Tutorial 24: Randomized Search

Level: Advanced · Part of: ML Learning Roadmap

← Back to roadmap

Case Study: Tuning Random Forest at Scale

Scenario

Random forests have many hyperparameters. Exhaustive grid search is too slow — sample random combinations instead.

Learning Objectives

• Use RandomizedSearchCV with distributions
• Compare wall-clock time vs GridSearchCV
• Use scipy.stats for continuous distributions
• Analyze cv_results_ DataFrame

Prerequisites

• Tutorial 23
• Python 3.9+, scikit-learn, NumPy, pandas, matplotlib

Dataset

load_wine()

Hands-On Solution

Copy and run the complete script below:

import time
import pandas as pd
from scipy.stats import randint, uniform
from sklearn.datasets import load_wine
from sklearn.model_selection import train_test_split, RandomizedSearchCV
from sklearn.ensemble import RandomForestClassifier

X, y = load_wine(return_X_y=True)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

param_dist = {
    "n_estimators": randint(50, 500),
    "max_depth": randint(2, 20),
    "min_samples_split": randint(2, 20),
    "max_features": uniform(0.1, 0.9),
}

start = time.time()
search = RandomizedSearchCV(
    RandomForestClassifier(random_state=42),
    param_dist, n_iter=30, cv=5, scoring="accuracy",
    random_state=42, n_jobs=-1,
)
search.fit(X_train, y_train)
elapsed = time.time() - start

print(f"Search time: {elapsed:.1f}s")
print(f"Best params: {search.best_params_}")
print(f"Test accuracy: {search.score(X_test, y_test):.4f}")

results = pd.DataFrame(search.cv_results_)
print(results[["param_n_estimators", "param_max_depth", "mean_test_score"]].head())

Expected Output

When you run the script, you should see evaluation metrics printed to the console. Some tutorials also save .png plot files in the current directory.

Exercises

1. Increase n_iter to 100. Does test accuracy improve?
2. Plot n_estimators vs mean_test_score as scatter plot.
3. When is RandomizedSearchCV preferred over GridSearchCV?

Key Takeaways

• Random search often finds good params faster than exhaustive grid
• Use distributions for continuous params, lists for discrete
• Always set random_state for reproducibility

Navigation

Part of the ML Learning Roadmap — Hands-On with scikit-learn