contraction

This is a wrapper for a loss which induces a contraction in the
latent space.

Implementation of several unsupervised neural networks,
from building their architecture to their training and evaluation. Available
networks are auto-encoders including their main variants: sparse, contractive,
denoising, robust and variational, as described in Charte et al. (2018)
<doi:10.1016/j.inffus.2017.12.007>.

David Charte

ruta

Implementation of Unsupervised Neural Architectures

Francisco Charte

Francisco Herrera

contraction function

<dl><dt>reconstruction_loss</dt>
<dd>Original reconstruction error to be combined with the
contractive loss (e.g. <code>"binary_crossentropy"</code>)</dd>
<dt>weight</dt>
<dd>Weight assigned to the contractive loss</dd></dl>

Arguments

Contractive loss — contraction

<dl>

<dt>reconstruction_loss</dt>
<dd>Original reconstruction error to be combined with the
contractive loss (e.g. <code>"binary_crossentropy"</code>)</dd>


<dt>weight</dt>
<dd>Weight assigned to the contractive loss</dd>

</dl>

contraction: Contractive loss

Description

Usage

Value

Arguments

See Also