gain_lift: Cumulative Gain, Lift and Response

In lares: Lean Analytics and Robust Exploration Sidekick

Cumulative Gain, Lift and Response

Description

This function calculates cumulative gain, lift, and response values for a predictive score of a specific target. You can use the mplot_gain() function to create a plot.

Usage

gain_lift(
  tag,
  score,
  target = "auto",
  splits = 10,
  plot = FALSE,
  quiet = FALSE
)

Arguments

tag	Vector. Real known label
score	Vector. Predicted value or model's result
target	Value. Which is your target positive value? If set to 'auto', the target with largest mean(score) will be selected. Change the value to overwrite. Only used when binary categorical model.
splits	Integer. Number of percentiles to split the data
plot	Boolean. Plot results? Uses mplot_gain()
quiet	Boolean. Keep quiet? If not, informative messages will be shown.

Value

data.frame when plot=FALSE or plot when plot=TRUE.

See Also

Other Machine Learning: ROC(), conf_mat(), export_results(), h2o_automl(), h2o_predict_MOJO(), h2o_selectmodel(), impute(), iter_seeds(), lasso_vars(), model_metrics(), model_preprocess(), msplit()

Other Model metrics: ROC(), conf_mat(), errors(), loglossBinary(), model_metrics()

Examples

data(dfr) # Results for AutoML Predictions
head(dfr$class2)

# Results for Binomial Model
gain_lift(dfr$class2$tag, dfr$class2$scores, target = "FALSE")
gain_lift(dfr$class2$tag, dfr$class2$scores, target = "TRUE", splits = 5)

lares documentation built on Aug. 18, 2025, 5:12 p.m.