https://github.com/cknd/pyesn

Echo State Networks in Python
https://github.com/cknd/pyesn

echo-state-networks python reservoir-computing

Last synced: 6 days ago
JSON representation

Echo State Networks in Python

• Host: GitHub
• URL: https://github.com/cknd/pyesn
• Owner: cknd
• License: mit
• Created: 2015-02-04T19:34:55.000Z (almost 10 years ago)
• Default Branch: master
• Last Pushed: 2020-05-06T00:54:25.000Z (over 4 years ago)
• Last Synced: 2024-12-08T10:40:47.145Z (14 days ago)
• Topics: echo-state-networks, python, reservoir-computing
• Language: Jupyter Notebook
• Homepage:
• Size: 1.14 MB
• Stars: 299
• Watchers: 12
• Forks: 121
• Open Issues: 10
• Metadata Files:
  • Readme: readme.md
  • License: license.md

Awesome Lists containing this project

README

        
# Echo State Networks in Python

[Echo State Networks](http://www.scholarpedia.org/article/Echo_state_network) are easy-to-train recurrent neural networks, a variant of [Reservoir Computing](https://en.wikipedia.org/wiki/Reservoir_computing). In some sense, these networks show how far you can get with nothing but a good weight initialisation.

This ESN implementation is relatively simple and self-contained, though it offers tricks like noise injection and teacher forcing (feedback connections), plus a zoo of dubious little hyperparameters.

However! If your aims are practical and your gradients automatic, consider using a fully trained network.

# Examples

- [learning to be a tunable frequency generator](http://nbviewer.ipython.org/github/cknd/pyESN/blob/master/freqgen.ipynb)

- [learning a Mackey-Glass system](http://nbviewer.ipython.org/github/cknd/pyESN/blob/master/mackey.ipynb)