Module data_split 

Data splitting utilities for machine learning workflows

This module provides tools for splitting datasets into training, validation, and test sets using various strategies:

• train_test_split - Simple random train/test split
• stratified_train_test_split - Stratified split preserving class proportions
• KFold - K-fold cross-validation
• StratifiedKFold - Stratified K-fold cross-validation
• LeaveOneOut - Leave-one-out cross-validation
• TimeSeriesSplit - Time series cross-validation (expanding or sliding window)
• GroupKFold - Group K-fold (keeps groups intact)
• ShuffleSplit - Repeated random train/test splits

Structs

GroupKFold

Group K-fold cross-validation.

KFold

K-fold cross-validation splitter.

LeaveOneOut

Leave-one-out cross-validation.

ShuffleSplit

Repeated random train/test splits.

StratifiedKFold

Stratified K-fold cross-validation.

TimeSeriesSplit

Time series cross-validation splitter.

Enums

TimeSeriesMode

Time series split mode.

Functions

stratified_train_test_split

Stratified train/test split that preserves the proportion of each class.

train_test_split

Split data indices into training and test sets.

Type Aliases

SplitIndices

Indices for a single split (train indices, test indices).