Simple Echo State Network implementations

These minimalist self-contained source codes in different programming languages demonstrate the simplicity and power of implementing and applying Echo State Networks “from scratch”. This is arguably the simplest form of recurrent neural network learning. The source codes are intended for education and instruction, but can also be easily adapted for practical purposes.

As a demo, they learn to predict Mackey-Glass chaotic time series (delay=17) with a remarkable accuracy. The data needed for the code are available here: MackeyGlass_t17.txt (or .zip, needs unpacking).

Mackey-Glass_curveThe source codes in plain:

The program codes above are distributed under a friendly MIT license.


If you want to learn more about applying Echo State Networks take a look at my free book chapter

Mantas Lukoševičius

A practical guide to applying echo state networks Book Section

In: Grégoire Montavon; Geneviève B. Orr; Klaus-Robert Müller (Ed.): Neural Networks: Tricks of the Trade, 2nd Edition, vol. 7700, pp. 659-686, Springer, 2012, ISBN: 978-3-642-35288-1, (simple source code samples available).

Links | BibTeX

and other Reservoir Computing publications.