7->engiopt==0.0.1) (0.4.0)\n", - "Requirement already satisfied: charset-normalizer<4,>=2 in ./.venv/lib/python3.12/site-packages (from requests->diffusers>=0.33.1->engiopt==0.0.1) (3.4.0)\n", - "Requirement already satisfied: idna<4,>=2.5 in ./.venv/lib/python3.12/site-packages (from requests->diffusers>=0.33.1->engiopt==0.0.1) (3.10)\n", - "Requirement already satisfied: urllib3<3,>=1.21.1 in ./.venv/lib/python3.12/site-packages (from requests->diffusers>=0.33.1->engiopt==0.0.1) (2.2.3)\n", - "Requirement already satisfied: certifi>=2017.4.17 in ./.venv/lib/python3.12/site-packages (from requests->diffusers>=0.33.1->engiopt==0.0.1) (2024.8.30)\n", - "Requirement already satisfied: markdown-it-py>=2.2.0 in ./.venv/lib/python3.12/site-packages (from rich>=11.1.0->tyro>=0.9.2->engiopt==0.0.1) (3.0.0)\n", - "Requirement already satisfied: pygments<3.0.0,>=2.13.0 in ./.venv/lib/python3.12/site-packages (from rich>=11.1.0->tyro>=0.9.2->engiopt==0.0.1) (2.18.0)\n", - "Requirement already satisfied: about-time==4.2.1 in ./.venv/lib/python3.12/site-packages (from alive-progress->pymoo>=0.6->engiopt==0.0.1) (4.2.1)\n", - "Requirement already satisfied: grapheme==0.6.0 in ./.venv/lib/python3.12/site-packages (from alive-progress->pymoo>=0.6->engiopt==0.0.1) (0.6.0)\n", - "Requirement already satisfied: wrapt<2,>=1.10 in ./.venv/lib/python3.12/site-packages (from Deprecated->pymoo>=0.6->engiopt==0.0.1) (1.17.2)\n", - "Collecting pyoptsparse@ git+https://github.com/mdolab/pyoptsparse#egg=7e862db49bbc8bde2b49dee686cce68c8d2ab9f0 (from engibench[all]->engiopt==0.0.1)\n", - " Cloning https://github.com/mdolab/pyoptsparse to /private/var/folders/fw/69hmbxm946l1xycg8w29nw0800l9gd/T/pip-install-zxekdyn0/pyoptsparse_f7cbacf77ad341919fb2519347802ce5\n", - " Running command git clone --filter=blob:none --quiet https://github.com/mdolab/pyoptsparse /private/var/folders/fw/69hmbxm946l1xycg8w29nw0800l9gd/T/pip-install-zxekdyn0/pyoptsparse_f7cbacf77ad341919fb2519347802ce5\n", - " Resolved https://github.com/mdolab/pyoptsparse to commit 9d417c740693dc706fef8ff2c064837fea04ec50\n", - " Installing build dependencies ... \u001b[?25ldone\n", - "\u001b[?25h Getting requirements to build wheel ... \u001b[?25ldone\n", - "\u001b[?25h Preparing metadata (pyproject.toml) ... \u001b[?25ldone\n", - "\u001b[?25hRequirement already satisfied: cvxopt>=1.3.2 in ./.venv/lib/python3.12/site-packages (from engibench[all]->engiopt==0.0.1) (1.3.2)\n", - "Requirement already satisfied: seaborn in ./.venv/lib/python3.12/site-packages (from engibench[all]->engiopt==0.0.1) (0.13.2)\n", - "Requirement already satisfied: mmapy>=0.3.0 in ./.venv/lib/python3.12/site-packages (from engibench[all]->engiopt==0.0.1) (0.3.0)\n", - "Requirement already satisfied: ceviche>=0.1.3 in ./.venv/lib/python3.12/site-packages (from engibench[all]->engiopt==0.0.1) (0.1.3)\n", - "Requirement already satisfied: zipp>=3.20 in ./.venv/lib/python3.12/site-packages (from importlib-metadata->diffusers>=0.33.1->engiopt==0.0.1) (3.21.0)\n", - "Requirement already satisfied: comm>=0.1.3 in ./.venv/lib/python3.12/site-packages (from ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (0.2.2)\n", - "Requirement already satisfied: ipython>=6.1.0 in ./.venv/lib/python3.12/site-packages (from ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (9.2.0)\n", - "Requirement