Supervised learning

Train on labelled (X, y) pairs to predict y for new X.

Code example

from sklearn.ensemble import RandomForestClassifier

model = RandomForestClassifier(n_estimators=100)
model.fit(X_train, y_train)        # learn
preds = model.predict(X_test)      # predict
acc  = (preds == y_test).mean()    # evaluate

How it works

Supervised learning is the bread-and-butter of ML: you have examples (X) with labels (y), and the model learns a function from X to y. Classification predicts a discrete label (spam/not-spam); regression predicts a continuous value (house price). OpenBenchML benchmarks supervised models by calling fit() then predict() on a holdout split, then computing accuracy/F1 (classification) or RMSE/MAE (regression).