categorical_to_one_hot_layer

The <code>categorical_to_one_hot_layer</code> module/layer expands categorical features into one-hot vectors,
because multi-layer perceptrons are known to work better with this data representation.
It also replaces NaNs with zeros in order so that further layers may work correctly.

internal

Complex machine learning models are often hard to interpret. However, in
many situations it is crucial to understand and explain why a model made a specific
prediction. Shapley values is the only method for such prediction explanation framework
with a solid theoretical foundation. Previously known methods for estimating the Shapley
values do, however, assume feature independence. This package implements methods which accounts for any feature
dependence, and thereby produces more accurate estimates of the true Shapley values.
An accompanying 'Python' wrapper ('shaprpy') is available through the GitHub repository.

Martin Jullum

shapr

Prediction Explanation with Dependence-Aware Shapley Values

Lars Henry Berge Olsen

Annabelle Redelmeier

Jon Lachmann

Nikolai Sellereite

Anders Løland

Jens Christian Wahl

Camilla Lingjærde

Norsk Regnesentral 

categorical_to_one_hot_layer function

<dl><dt>one_hot_max_sizes</dt>
<dd>A torch tensor of dimension <code>n_features</code> containing the one hot sizes of the <code>n_features</code>
features. That is, if the <code>i</code>th feature is a categorical feature with 5 levels, then <code>one_hot_max_sizes[i] = 5</code>.
While the size for continuous features can either be <code>0</code> or <code>1</code>.</dd>
<dt>add_nans_map_for_columns</dt>
<dd>Optional list which contains indices of columns which is_nan masks are to be appended
to the result tensor. This option is necessary for the full encoder to distinguish whether value is to be
reconstructed or not.</dd></dl>

Arguments

Author

A torch::nn_module() Representing a categorical_to_one_hot_layer — categorical_to_one_hot_layer

<dl>

<dt>one_hot_max_sizes</dt>
<dd>A torch tensor of dimension <code>n_features</code> containing the one hot sizes of the <code>n_features</code>
features. That is, if the <code>i</code>th feature is a categorical feature with 5 levels, then <code>one_hot_max_sizes[i] = 5</code>.
While the size for continuous features can either be <code>0</code> or <code>1</code>.</dd>


<dt>add_nans_map_for_columns</dt>
<dd>Optional list which contains indices of columns which is_nan masks are to be appended
to the result tensor. This option is necessary for the full encoder to distinguish whether value is to be
reconstructed or not.</dd>

</dl>

A <code>torch::nn_module()</code> Representing a <code>categorical_to_one_hot_layer</code>

Get 50% off unlimited learning

categorical_to_one_hot_layer: A `torch::nn_module()` Representing a `categorical_to_one_hot_layer`

Description

Usage

Arguments

Author

Details