already satisfied: traitlets>=4.3.1 in ./.venv/lib/python3.12/site-packages (from ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (5.14.3)\n", - "Requirement already satisfied: widgetsnbextension~=4.0.14 in ./.venv/lib/python3.12/site-packages (from ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (4.0.14)\n", - "Requirement already satisfied: jupyterlab_widgets~=3.0.14 in ./.venv/lib/python3.12/site-packages (from ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (3.0.14)\n", - "Requirement already satisfied: MarkupSafe>=2.0 in ./.venv/lib/python3.12/site-packages (from jinja2->ax-platform>=0.5->engiopt==0.0.1) (3.0.2)\n", - "Requirement already satisfied: typing-inspect in ./.venv/lib/python3.12/site-packages (from pyre-extensions->ax-platform>=0.5->engiopt==0.0.1) (0.9.0)\n", - "Requirement already satisfied: pyMKL>=0.0.3 in ./.venv/lib/python3.12/site-packages (from ceviche>=0.1.3->engibench[all]->engiopt==0.0.1) (0.0.3)\n", - "Requirement already satisfied: aiohappyeyeballs>=2.3.0 in ./.venv/lib/python3.12/site-packages (from aiohttp->datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (2.4.3)\n", - "Requirement already satisfied: aiosignal>=1.1.2 in ./.venv/lib/python3.12/site-packages (from aiohttp->datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (1.3.1)\n", - "Requirement already satisfied: attrs>=17.3.0 in ./.venv/lib/python3.12/site-packages (from aiohttp->datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (24.2.0)\n", - "Requirement already satisfied: frozenlist>=1.1.1 in ./.venv/lib/python3.12/site-packages (from aiohttp->datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (1.5.0)\n", - "Requirement already satisfied: multidict<7.0,>=4.5 in ./.venv/lib/python3.12/site-packages (from aiohttp->datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (6.1.0)\n", - "Requirement already satisfied: propcache>=0.2.0 in ./.venv/lib/python3.12/site-packages (from aiohttp->datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (0.2.0)\n", - "Requirement already satisfied: yarl<2.0,>=1.17.0 in ./.venv/lib/python3.12/site-packages (from aiohttp->datasets>=3.1.0->datasets[vision]>=3.1.0->engibench[all]->engiopt==0.0.1) (1.18.0)\n", - "Requirement already satisfied: smmap<6,>=3.0.1 in ./.venv/lib/python3.12/site-packages (from gitdb<5,>=4.0.1->gitpython!=3.1.29,>=1.0.0->wandb>=0.18.7->engiopt==0.0.1) (5.0.1)\n", - "Requirement already satisfied: decorator in ./.venv/lib/python3.12/site-packages (from ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (5.2.1)\n", - "Requirement already satisfied: ipython-pygments-lexers in ./.venv/lib/python3.12/site-packages (from ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (1.1.1)\n", - "Requirement already satisfied: jedi>=0.16 in ./.venv/lib/python3.12/site-packages (from ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (0.19.2)\n", - "Requirement already satisfied: matplotlib-inline in ./.venv/lib/python3.12/site-packages (from ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (0.1.7)\n", - "Requirement already satisfied: pexpect>4.3 in ./.venv/lib/python3.12/site-packages (from ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (4.9.0)\n", - "Requirement already satisfied: prompt_toolkit<3.1.0,>=3.0.41 in ./.venv/lib/python3.12/site-packages (from ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (3.0.51)\n", - "Requirement already satisfied: stack_data in ./.venv/lib/python3.12/site-packages (from ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (0.6.3)\n", - "Requirement already satisfied: mdurl~=0.1 in ./.venv/lib/python3.12/site-packages (from markdown-it-py>=2.2.0->rich>=11.1.0->tyro>=0.9.2->engiopt==0.0.1) (0.1.2)\n", - "Requirement already satisfied: opt-einsum>=2.3.2 in ./.venv/lib/python3.12/site-packages (from pyro-ppl>=1.8.4->botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (3.4.0)\n", - "Requirement already satisfied: pyro-api>=0.1.1 in ./.venv/lib/python3.12/site-packages (from pyro-ppl>=1.8.4->botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (0.1.2)\n", - "Requirement already satisfied: sqlitedict>=1.6 in ./.venv/lib/python3.12/site-packages (from pyoptsparse@ git+https://github.com/mdolab/pyoptsparse#egg=7e862db49bbc8bde2b49dee686cce68c8d2ab9f0->engibench[all]->engiopt==0.0.1) (2.1.0)\n", - "Requirement already satisfied: mdolab-baseclasses>=1.3.1 in ./.venv/lib/python3.12/site-packages (from pyoptsparse@ git+https://github.com/mdolab/pyoptsparse#egg=7e862db49bbc8bde2b49dee686cce68c8d2ab9f0->engibench[all]->engiopt==0.0.1) (1.8.0)\n", - "Requirement already satisfied: mypy-extensions>=0.3.0 in ./.venv/lib/python3.12/site-packages (from typing-inspect->pyre-extensions->ax-platform>=0.5->engiopt==0.0.1) (1.1.0)\n", - "Requirement already satisfied: parso<0.9.0,>=0.8.4 in ./.venv/lib/python3.12/site-packages (from jedi>=0.16->ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (0.8.4)\n", - "Requirement already satisfied: ptyprocess>=0.5 in ./.venv/lib/python3.12/site-packages (from pexpect>4.3->ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (0.7.0)\n", - "Requirement already satisfied: wcwidth in ./.venv/lib/python3.12/site-packages (from prompt_toolkit<3.1.0,>=3.0.41->ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (0.2.13)\n", - "Requirement already satisfied: future in ./.venv/lib/python3.12/site-packages (from pyMKL>=0.0.3->ceviche>=0.1.3->engibench[all]->engiopt==0.0.1) (1.0.0)\n", - "Requirement already satisfied: wadler-lindig>=0.1.3 in ./.venv/lib/python3.12/site-packages (from jaxtyping->gpytorch==1.14->botorch==0.13.0->ax-platform>=0.5->engiopt==0.0.1) (0.1.5)\n", - "Requirement already satisfied: executing>=1.2.0 in ./.venv/lib/python3.12/site-packages (from stack_data->ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (2.2.0)\n", - "Requirement already satisfied: asttokens>=2.1.0 in ./.venv/lib/python3.12/site-packages (from stack_data->ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (3.0.0)\n", - "Requirement already satisfied: pure-eval in ./.venv/lib/python3.12/site-packages (from stack_data->ipython>=6.1.0->ipywidgets->ax-platform>=0.5->engiopt==0.0.1) (0.2.3)\n", - "Building wheels for collected packages: engiopt\n", - " Building editable for engiopt (pyproject.toml) ... \u001b[?25ldone\n", - "\u001b[?25h Created wheel for engiopt: filename=engiopt-0.0.1-0.editable-py3-none-any.whl size=16828 sha256=d2bc822e152627edd5203df85f86232675450ffc27a4116b36a094823796fa5a\n", - " Stored in directory: /private/var/folders/fw/69hmbxm946l1xycg8w29nw0800l9gd/T/pip-ephem-wheel-cache-catulojk/wheels/b6/44/39/3c3f0715f4343d6ef24ce1d8f1606494f927246a0d3d312dcc\n", - "Successfully