reservoirR_fit

A simple user-friendly library based on the 'python' module 'reservoirpy'.
It provides a flexible interface to implement efficient Reservoir
Computing (RC) architectures with a particular focus on Echo State Networks
(ESN). 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. Finally, it includes several tutorials exploring
time series forecasting, classification and hyperparameter tuning. For more information
about 'reservoirpy', please see Trouvain et al. (2020) <doi:10.1007/978-3-030-61616-8_40>.
This package was developed in the framework of the University of Bordeaux’s IdEx
"Investments for the Future" program / RRI PHDS.

Thomas Ferte

reservoirnet

Reservoir Computing and Echo State Networks

Kalidou Ba

Nathan Trouvain

Rodolphe Thiebaut

Xavier Hinaut

Boris Hejblum

reservoirR_fit function

<dl><dt>node</dt>
<dd>node</dd>
<dt>X</dt>
<dd><code>array-like</code> of shape <code>[n_inputs], [series], timesteps, input_dim)</code>, optional
Input sequences dataset. If None, the method will try to fit
the parameters of the Node using the precomputed values returned
by previous call of :py:meth:<code>partial_fit</code>.</dd>
<dt>Y</dt>
<dd>array-like of shape <code>([series], timesteps, output_dim)</code>, optional
Teacher signals dataset. If None, the method will try to fit
the parameters of the Node using the precomputed values returned
by previous call of :py:meth: <code>partial_fit</code>, or to fit the Node in
an unsupervised way, if possible.</dd>
<dt>warmup</dt>
<dd>: <code>int</code>, default to 0 
Number of timesteps to consider as warmup and 
discard at the begining of each timeseries before training.</dd>
<dt>stateful</dt>
<dd>is boolen</dd>
<dt>reset</dt>
<dd>is boolean. Should the node status be reset before fitting.</dd></dl>

Arguments

Offline fitting method of a Node — reservoirR_fit

<dl>

<dt>node</dt>
<dd>node</dd>


<dt>X</dt>
<dd><code>array-like</code> of shape <code>[n_inputs], [series], timesteps, input_dim)</code>, optional
Input sequences dataset. If None, the method will try to fit
the parameters of the Node using the precomputed values returned
by previous call of :py:meth:<code>partial_fit</code>.</dd>


<dt>Y</dt>
<dd>array-like of shape <code>([series], timesteps, output_dim)</code>, optional
Teacher signals dataset. If None, the method will try to fit
the parameters of the Node using the precomputed values returned
by previous call of :py:meth: <code>partial_fit</code>, or to fit the Node in
an unsupervised way, if possible.</dd>


<dt>warmup</dt>
<dd>: <code>int</code>, default to 0 
Number of timesteps to consider as warmup and 
discard at the begining of each timeseries before training.</dd>


<dt>stateful</dt>
<dd>is boolen</dd>


<dt>reset</dt>
<dd>is boolean. Should the node status be reset before fitting.</dd>

</dl>

Offline fitting method of a Node

reservoirR_fit: Offline fitting method of a Node

Description

Usage

Value

Arguments

Examples