familiarMetric-class

Superclass for model performance objects.

Single unified interface for end-to-end modelling of regression,
categorical and time-to-event (survival) outcomes. Models created using
familiar are self-containing, and their use does not require additional
information such as baseline survival, feature clustering, or feature
transformation and normalisation parameters. Model performance,
calibration, risk group stratification, (permutation) variable importance,
individual conditional expectation, partial dependence, and more, are
assessed automatically as part of the evaluation process and exported in
tabular format and plotted, and may also be computed manually using export
and plot functions. Where possible, metrics and values obtained during the
evaluation process come with confidence intervals.

Alex Zwanenburg

familiar

End-to-End Automated Machine Learning and Model Evaluation

Steffen Löck

Stefan Leger

Iram Shahzadi

Asier Rabasco Meneghetti

Sebastian Starke

Technische Universität Dresden 

German Cancer Research Center (DKFZ) 

familiarMetric-class function

<dl>
<dt><code>metric</code></dt>
<dd>Performance metric.</dd></dl>
<dt><code>outcome_type</code></dt>
<dd>Type of outcome being predicted.</dd>
<dt><code>name</code></dt>
<dd>Name of the performance metric.</dd>
<dt><code>value_range</code></dt>
<dd>Range of the performance metric. Can be half-open.</dd>
<dt><code>baseline_value</code></dt>
<dd>Value of the metric for trivial models, e.g. models that
always predict the median value, the majority class, or the mean hazard,
etc.</dd>
<dt><code>higher_better</code></dt>
<dd>States whether higher metric values correspond to better
predictive model performance (e.g. accuracy) or not (e.g. root mean squared
error).</dd>

Slots

Model performance metric. — familiarMetric-class

<dl>
<dt><code>metric</code></dt>
<dd>Performance metric.</dd>


<dt><code>outcome_type</code></dt>
<dd>Type of outcome being predicted.</dd>


<dt><code>name</code></dt>
<dd>Name of the performance metric.</dd>


<dt><code>value_range</code></dt>
<dd>Range of the performance metric. Can be half-open.</dd>


<dt><code>baseline_value</code></dt>
<dd>Value of the metric for trivial models, e.g. models that
always predict the median value, the majority class, or the mean hazard,
etc.</dd>


<dt><code>higher_better</code></dt>
<dd>States whether higher metric values correspond to better
predictive model performance (e.g. accuracy) or not (e.g. root mean squared
error).</dd>


</dl>

familiarMetric-class: Model performance metric.

Description

Arguments

Slots