explain: Local Interpretable Model-agnostic Explanations.

Description Usage Arguments Value

View source: R/explain.R

Description

Local Interpretable Model-agnostic Explanations.

Generate an explanation for a classifier's prediction.

Usage

1
2
3
explain(classifier, onerow, variable_summaries, num_features = 3,
  num_samples = 100, explainer = MASS::lm.ridge,
  feature_selection = "auto", ...)

Arguments

classifier

tundraContainer. Can be either a tundraContainer or a list with a predict method attached.

onerow

data.frame. A one-row dataframe around which we want to get a local approximation.

variable_summaries

list. A named list that defines means and standard deviations for the variables that need to be perturbed.

Value

an explanation object.


kirillseva/explain documentation built on May 20, 2019, 10:23 a.m.