From 3fe5894181e6c845b590465053912d19171c8916 Mon Sep 17 00:00:00 2001
From: alexwong <11878166+alexwong@users.noreply.github.com>
Date: Mon, 24 Feb 2020 21:57:06 -0800
Subject: [PATCH 1/8] Add a tutorial for PyTorch

---
 tutorials/frontend/from_pytorch.py | 166 +++++++++++++++++++++++++++++
 1 file changed, 166 insertions(+)
 create mode 100644 tutorials/frontend/from_pytorch.py

diff --git a/tutorials/frontend/from_pytorch.py b/tutorials/frontend/from_pytorch.py
new file mode 100644
index 000000000000..675cf06d8a77
--- /dev/null
+++ b/tutorials/frontend/from_pytorch.py
@@ -0,0 +1,166 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License.  You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied.  See the License for the
+# specific language governing permissions and limitations
+# under the License.
+"""
+Compile PyTorch Models
+=====================
+**Author**: `Alex Wong <https://github.com/alexwong/>`_
+
+This article is an introductory tutorial to deploy PyTorch models with Relay.
+
+For us to begin with, PyTorch should be installed.
+TorchVision is also required since we will be using it as our model zoo.
+
+A quick solution is to install via pip
+
+.. code-block:: bash
+
+    pip install torch==1.4.0
+    pip install torchvision==0.5.0
+
+or please refer to official site
+https://pytorch.org/get-started/locally/
+"""
+
+# tvm, relay
+import tvm
+from tvm import relay
+
+# numpy, packaging
+import numpy as np
+from packaging import version
+from tvm.contrib.download import download_testdata
+
+# PyTorch imports
+import torch
+import torchvision
+if (version.parse(torch.__version__) > version.parse("1.4.0")) \
+    or (version.parse(torchvision.__version__) > version.parse("0.5.0")):
+    assert "Please ensure version of PyTorch is supported by TVM"
+
+######################################################################
+# Load a pretrained PyTorch model
+# ------------
+model_name = 'resnet18'
+model = getattr(torchvision.models, model_name)(pretrained=True)
+model = model.float().eval()
+
+# We grab the TorchScripted model via tracing
+input_shape = [1, 3, 224, 224]
+input_data = torch.randn(input_shape).float()
+scripted_model = torch.jit.trace(model, input_data).float().eval()
+
+######################################################################
+# Load a test image
+# ------------------
+# Classic cat example!
+from PIL import Image
+img_url = 'https://raw.githubusercontent.com/Cadene/pretrained-models.pytorch/master/data/cat_224.jpg'
+img_path = download_testdata(img_url, 'cat_224.png', module='data')
+img = Image.open(img_path)
+
+# Preprocess the image and convert to tensor
+from torchvision import transforms
+my_preprocess = transforms.Compose([
+    transforms.Resize(256),
+    transforms.CenterCrop(224),
+    transforms.ToTensor(),
+    transforms.Normalize(mean=[0.485, 0.456, 0.406],
+                         std=[0.229, 0.224, 0.225])
+])
+img = my_preprocess(img)
+img = np.expand_dims(img, 0)
+
+######################################################################
+# Import the graph to Relay
+# -------------------------
+# Convert PyTorch graph to Relay graph.
+shape_dict = {'input0': img.shape}
+mod, params = relay.frontend.from_pytorch(scripted_model,
+                                          shape_dict)
+
+######################################################################
+# Relay Build
+# -----------
+# Compile the graph to llvm target with given input specification.
+target = 'llvm'
+target_host = 'llvm'
+layout = None
+ctx = tvm.cpu(0)
+with relay.build_config(opt_level=3):
+    graph, lib, params = relay.build(mod,
+                                     target=target,
+                                     target_host=target_host,
+                                     params=params)
+
+######################################################################
+# Execute the portable graph on TVM
+# ---------------------------------
+# Now we can try deploying the compiled model on target.
+
+from tvm.contrib import graph_runtime
+dtype = 'float32'
+m = graph_runtime.create(graph, lib, ctx)
+# Set inputs
+m.set_input('input0', tvm.nd.array(img.astype(dtype)))
+m.set_input(**params)
+# Execute
+m.run()
+# Get outputs
+tvm_output = m.get_output(0, tvm.nd.empty(((1, 1000)), 'float32'))
+
+#####################################################################
+# Look up synset name
+# -------------------
+# Look up prediction top 1 index in 1000 class synset.
+synset_url = ''.join(['https://raw.githubusercontent.com/Cadene/',
+                      'pretrained-models.pytorch/master/data/',
+                      'imagenet_synsets.txt'])
+synset_name = 'imagenet_synsets.txt'
+synset_path = download_testdata(synset_url, synset_name, module='data')
+with open(synset_path) as f:
+    synsets = f.readlines()
+
+synsets = [x.strip() for x in synsets]
+splits = [line.split(' ') for line in synsets]
+key_to_classname = {spl[0]:' '.join(spl[1:]) for spl in splits}
+
+class_url = ''.join(['https://raw.githubusercontent.com/Cadene/',
+                      'pretrained-models.pytorch/master/data/',
+                      'imagenet_classes.txt'])
+class_name = 'imagenet_classes.txt'
+class_path = download_testdata(class_url, class_name, module='data')
+with open(class_path) as f:
+    class_id_to_key = f.readlines()
+
+class_id_to_key = [x.strip() for x in class_id_to_key]
+
+# Get top-1 result for TVM
+top1_tvm = np.argmax(tvm_output.asnumpy()[0])
+tvm_class_key = class_id_to_key[top1_tvm]
+
+# Convert input to PyTorch variable and get PyTorch result for comparison
+torch_img = torch.from_numpy(img)
+from torch.autograd import Variable
+torch_img = Variable(torch_img).float()
+output = model(torch_img)
+
+# Get top-1 result for PyTorch
+top1_torch = np.argmax(output.detach().numpy())
+torch_class_key = class_id_to_key[top1_torch]
+
+print('Relay top-1 id: {}, class name: {}'.format(top1_tvm, key_to_classname[tvm_class_key]))
+print('Torch top-1 id: {}, class name: {}'.format(top1_torch, key_to_classname[torch_class_key]))
\ No newline at end of file

