sage_batch_predict

Performs batched prediction on combined data to manage memory usage.
Supports both classification (probability predictions) and regression.

internal

Provides a consistent interface for common feature importance
methods as described in Ewald et al. (2024)
<doi:10.1007/978-3-031-63797-1_22>, including permutation feature
importance (PFI), conditional and relative feature importance
(CFI, RFI), leave one covariate out (LOCO), and Shapley
additive global importance (SAGE), as well as feature sampling
mechanisms to support conditional importance methods.

Lukas Burk

xplainfi

Feature Importance Methods for Global Explanations

sage_batch_predict function

<dl><dt>learner</dt>
<dd>(<code>Learner</code>) Trained mlr3 learner.</dd>
<dt>combined_data</dt>
<dd>(<code>data.table</code>) Data with feature columns to predict on.</dd>
<dt>task</dt>
<dd>(<code>Task</code>) mlr3 task object.</dd>
<dt>batch_size</dt>
<dd>(<code>integer(1)</code> or <code>NULL</code>) Batch size for predictions. If <code>NULL</code> or if
<code>total_rows &lt;= batch_size</code>, processes all data at once.</dd>
<dt>task_type</dt>
<dd>(<code>character(1)</code>) Task type, either <code>"classif"</code> or <code>"regr"</code>.</dd></dl>

Arguments

Batch Predict for SAGE — sage_batch_predict

<dl>

<dt>learner</dt>
<dd>(<code>Learner</code>) Trained mlr3 learner.</dd>


<dt>combined_data</dt>
<dd>(<code>data.table</code>) Data with feature columns to predict on.</dd>


<dt>task</dt>
<dd>(<code>Task</code>) mlr3 task object.</dd>


<dt>batch_size</dt>
<dd>(<code>integer(1)</code> or <code>NULL</code>) Batch size for predictions. If <code>NULL</code> or if
<code>total_rows &lt;= batch_size</code>, processes all data at once.</dd>


<dt>task_type</dt>
<dd>(<code>character(1)</code>) Task type, either <code>"classif"</code> or <code>"regr"</code>.</dd>

</dl>

sage_batch_predict: Batch Predict for SAGE

Description

Usage

Value

Arguments