Scikit-Learn RandomizedSearchCV HistGradientBoostingRegressor

Hyperparameter tuning is essential for optimizing machine learning models. In this example, we’ll demonstrate how to use scikit-learn’s RandomizedSearchCV for hyperparameter tuning of a HistGradientBoostingRegressor model, commonly used for regression tasks.

Random search is a method for evaluating different combinations of model hyperparameters. Unlike grid search, it samples a fixed number of hyperparameter combinations from a specified distribution, making it more efficient when searching over a large hyperparameter space.

HistGradientBoostingRegressor is a type of gradient boosting algorithm designed for regression tasks. It builds an ensemble of decision trees in a sequential manner to minimize the loss function. This implementation is highly efficient, especially for large datasets.

Key hyperparameters for HistGradientBoostingRegressor include the learning_rate, which controls the contribution of each tree to the final model; max_iter, which specifies the number of boosting iterations; max_leaf_nodes, which sets the maximum number of leaves in each tree; and min_samples_leaf, which determines the minimum number of samples required to form a leaf.

from sklearn.datasets import make_regression
from sklearn.model_selection import train_test_split, RandomizedSearchCV
from sklearn.ensemble import HistGradientBoostingRegressor
from scipy.stats import uniform, randint

# Generate synthetic regression dataset
X, y = make_regression(n_samples=100, n_features=20, noise=0.1, random_state=42)

# Split into train and test sets
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# Define the model
model = HistGradientBoostingRegressor(random_state=42)

# Define hyperparameter distribution
param_dist = {
    'learning_rate': uniform(0.01, 0.3),
    'max_iter': randint(10, 50),
    'max_leaf_nodes': randint(10, 50),
    'min_samples_leaf': randint(1, 20)
}

# Perform random search
random_search = RandomizedSearchCV(estimator=model,
                                   param_distributions=param_dist,
                                   n_iter=50,
                                   cv=5,
                                   scoring='neg_mean_squared_error',
                                   random_state=42)
random_search.fit(X_train, y_train)

# Report best score and parameters
print(f"Best score: {random_search.best_score_:.3f}")
print(f"Best parameters: {random_search.best_params_}")

# Evaluate on test set
best_model = random_search.best_estimator_
mse = best_model.score(X_test, y_test)
print(f"Test set mean squared error: {-mse:.3f}")

Running the example gives an output like:

Best score: -8098.917
Best parameters: {'learning_rate': 0.268219174976903, 'max_iter': 48, 'max_leaf_nodes': 15, 'min_samples_leaf': 8}
Test set mean squared error: -0.440

The steps are as follows:

1. Generate a synthetic regression dataset using scikit-learn’s make_regression function.
2. Split the dataset into train and test sets using train_test_split.
3. Define the HistGradientBoostingRegressor model.
4. Define the hyperparameter distribution, specifying ranges for learning_rate, max_iter, max_leaf_nodes, and min_samples_leaf.
5. Perform random search using RandomizedSearchCV, specifying the HistGradientBoostingRegressor model, hyperparameter distribution, 100 iterations, 5-fold cross-validation, and mean squared error scoring metric.
6. Report the best cross-validation score and best set of hyperparameters found by the random search.
7. Evaluate the best model on the hold-out test set and report the mean squared error.

By using RandomizedSearchCV, we can efficiently explore different hyperparameter settings and find the combination that maximizes the model’s performance. This automated approach saves time and effort compared to manual hyperparameter tuning and helps ensure we select the best configuration for our HistGradientBoostingRegressor model.