prepare_data_gaussian_cpp_caus

Generate Gaussian MC samples for the causal setup with a single MC sample for each explicand

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 

prepare_data_gaussian_cpp_caus function

<dl><dt>MC_samples_mat</dt>
<dd>arma::mat.
Matrix of dimension (<code>n_MC_samples</code>, <code>n_features</code>) containing samples from the univariate standard normal.</dd>
<dt>x_explain_mat</dt>
<dd>arma::mat.
Matrix of dimension (<code>n_explain</code>, <code>n_features</code>) containing the observations to explain.</dd>
<dt>S</dt>
<dd>arma::mat.
Matrix of dimension (<code>n_coalitions</code>, <code>n_features</code>) containing binary representations of the used coalitions.
S cannot contain the empty or grand coalition, i.e., a row containing only zeros or ones.
This is not a problem internally in shapr as the empty and grand coalitions are treated differently.</dd>
<dt>mu</dt>
<dd>arma::vec.
Vector of length <code>n_features</code> containing the mean of each feature.</dd>
<dt>cov_mat</dt>
<dd>arma::mat.
Matrix of dimension (<code>n_features</code>, <code>n_features</code>) containing the covariance matrix of the features.</dd></dl>

Arguments

Author

Generate Gaussian MC samples for the causal setup with a single MC sample for each explicand — prepare_data_gaussian_cpp_caus

<dl>

<dt>MC_samples_mat</dt>
<dd>arma::mat.
Matrix of dimension (<code>n_MC_samples</code>, <code>n_features</code>) containing samples from the univariate standard normal.</dd>


<dt>x_explain_mat</dt>
<dd>arma::mat.
Matrix of dimension (<code>n_explain</code>, <code>n_features</code>) containing the observations to explain.</dd>


<dt>S</dt>
<dd>arma::mat.
Matrix of dimension (<code>n_coalitions</code>, <code>n_features</code>) containing binary representations of the used coalitions.
S cannot contain the empty or grand coalition, i.e., a row containing only zeros or ones.
This is not a problem internally in shapr as the empty and grand coalitions are treated differently.</dd>


<dt>mu</dt>
<dd>arma::vec.
Vector of length <code>n_features</code> containing the mean of each feature.</dd>


<dt>cov_mat</dt>
<dd>arma::mat.
Matrix of dimension (<code>n_features</code>, <code>n_features</code>) containing the covariance matrix of the features.</dd>

</dl>

Generate Gaussian MC samples for the causal setup with a single MC sample for each explicand

prepare_data_gaussian_cpp_caus: Generate Gaussian MC samples for the causal setup with a single MC sample for each explicand

Description

Usage

Value

Arguments

Author