mplot_roc: ROC Curve Plot

In laresbernardo/lares: Analytics & Machine Learning Sidekick

ROC Curve Plot

Description

This function plots ROC Curves with AUC values with 95% confidence range. It also works for multi-categorical models.

Usage

mplot_roc(
  tag,
  score,
  multis = NA,
  sample = 1000,
  model_name = NA,
  subtitle = NA,
  interval = 0.2,
  squared = TRUE,
  plotly = FALSE,
  save = FALSE,
  subdir = NA,
  file_name = "viz_roc.png"
)

Arguments

tag	Vector. Real known label.
score	Vector. Predicted value or model's result.
multis	Data.frame. Containing columns with each category probability or score (only used when more than 2 categories coexist).
sample	Integer. Number of samples to use for rendering plot.
model_name	Character. Model's name
subtitle	Character. Subtitle to show in plot
interval	Numeric. Interval for breaks in plot
squared	Boolean. Keep proportions?
plotly	Boolean. Use plotly for plot's output for an interactive plot
save	Boolean. Save output plot into working directory
subdir	Character. Sub directory on which you wish to save the plot
file_name	Character. File name as you wish to save the plot

Value

Plot with ROC curve and AUC performance results.

See Also

Other ML Visualization: mplot_conf(), mplot_cuts(), mplot_cuts_error(), mplot_density(), mplot_full(), mplot_gain(), mplot_importance(), mplot_lineal(), mplot_metrics(), mplot_response(), mplot_splits(), mplot_topcats()

Examples

Sys.unsetenv("LARES_FONT") # Temporal
data(dfr) # Results for AutoML Predictions
lapply(dfr[c(1, 2)], head)

# ROC Curve for Binomial Model
mplot_roc(dfr$class2$tag, dfr$class2$scores,
  model_name = "Titanic Survived Model"
)

# ROC Curves for Multi-Categorical Model
mplot_roc(dfr$class3$tag, dfr$class3$score,
  multis = subset(dfr$class3, select = -c(tag, score)),
  squared = FALSE,
  model_name = "Titanic Class Model"
)

laresbernardo/lares documentation built on Oct. 23, 2024, 12:05 p.m.