loss_adapt_mlr3proba

This function allows for usage of standardized measures from the mlr3proba package with <code>survex</code>.

Survival analysis models are commonly used in medicine and other areas. Many of them
are too complex to be interpreted by human. Exploration and explanation is needed, but
standard methods do not give a broad enough picture. 'survex' provides easy-to-apply
methods for explaining survival models, both complex black-boxes and simpler statistical models.
They include methods specific to survival analysis such as SurvSHAP(t) introduced in Krzyzinski et al., (2023)
<doi:10.1016/j.knosys.2022.110234>, SurvLIME described in Kovalev et al., (2020) <doi:10.1016/j.knosys.2020.106164> as well as
extensions of existing ones described in Biecek et al., (2021) <doi:10.1201/9780429027192>.

Mikołaj Spytek

survex

Explainable Machine Learning in Survival Analysis

Mateusz Krzyziński

Sophie Langbein

Hubert Baniecki

Lorenz A. Kapsner

Przemyslaw Biecek

loss_adapt_mlr3proba function

<dl><dt>measure</dt>
<dd><ul>
<li>a <code>MeasureSurv</code> object from the <code>mlr3proba</code> package, the object to adapt</li>
</ul></dd>
<dt>reverse</dt>
<dd><ul>
<li>boolean, FALSE by default, whether the metric should be reversed in order to be treated as loss (for permutational variable importance we need functions with lower values indicating better performance). If TRUE, the new metric value will be (1 - metric_value)</li>
</ul></dd>
<dt>...</dt>
<dd><ul>
<li>other parameters, currently ignored</li>
</ul></dd></dl>

Arguments

Adapt mlr3proba measures for use with survex — loss_adapt_mlr3proba

<dl>

<dt>measure</dt>
<dd><ul>
<li>a <code>MeasureSurv</code> object from the <code>mlr3proba</code> package, the object to adapt</li>
</ul></dd>


<dt>reverse</dt>
<dd><ul>
<li>boolean, FALSE by default, whether the metric should be reversed in order to be treated as loss (for permutational variable importance we need functions with lower values indicating better performance). If TRUE, the new metric value will be (1 - metric_value)</li>
</ul></dd>


<dt>...</dt>
<dd><ul>
<li>other parameters, currently ignored</li>
</ul></dd>

</dl>

loss_adapt_mlr3proba: Adapt mlr3proba measures for use with survex

Description

Usage

Value

Arguments

Examples