Skip to main content

EchoStateNetwork

DOI PyPI Docs

Echo state networks / reservoir computing in pure numpy. One class, EchoStateNetwork: ridge-regression training with chaotic recycle validation (contiguous runs or ragged dwell segments), parametric inputs, optional Bayesian hyperparameter search (scikit-optimize), closed-loop prediction, and Jacobians for data assimilation. For spatially extended systems, the parallel layout (patch_size, halo, periodic) runs one reservoir per patch of neighbouring sites (Pathak et al. 2018), with shared matrices and a position for every reservoir row, which covariance localization requires; see the parallel ESN page.

This is the single shared reservoir core behind qlroms (quantized-local ESN families) and romda (real-time ESN forecasting and bias-aware data assimilation).

Documentation: https://andreanovoa.github.io/EchoStateNetwork/

Tutorials

Install

pip install echostatenetwork          # core (numpy/scipy/matplotlib)
pip install "echostatenetwork[opt]"   # + scikit-optimize for Bayesian hyperparameter search
pip install -e ".[dev]"               # development

Quickstart

import numpy as np
from echostatenetwork import EchoStateNetwork

y = my_time_series                       # (Nt, N_dim)
esn = EchoStateNetwork(y[:1].T, dt=0.01, N_units=200, t_train=40.0, t_val=4.0)   # y[:1].T sets N_dim
esn.train(y)
u_wash, r = ...                          # see docstrings: washout then closed loop

Citation

If this package contributes to your work, please cite the software archive on Zenodo:

@software{novoa_echostatenetwork,
  author = {Nóvoa, Andrea},
  title = {echostatenetwork: echo state networks in pure numpy},
  publisher = {Zenodo},
  doi = {10.5281/zenodo.23266679},
  url = {https://doi.org/10.5281/zenodo.23266679},
}

and the paper that introduced chaotic recycle validation, which train() employs by default:

@article{racca2021robust,
  author = {Racca, Alberto and Magri, Luca},
  title = {Robust optimization and validation of echo state networks for learning chaotic dynamics},
  journal = {Neural Networks},
  volume = {142},
  pages = {252--268},
  year = {2021},
  doi = {10.1016/j.neunet.2021.05.004},
  url = {https://doi.org/10.1016/j.neunet.2021.05.004},
}

Contributors

Acknowledgements

This repository is based on alberacca/Echo-State-Networks, the reference implementation of the recycle-validation echo state network by Racca & Magri (2021). The parallel layout follows Pathak et al. (2018) and Vlachas et al. (2020).

  • Pathak, J., Hunt, B., Girvan, M., Lu, Z., & Ott, E. (2018). Model-free prediction of large spatiotemporally chaotic systems from data: a reservoir computing approach. Physical Review Letters, 120, 024102. doi:10.1103/PhysRevLett.120.024102
  • Vlachas, P. R., Pathak, J., Hunt, B. R., Sapsis, T. P., Girvan, M., Ott, E., & Koumoutsakos, P. (2020). Backpropagation algorithms and reservoir computing in recurrent neural networks for the forecasting of complex spatiotemporal dynamics. Neural Networks, 126, 191-217. doi:10.1016/j.neunet.2020.02.016

Metadata

Release files for echostatenetwork 0.4.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for echostatenetwork 0.4.0
File Size Uploaded Provenance Actions
echostatenetwork-0.4.0.tar.gz 104.3 kB With provenance Details for echostatenetwork-0.4.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for echostatenetwork 0.4.0
File Interpreter ABI Platform Actions
echostatenetwork-0.4.0-py3-none-any.whl Python 3 none any Details for echostatenetwork-0.4.0-py3-none-any.whl

Total release size: 175.1 kB

Release files / echostatenetwork-0.4.0.tar.gz

Download URL echostatenetwork-0.4.0.tar.gz
Size 104.3 kB
Tags Source
SHA-256 checksum
How to use checksums
3c764234c78ca1e578969ec29f6b6f04263ab2e0e13e2fd0cbcf585329822502
BLAKE2b-256 checksum
How to use checksums
210140dcc8162e13c5d8bee36a1b8aba9a0516564d1d920e6d1955d2c1ea5403
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 11, 2026.

Transparency log

Release files / echostatenetwork-0.4.0-py3-none-any.whl

Download URL echostatenetwork-0.4.0-py3-none-any.whl
Size 70.7 kB
Tags Python 3
SHA-256 checksum
How to use checksums
1d2e5b6b4dadf3106b881dc104327e3bde19fc36a7682537067969cf471203ee
BLAKE2b-256 checksum
How to use checksums
22dfaf9038854cc036a2e3c3fea3100d4c8e6234882faadfc906fd843df3e571
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 11, 2026.

Transparency log

Security for echostatenetwork 0.4.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

Both files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for echostatenetwork-0.4.0.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 11, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page