https://github.com/nengo/nengo-gyrus

Recursively generate large-scale Nengo models using NumPy semantics.
https://github.com/nengo/nengo-gyrus

functional-programming nengo neural-networks numpy python

Last synced: about 1 month ago
JSON representation

Recursively generate large-scale Nengo models using NumPy semantics.

• Host: GitHub
• URL: https://github.com/nengo/nengo-gyrus
• Owner: nengo
• License: other
• Created: 2021-01-08T15:09:29.000Z (about 4 years ago)
• Default Branch: master
• Last Pushed: 2021-02-13T11:36:52.000Z (almost 4 years ago)
• Last Synced: 2024-03-27T03:53:33.434Z (10 months ago)
• Topics: functional-programming, nengo, neural-networks, numpy, python
• Language: Python
• Homepage:
• Size: 300 KB
• Stars: 3
• Watchers: 5
• Forks: 0
• Open Issues: 2
• Metadata Files:
  • Readme: README.rst
  • Changelog: CHANGES.rst
  • Contributing: CONTRIBUTING.rst
  • License: LICENSE.rst

Awesome Lists containing this project

README

        
.. image:: https://i.imgur.com/xef9dhv.png

  :width: 100%

  :target: https://github.com/nengo-labs/nengo-gyrus

  :alt: Recursively generate large-scale Nengo models using NumPy semantics.

Recursively generate large-scale `Nengo `_ models using

`NumPy `_ semantics.

Quick Start

===========

.. code-block:: bash

   pip install git+https://github.com/nengo-labs/nengo-gyrus

An example of computing the square of a two-dimensional vector with Gyrus:

.. code-block:: python

  import gyrus

  import matplotlib.pyplot as plt

  import numpy as np

  u = gyrus.stimuli([np.cos, np.sin])

  x = (u ** 2).filter(0.01)

  y = np.asarray(x.run(np.pi))  # shape: (fold, time, size_out)

  plt.figure()

  plt.plot(y.squeeze(axis=-1).T)

  plt.xlabel("Time-step")

  plt.show()

.. image:: https://i.imgur.com/KyDKeyc.png

  :width: 100%

  :target: https://github.com/nengo-labs/nengo-gyrus

  :alt: Computing the square of a two-dimensional vector with Gyrus.

This code is automagically converted to `Nengo `_ and implemented

via two spiking LIF ensembles and a lowpass synapse.

Gyrus supports many common NumPy 'ufuncs', array functions, and numeric Python

operators. Thus, code can be written in a functional style using N-D arrays and then

realized as a Nengo neural network. This enables algorithms to be written in NumPy and

then compiled onto `Nengo's supported backends `_

(e.g., GPUs, microcontrollers, neuromorphic hardware, and other neural network

accelerators).

Documentation

=============

Check out and render the Jupyter notebooks located in

`docs/examples `_.

The `gyrus_overview

`_

notebook is currently the best starting point to learn the Gyrus API and see a variety

of examples.

Support

=======

Tested against ``nengo>=3.0.0`` and requires ``numpy>=1.17``.

This project is currently pre-alpha. Pull requests are welcome, as are breaking (i.e.,

reverse-incompatible) changes.

If something doesn't work quite as you thought it should, or if you have ideas for

improvements, please feel free to open up a `GitHub issue

`_ or post on the `Nengo Forum

`_.