From dac502154e5494db7392ed7a23085d255193bfb4 Mon Sep 17 00:00:00 2001
From: alexwong <11878166+alexwong@users.noreply.github.com>
Date: Tue, 25 Feb 2020 11:20:57 -0800
Subject: [PATCH 2/8] Fix sphinx formatting, add version support

---
 tutorials/frontend/from_pytorch.py | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/tutorials/frontend/from_pytorch.py b/tutorials/frontend/from_pytorch.py
index 675cf06d8a77..0d11e3e04459 100644
--- a/tutorials/frontend/from_pytorch.py
+++ b/tutorials/frontend/from_pytorch.py
@@ -16,7 +16,7 @@
 # under the License.
 """
 Compile PyTorch Models
-=====================
+======================
 **Author**: `Alex Wong <https://github.com/alexwong/>`_
 
 This article is an introductory tutorial to deploy PyTorch models with Relay.
@@ -33,6 +33,12 @@
 
 or please refer to official site
 https://pytorch.org/get-started/locally/
+
+PyTorch versions should be backwards compatible but should be used
+with the proper TorchVision version.
+
+Currently, TVM supports PyTorch 1.4, 1.3, and 1.2. Other versions may
+be unstable.
 """
 
 # tvm, relay
@@ -53,7 +59,7 @@
 
 ######################################################################
 # Load a pretrained PyTorch model
-# ------------
+# -------------------------------
 model_name = 'resnet18'
 model = getattr(torchvision.models, model_name)(pretrained=True)
 model = model.float().eval()
@@ -65,7 +71,7 @@
 
