ReservoirPy: Efficient Training of Recurrent Neural Networks for Timeseries Processing - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Poster Communications Year : 2022

ReservoirPy: Efficient Training of Recurrent Neural Networks for Timeseries Processing

Xavier Hinaut
Nathan Trouvain

Abstract

ReservoirPy is a simple user-friendly library based on Python scientific modules. It provides a flexible interface to implement efficient Reservoir Computing (RC) [2] architectures with a particular focus on Echo State Networks (ESN) [1]. Advanced features of ReservoirPy allow to improve computation time efficiency on a simple laptop compared to basic Python implementation. Some of its features are: offline and online training, parallel implementation, sparse matrix computation, fast spectral initialization, advanced learning rules (e.g. Intrinsic Plasticity) etc. It also makes possible to easily create complex architectures with multiple reservoirs (e.g. deep reservoirs), readouts, and complex feedback loops. Moreover, graphical tools are included to easily explore hyperparameters with the help of the hyperopt library. It includes several tutorials exploring exotic architectures and examples of scientific papers reproduction.
Fichier principal
Vignette du fichier
Hinaut2022_Euroscipy_Poster_Reservoirpy_v1-HAL.pdf (4.3 Mo) Télécharger le fichier
spotlight-hinaut-euroscipy2022.pdf (365.45 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03780006 , version 1 (18-09-2022)

Identifiers

  • HAL Id : hal-03780006 , version 1

Cite

Xavier Hinaut, Nathan Trouvain. ReservoirPy: Efficient Training of Recurrent Neural Networks for Timeseries Processing. EuroSciPy 2022 - 14th European Conference on Python in Science, Aug 2022, Basel, Switzerland. ⟨hal-03780006⟩
141 View
147 Download

Share

Gmail Facebook X LinkedIn More