cli_startup

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 

cli_startup function

<dl><dt>internal</dt>
<dd>List.
Not used directly, but passed through from <code>explain()</code>.</dd>
<dt>verbose</dt>
<dd>String vector or NULL.
Controls verbosity (printout detail level) via one or more of <code>"basic"</code>, <code>"progress"</code>,
<code>"convergence"</code>, <code>"shapley"</code> and <code>"vS_details"</code>.
<code>"basic"</code> (default) displays basic information about the computation and messages about parameters/checks.
<code>"progress"</code> displays where in the calculation process the function currently is.
<code>"convergence"</code> displays how close the Shapley value estimates are to convergence
(only when <code>iterative = TRUE</code>).
<code>"shapley"</code> displays intermediate Shapley value estimates and standard deviations (only when <code>iterative = TRUE</code>),
and the final estimates.
<code>"vS_details"</code> displays information about the v(S) estimates,
most relevant for <code>approach %in% c("regression_separate", "regression_surrogate", "vaeac")</code>.
<code>NULL</code> means no printout.
Any combination can be used, e.g., <code>verbose = c("basic", "vS_details")</code>.</dd></dl>

Arguments

Print Startup Messages with CLI — cli_startup

<dl>

<dt>internal</dt>
<dd>List.
Not used directly, but passed through from <code>explain()</code>.</dd>


<dt>verbose</dt>
<dd>String vector or NULL.
Controls verbosity (printout detail level) via one or more of <code>"basic"</code>, <code>"progress"</code>,
<code>"convergence"</code>, <code>"shapley"</code> and <code>"vS_details"</code>.
<code>"basic"</code> (default) displays basic information about the computation and messages about parameters/checks.
<code>"progress"</code> displays where in the calculation process the function currently is.
<code>"convergence"</code> displays how close the Shapley value estimates are to convergence
(only when <code>iterative = TRUE</code>).
<code>"shapley"</code> displays intermediate Shapley value estimates and standard deviations (only when <code>iterative = TRUE</code>),
and the final estimates.
<code>"vS_details"</code> displays information about the v(S) estimates,
most relevant for <code>approach %in% c("regression_separate", "regression_surrogate", "vaeac")</code>.
<code>NULL</code> means no printout.
Any combination can be used, e.g., <code>verbose = c("basic", "vS_details")</code>.</dd>

</dl>

Print Startup Messages with CLI

cli_startup: Print Startup Messages with CLI

Description

Usage

Value

Arguments