built engiopt\n", - "Installing collected packages: engiopt\n", - " Attempting uninstall: engiopt\n", - " Found existing installation: engiopt 0.0.1\n", - " Uninstalling engiopt-0.0.1:\n", - " Successfully uninstalled engiopt-0.0.1\n", - "Successfully installed engiopt-0.0.1\n", - "\n", - "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m24.2\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m25.1.1\u001b[0m\n", - "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n", - "Note: you may need to restart the kernel to use updated packages.\n" - ] - } - ], + "outputs": [], "source": [ "%pip install -e ." ] }, { "cell_type": "code", - "execution_count": 2, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/gymnasium/spaces/box.py:235: UserWarning: \u001b[33mWARN: Box low's precision lowered by casting to float32, current low.dtype=float64\u001b[0m\n", - " gym.logger.warn(\n", - "/Users/ffelte/Documents/EngiLearn/.venv/lib/python3.12/site-packages/gymnasium/spaces/box.py:305: UserWarning: \u001b[33mWARN: Box high's precision lowered by casting to float32, current high.dtype=float64\u001b[0m\n", - " gym.logger.warn(\n" - ] - } - ], + "outputs": [], "source": [ "# Some imports we will need\n", "import os\n", @@ -435,21 +249,9 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "\u001b[34m\u001b[1mwandb\u001b[0m: Downloading large artifact heatconduction2d_diffusion_2d_cond_model:seed_1, 200.90MB. 1 files... \n", - "\u001b[34m\u001b[1mwandb\u001b[0m: 1 of 1 files downloaded. \n", - "Done. 0:0:0.8\n", - "/var/folders/fw/69hmbxm946l1xycg8w29nw0800l9gd/T/ipykernel_6077/97458895.py:20: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n", - " ckpt = th.load(ckpt_path, map_location=device)\n" - ] - } - ], + "outputs": [], "source": [ "## Downloading a model from wandb\n", "problem_id = \"heatconduction2d\"\n", @@ -555,7 +357,6 @@ } ], "source": [ - "\n", "n_samples = 3\n", "\n", "conditions = [\n", From 0ce353acfe6ac1791fd65d63654e905320adcb22 Mon Sep 17 00:00:00 2001 From: Florian Felten Date: Wed, 9 Jul 2025 14:20:49 +0200 Subject: [PATCH 5/5] Remove some outputs from saved notebook --- example_hard_model.ipynb | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/example_hard_model.ipynb b/example_hard_model.ipynb index 9680c084..4c4f3ad8 100644 --- a/example_hard_model.ipynb +++ b/example_hard_model.ipynb @@ -376,7 +376,7 @@ " gen_designs = ddm_sampler.sample_timestep(model, gen_designs, t, conditions_tensor)\n", "\n", "gen_designs = gen_designs.cpu().numpy()\n", - "gen_designs = gen_designs.squeeze() # Remove the channel dim\n", + "gen_designs = gen_designs.squeeze() # Remove the channel dim\n", "\n", "# Render the designs\n", "for i in range(n_samples):\n", @@ -6331,7 +6331,6 @@ } ], "source": [ - "\n", "opt_designs = []\n", "opt_histories = []\n", "for i in range(n_samples):\n", @@ -6419,7 +6418,6 @@ } ], "source": [ - "\n", "for i in range(n_samples):\n", " fig_opt, ax_opt = problem.render(opt_designs[i])\n", " fig_gen, ax_gen = problem.render(gen_designs[i])\n", @@ -6879,7 +6877,6 @@ } ], "source": [ - "\n", "objective_values_opt = [problem.simulate(opt_designs[i], config=conditions[i]) for i in range(n_samples)]\n", "optimality_gaps = [metrics.optimality_gap(opt_histories[i], baseline=objective_values_opt[i]) for i in range(n_samples)]\n", "\n",