predict-Surrogate

predict.Surrogate

Provides several methods for model distillation and interpretability
for general black box machine learning models and treatment effect estimation
methods. For details on the algorithms implemented, see <https://forestry-labs.github.io/distillML/index.html>
Brian Cho, Theo F. Saarinen, Jasjeet S. Sekhon, Simon Walter.

Theo Saarinen

distillML

Model Distillation and Interpretability Methods for Machine
Learning Models

Brian Cho

Jasjeet Sekhon

Simon Walter

predict-Surrogate function

<dl><dt>object</dt>
<dd>A surrogate object distilled from the interpreter</dd>
<dt>newdata</dt>
<dd>The dataframe to use for the predictions</dd>
<dt>...</dt>
<dd>Additional parameters to pass to predict</dd></dl>

Arguments

Prediction method for the distilled surrogate model — predict-Surrogate

<dl>

<dt>object</dt>
<dd>A surrogate object distilled from the interpreter</dd>


<dt>newdata</dt>
<dd>The dataframe to use for the predictions</dd>


<dt>...</dt>
<dd>Additional parameters to pass to predict</dd>

</dl>

Prediction method for the distilled surrogate model

predict-Surrogate: Prediction method for the distilled surrogate model

Description

Usage

Value

Arguments