Cross-Validation Explained

Why one split isn't enough

A single test split can be lucky or unlucky. k-fold cross-validation splits data into k parts, trains on k−1 and tests on the remaining fold, then rotates — averaging the scores.

from sklearn.model_selection import cross_val_score
scores = cross_val_score(model, X, y, cv=5, scoring="f1")
print(scores.mean(), "+/-", scores.std())

When to use which

• k-fold for most cases (k=5 or 10).
• Stratified k-fold for imbalanced classification (keeps class ratios).
• TimeSeriesSplit for temporal data — never shuffle time.

Report the mean and the spread; a high average with huge variance isn't reliable.