 ######################################################################
 # Load a test image
-# ------------------
+# -----------------
 # Classic cat example!
 from PIL import Image
 img_url = 'https://raw.githubusercontent.com/Cadene/pretrained-models.pytorch/master/data/cat_224.jpg'

From 72769430f2a56b605c2ab43cb925bc0ffb90c8d2 Mon Sep 17 00:00:00 2001
From: alexwong <11878166+alexwong@users.noreply.github.com>
Date: Tue, 25 Feb 2020 11:25:20 -0800
Subject: [PATCH 3/8] Remove space

---
 tutorials/frontend/from_pytorch.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/tutorials/frontend/from_pytorch.py b/tutorials/frontend/from_pytorch.py
index 0d11e3e04459..01cba505846c 100644
--- a/tutorials/frontend/from_pytorch.py
+++ b/tutorials/frontend/from_pytorch.py
@@ -116,7 +116,6 @@
 # Execute the portable graph on TVM
 # ---------------------------------
 # Now we can try deploying the compiled model on target.
-
 from tvm.contrib import graph_runtime
 dtype = 'float32'
 m = graph_runtime.create(graph, lib, ctx)

From 85368d36f2eda0767d4efb04649be758abf8dada Mon Sep 17 00:00:00 2001
From: alexwong <11878166+alexwong@users.noreply.github.com>
Date: Tue, 25 Feb 2020 15:31:49 -0800
Subject: [PATCH 4/8] Remove version check

---
 tutorials/frontend/from_pytorch.py | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/tutorials/frontend/from_pytorch.py b/tutorials/frontend/from_pytorch.py
index 01cba505846c..8f3e708eb79f 100644
--- a/tutorials/frontend/from_pytorch.py
+++ b/tutorials/frontend/from_pytorch.py
@@ -53,9 +53,6 @@
 # PyTorch imports
 import torch
 import torchvision
-if (version.parse(torch.__version__) > version.parse("1.4.0")) \
-    or (version.parse(torchvision.__version__) > version.parse("0.5.0")):
-    assert "Please ensure version of PyTorch is supported by TVM"
 
 ######################################################################
 # Load a pretrained PyTorch model

From 2ba82c4ccf0892d950777f5b77bad3f963c2a64b Mon Sep 17 00:00:00 2001
From: alexwong <11878166+alexwong@users.noreply.github.com>
Date: Tue, 25 Feb 2020 16:16:35 -0800
Subject: [PATCH 5/8] Some refactoring

---
 tutorials/frontend/from_pytorch.py | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/tutorials/frontend/from_pytorch.py b/tutorials/frontend/from_pytorch.py
index 8f3e708eb79f..d58c3b7a76fe 100644
--- a/tutorials/frontend/from_pytorch.py
+++ b/tutorials/frontend/from_pytorch.py
@@ -59,12 +59,12 @@
 # -------------------------------
 model_name = 'resnet18'
 model = getattr(torchvision.models, model_name)(pretrained=True)
-model = model.float().eval()
+model = model.eval()
 
 # We grab the TorchScripted model via tracing
 input_shape = [1, 3, 224, 224]
-input_data = torch.randn(input_shape).float()
-scripted_model = torch.jit.trace(model, input_data).float().eval()
+input_data = torch.randn(input_shape)
+scripted_model = torch.jit.trace(model, input_data).eval()
 
 ######################################################################
 # Load a test image
@@ -101,7 +101,6 @@
 # Compile the graph to llvm target with given input specification.
 target = 'llvm'
 target_host = 'llvm'
-layout = None
 ctx = tvm.cpu(0)
 with relay.build_config(opt_level=3):
     graph, lib, params = relay.build(mod,
@@ -156,8 +155,6 @@
 
 # Convert input to PyTorch variable and get PyTorch result for comparison
 torch_img = torch.from_numpy(img)
-from torch.autograd import Variable
-torch_img = Variable(torch_img).float()
 output = model(torch_img)
 
