localSurrogate

Plots and returns a Rforestry object with a single tree explaining
 the PDP surface.

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

localSurrogate function

<dl><dt>object</dt>
<dd>Interpreter object to make plots + surrogate for.</dd>
<dt>features.2d</dt>
<dd>A two-column dataframe of pairs of features to make local surrogates for.
Each row represents a pair of features, with the names of features as the
entries.</dd>
<dt>interact</dt>
<dd>An indicator specifying if the surrogate model should also be
given the interaction terms to create the surrogate models with.
Default is FALSE.</dd>
<dt>params.forestry</dt>
<dd>Optional list of parameters to pass to the surrogate
model. Defaults to the standard Rforestry parameters with ntree = 1
and maxDepth = 2.</dd></dl>

Arguments

Given a interpreter object with at least one pair of features,
  create a small surrogate model for the two features using the PDP function
  as the output and the two features as the independent variables. — localSurrogate

<dl>

<dt>object</dt>
<dd>Interpreter object to make plots + surrogate for.</dd>


<dt>features.2d</dt>
<dd>A two-column dataframe of pairs of features to make local surrogates for.
Each row represents a pair of features, with the names of features as the
entries.</dd>


<dt>interact</dt>
<dd>An indicator specifying if the surrogate model should also be
given the interaction terms to create the surrogate models with.
Default is FALSE.</dd>


<dt>params.forestry</dt>
<dd>Optional list of parameters to pass to the surrogate
model. Defaults to the standard Rforestry parameters with ntree = 1
and maxDepth = 2.</dd>

</dl>

Given a interpreter object with at least one pair of features,
  create a small surrogate model for the two features using the PDP function
  as the output and the two features as the independent variables.

localSurrogate: Given a interpreter object with at least one pair of features, create a small surrogate model for the two features using the PDP function as the output and the two features as the independent variables.

Description

Usage

Value

Arguments