35. Decision Trees and Random Forests in Python
35.4 Copy-Run a Small Random Forest
Why. Same data, stronger committee — RandomForest with a small n_estimators for a laptop-friendly demo.
What. RandomForestClassifier(n_estimators=50) builds 50 trees and majority-votes for the class.
How.
Fit a small forest
- Use the same X_train, y_train.
- Fit RandomForestClassifier.
- Compare predict on X_test with the single tree (spot-check only).
from sklearn.ensemble import RandomForestClassifier
forest = RandomForestClassifier(n_estimators=50, random_state=0)
forest.fit(X_train, y_train)
print("tree preds: ", tree.predict(X_test))
print("forest preds:", forest.predict(X_test))
print("feature_importances_:", forest.feature_importances_.round(3))
What you should see. Predictions from both models; importances that sum to about 1.0 — a rough “which column the forest used more,” not a causal proof.
Fictional data
Same teaching churn table — not a production CRM extract.