spks

Electrophysiology analysis package for large-scale electrophysiology

! This package is under heavy development - we'll post here when stable!

This package aims to provide transparent access to the code being used to process and analyse data; the documentation will provide references and (hopefully) a description of what the code is doing in each step.

Installation

We typically install spks in environments with other packages so kept dependencies to a minimum

Dependencies:

• numpy - array handling and loading
• scipy - interpolation and other
• matplotlib - plotting
• torch - speed up some tasks
• pandas - make tables and save files
• h5py - save dictionaries
• natsort - sort files
• tqdm - progress bars
• joblib - multiprocess pools

Installation for the brave:

1. clone the repository git clone https://github.com/spkware/spks.git usually in a separate folder
2. install dependencies with pip or which ever way you want
3. go into the newly created spks folder and install with python setup.py develop

Install with anaconda in a new environment:

Recommended to avoid interfering with other packages or for users new to python.

1. install the anaconda python distribution
2. create a new environment from a terminal conda create -n spks and activate it conda activate spks
3. install torch in the environment conda install pytorch torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia and then install the dependencies conda install pandas h5py natsort tqdm scipy joblib jupyter matplotlib you can use other versions of pytorch depending on which graphics driver is installed. A gpu is not required.
4. install spks using pip pip install git+https://github.com/spkware/spks.git@main