Data Science · Chapter 32 of 43

Random Forest

A RANDOM FOREST averages many decision trees trained on random subsets of the data and features.

Strong out-of-the-box performance on tabular data — a common first serious model.

Example 1 (python)

from sklearn.ensemble import RandomForestClassifier
m = RandomForestClassifier(n_estimators=200, random_state=42).fit(X_train, y_train)
print(m.score(X_test, y_test))

Output

0.91

Usually beats a single tree.

Example 2 (python)

print(m.feature_importances_[:5])

Output

[0.12 0.08 0.35 0.02 0.05]

Built-in feature importance.

Key points

Ensemble of many trees.
Reduces variance vs single tree.
Great on tabular data.
Provides feature importances.

💡 Note: Random forests handle mixed data types and missing-ish patterns with less pre-processing than most models.

Random Forest

Key points

📝 Quick Quiz