SINABS

Getting started

Sinabs Is Not A Brain Simulator

NOTE: This project is mirrored to gitlab.com/aiCTX/sinabs and is managed by aiCTX AG.

sinabs (pytorch based library) is developed to design and implement Spiking Convolutional Neural Networks (SCNNs). The library implements several layers that are spiking equivalents of CNN layers. In addition it provides support to import CNN models implemented in keras conveniently to test their spiking equivalent implementation.

NOTE: The conversion of CNNs to SCNNs is still a subject of research and we strive to keep the library updated to the state-of-the art in addition to providing options to compare various approaches both at a high level abstraction to low level implementation details.

NOTE: This library is an alpha release state and is subject to API changes.

Installation

You can install sinabs with pip:

pip install sinabs

If you want to develop or have access to source code of sinabs, download the package from the git repository:

$ cd <to/your/software/folder>
$ git clone https://gitlab.com/aiCTX/sinabs.git>
$ cd sinabs
$ pip install -e . --user

For developers, we recommend that you install this package as a development version so that you can update the package without reinstalling the package. Checkout our quick instructional on how to create a project based on sinabs within a virtual environment using pyenv+pipenv

Documentation and Examples

https://aictx.gitlab.io/sinabs

If you would like to generate documentation locally, you can do that using sphynx.

You can generate a sphynx documentation for this package by running the the following command.

Requirements: sphinx, pandoc, nbsphinx

$ cd /path/to/sinabs/docs/
$ make html
$ firefox build/html/index.html

This will build and auto generate html documentation at sinabs/docs/build/html/index.html

License

sinabs is published under AGPL v3.0. See the LICENSE file for details.

Contributing to sinabs

Checkout CONTRIBUTING.md