plot_fairness_heatmap: Plot Heatmap

In ModelOriented/FairModels: Flexible Tool for Bias Detection, Visualization, and Mitigation

Plot Heatmap

Description

Heatmap shows all parity loss metrics across all models while displaying similarity between variables (in form of dendograms). All metrics are visible. Some have identical values as it should be in terms of their parity loss (eg. TPR parity loss == FNR parity loss, because TPR = 1 - FNR ). NA's in metrics are gray.

Usage

## S3 method for class 'fairness_heatmap'
plot(
  x,
  ...,
  midpoint = NULL,
  title = NULL,
  subtitle = NULL,
  text = TRUE,
  text_size = 3,
  flip_axis = FALSE
)

Arguments

x	fairness_heatmap
...	other fairness_heatmap objects
midpoint	numeric, midpoint on gradient scale
title	character, title of the plot
subtitle	character, subtitle of the plot
text	logical, default TRUE means it shows values on tiles
text_size	numeric, size of text
flip_axis	logical, whether to change axis with metrics on axis with models

Value

list of ggplot2 objects

Examples

data("german")

y_numeric <- as.numeric(german$Risk) - 1

lm_model <- glm(Risk ~ .,
  data = german,
  family = binomial(link = "logit")
)

rf_model <- ranger::ranger(Risk ~ .,
  data = german,
  probability = TRUE,
  num.trees = 200,
  num.threads = 1,
  seed = 1
)

explainer_lm <- DALEX::explain(lm_model, data = german[, -1], y = y_numeric)
explainer_rf <- DALEX::explain(rf_model, data = german[, -1], y = y_numeric)

fobject <- fairness_check(explainer_lm, explainer_rf,
  protected = german$Sex,
  privileged = "male"
)

# same explainers with different cutoffs for female
fobject <- fairness_check(explainer_lm, explainer_rf, fobject,
  protected = german$Sex,
  privileged = "male",
  cutoff = list(female = 0.4),
  label = c("lm_2", "rf_2")
)


fh <- fairness_heatmap(fobject)

plot(fh)

ModelOriented/FairModels documentation built on Aug. 30, 2022, 5:48 p.m.