plot_local_explanations_gradient

Plots the contribution of each covariate, and the prediction, with error bars.

Latent binary Bayesian neural networks (LBBNNs) are implemented using
'torch', an R interface to the LibTorch backend. Supports mean-field variational
inference as well as flexible variational posteriors using normalizing flows.
The standard LBBNN implementation follows Hubin and Storvik (2024) <doi:10.3390/math12060788>,
using the local reparametrization trick as in Skaaret-Lund et al. (2024)
<https://openreview.net/pdf?id=d6kqUKzG3V>. Input-skip connections are also supported,
as described in Høyheim et al. (2025) <doi:10.48550/arXiv.2503.10496>.

Lars Skaaret-Lund

LBBNN

Latent Binary Bayesian Neural Networks Using 'torch'

Aliaksandr Hubin

Eirik Høyheim

plot_local_explanations_gradient function

<dl><dt>model</dt>
<dd>An instance of <code>LBBNN_Net</code> with input-skip enabled.</dd>
<dt>input_data</dt>
<dd>The data to be explained (one sample).</dd>
<dt>num_samples</dt>
<dd>integer, how many sample to use to produce credible intervals.</dd>
<dt>device</dt>
<dd>character, the device to be trained on. Default is cpu. Can be 'mps' or 'gpu'.</dd>
<dt>save_svg</dt>
<dd>the path where the plot will be saved as svg, if save_svg is not NULL.</dd></dl>

Arguments

Plot the gradient based local explanations for one sample with input-skip LBBNNs. — plot_local_explanations_gradient

<dl>

<dt>model</dt>
<dd>An instance of <code>LBBNN_Net</code> with input-skip enabled.</dd>


<dt>input_data</dt>
<dd>The data to be explained (one sample).</dd>


<dt>num_samples</dt>
<dd>integer, how many sample to use to produce credible intervals.</dd>


<dt>device</dt>
<dd>character, the device to be trained on. Default is cpu. Can be 'mps' or 'gpu'.</dd>


<dt>save_svg</dt>
<dd>the path where the plot will be saved as svg, if save_svg is not NULL.</dd>

</dl>

plot_local_explanations_gradient: Plot the gradient based local explanations for one sample with input-skip LBBNNs.

Description

Usage

Value

Arguments