ePerformance {explainer}

Enhanced Performance Evaluation

Description

This function generates Precision-Recall and ROC curves, including threshold information for binary classification models.

Usage

ePerformance(task, trained_model, splits)

Arguments

task

mlr3 binary classification task object specifying the task details

trained_model

mlr3 trained learner (model) object obtained after training

splits

mlr3 object defining data splits for train and test sets

Value

ROC and Precision-Recall curves with threshold information

Examples

# Set environment variables for reproducibility
Sys.setenv(LANG = "en") # Change R language to English!
RNGkind("L'Ecuyer-CMRG") # Change to L'Ecuyer-CMRG instead of the default "Mersenne-Twister"

# Load required libraries
library("explainer")

# Set seed for reproducibility
seed <- 246
set.seed(seed)

# Load necessary packages
if (!requireNamespace("mlbench", quietly = TRUE)) stop("mlbench not installed.")
if (!requireNamespace("mlr3learners", quietly = TRUE)) stop("mlr3learners not installed.")
if (!requireNamespace("ranger", quietly = TRUE)) stop("ranger not installed.")
# Load BreastCancer dataset
utils::data("BreastCancer", package = "mlbench")

# Keep the target column as "Class"
target_col <- "Class"

# Change the positive class to "malignant"
positive_class <- "malignant"

# Keep only the predictor variables and outcome
mydata <- BreastCancer[, -1] # 1 is ID

# Remove rows with missing values
mydata <- na.omit(mydata)

# Create a vector of sex categories
sex <- sample(c("Male", "Female"), size = nrow(mydata), replace = TRUE)

# Create a vector of age categories
mydata$age <- as.numeric(sample(seq(18, 60), size = nrow(mydata), replace = TRUE))

# Add a sex column to the mydata data frame (for fairness analysis)
mydata$sex <- factor(sex, levels = c("Male", "Female"), labels = c(1, 0))

# Create a classification task
maintask <- mlr3::TaskClassif$new(
  id = "my_classification_task",
  backend = mydata,
  target = target_col,
  positive = positive_class
)

# Create a train-test split
set.seed(seed)
splits <- mlr3::partition(maintask)

# Add a learner (machine learning model base)
# Here we use random forest for example (you can use any other available model)
mylrn <- mlr3::lrn("classif.ranger", predict_type = "prob")

# Train the model
mylrn$train(maintask, splits$train)

# Make predictions on new data
mylrn$predict(maintask, splits$test)
ePerformance(task = maintask, trained_model = mylrn, splits = splits)