Random Forests, Explained

Wisdom of crowds

A single decision tree overfits easily. A random forest trains many trees on random subsets of rows and features, then averages their votes. The errors cancel out, giving a robust model that works well with little tuning.

from sklearn.ensemble import RandomForestClassifier
model = RandomForestClassifier(n_estimators=300, random_state=42)
model.fit(X_train, y_train)
print(sorted(zip(model.feature_importances_, X.columns), reverse=True)[:5])

Why people love it

Minimal preprocessing, handles non-linear relationships, resists overfitting, and reports feature importances. For tabular problems it's an excellent first model — and often the last one you need.