 # Get top-1 result for PyTorch

From ef416260ac355a07bd2c80aee1e11ab153ae0dd2 Mon Sep 17 00:00:00 2001
From: alexwong <11878166+alexwong@users.noreply.github.com>
Date: Tue, 25 Feb 2020 16:23:55 -0800
Subject: [PATCH 6/8] Use no grad

---
 tutorials/frontend/from_pytorch.py | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/tutorials/frontend/from_pytorch.py b/tutorials/frontend/from_pytorch.py
index d58c3b7a76fe..dc92e835999b 100644
--- a/tutorials/frontend/from_pytorch.py
+++ b/tutorials/frontend/from_pytorch.py
@@ -154,12 +154,13 @@
 tvm_class_key = class_id_to_key[top1_tvm]
 
 # Convert input to PyTorch variable and get PyTorch result for comparison
-torch_img = torch.from_numpy(img)
-output = model(torch_img)
+with torch.no_grad():
+    torch_img = torch.from_numpy(img)
+    output = model(torch_img)
 
-# Get top-1 result for PyTorch
-top1_torch = np.argmax(output.detach().numpy())
-torch_class_key = class_id_to_key[top1_torch]
+    # Get top-1 result for PyTorch
+    top1_torch = np.argmax(output.numpy())
+    torch_class_key = class_id_to_key[top1_torch]
 
 print('Relay top-1 id: {}, class name: {}'.format(top1_tvm, key_to_classname[tvm_class_key]))
 print('Torch top-1 id: {}, class name: {}'.format(top1_torch, key_to_classname[torch_class_key]))
\ No newline at end of file

From 3097ddbd146f9182bb143cc45233f9590469ff73 Mon Sep 17 00:00:00 2001
From: alexwong <11878166+alexwong@users.noreply.github.com>
Date: Tue, 25 Feb 2020 16:27:47 -0800
Subject: [PATCH 7/8] Rename input

---
 tutorials/frontend/from_pytorch.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tutorials/frontend/from_pytorch.py b/tutorials/frontend/from_pytorch.py
index dc92e835999b..22c649712af0 100644
--- a/tutorials/frontend/from_pytorch.py
+++ b/tutorials/frontend/from_pytorch.py
@@ -91,7 +91,7 @@
 # Import the graph to Relay
 # -------------------------
 # Convert PyTorch graph to Relay graph.
-shape_dict = {'input0': img.shape}
+shape_dict = {'img': img.shape}
 mod, params = relay.frontend.from_pytorch(scripted_model,
                                           shape_dict)
 
@@ -116,7 +116,7 @@
 dtype = 'float32'
 m = graph_runtime.create(graph, lib, ctx)
 # Set inputs
-m.set_input('input0', tvm.nd.array(img.astype(dtype)))
+m.set_input('img', tvm.nd.array(img.astype(dtype)))
 m.set_input(**params)
 # Execute
 m.run()

From 38464d7f9fd478266a7b35d2931ca002b6032cdd Mon Sep 17 00:00:00 2001
From: alexwong <11878166+alexwong@users.noreply.github.com>
Date: Tue, 25 Feb 2020 16:44:02 -0800
Subject: [PATCH 8/8] Update cat img source

---
 tutorials/frontend/from_pytorch.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/tutorials/frontend/from_pytorch.py b/tutorials/frontend/from_pytorch.py
index 22c649712af0..c280c259c1fe 100644
--- a/tutorials/frontend/from_pytorch.py
+++ b/tutorials/frontend/from_pytorch.py
@@ -71,9 +71,9 @@
 # -----------------
 # Classic cat example!
 from PIL import Image
-img_url = 'https://raw.githubusercontent.com/Cadene/pretrained-models.pytorch/master/data/cat_224.jpg'
-img_path = download_testdata(img_url, 'cat_224.png', module='data')
-img = Image.open(img_path)
+img_url = 'https://github.com/dmlc/mxnet.js/blob/master/data/cat.png?raw=true'
+img_path = download_testdata(img_url, 'cat.png', module='data')
+img = Image.open(img_path).resize((224, 224))
 
 # Preprocess the image and convert to tensor
 from torchvision import transforms