Scikit-Learn RandomizedSearchCV RANSACRegressor

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 RANSACRegressor model, which is robust to outliers and suitable 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.

RANSACRegressor (RANdom SAmple Consensus) is a robust regression model that iteratively fits the model on random subsets of the data, effectively handling outliers. The model identifies inliers and outliers by fitting the model to random samples and selecting the subset that yields the best fit.

Key hyperparameters for RANSACRegressor include min_samples, which determines the minimum number of samples required to fit the model, and max_trials, the maximum number of iterations the algorithm will perform to find a valid consensus set.

from sklearn.datasets import make_regression
from sklearn.model_selection import train_test_split, RandomizedSearchCV
from sklearn.linear_model import RANSACRegressor
from scipy.stats import randint

# Generate synthetic regression dataset with outliers
X, y = make_regression(n_samples=100, n_features=10, noise=0.1, random_state=42)
# Introduce outliers
X[:50] += 3
y[:50] += 50

# 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 = RANSACRegressor(random_state=42)

# Define hyperparameter distribution
param_dist = {
    'min_samples': randint(50, 100),
    'max_trials': randint(10, 30)
}

# 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_
test_score = best_model.score(X_test, y_test)
print(f"Test set R^2 score: {test_score:.3f}")

Running the example gives an output like:

Best score: -26349.088
Best parameters: {'max_trials': 17, 'min_samples': 53}
Test set R^2 score: 0.593

The steps are as follows:

1. Generate a synthetic regression dataset with outliers using scikit-learn’s make_regression function and introduce outliers.
2. Split the dataset into train and test sets using train_test_split.
3. Define the model and specify the hyperparameter distribution for min_samples and max_trials.
4. Perform random search using RandomizedSearchCV, specifying the RANSACRegressor model, hyperparameter distribution, 100 iterations, 5-fold cross-validation, and the negative mean squared error scoring metric.
5. Report the best cross-validation score and the best set of hyperparameters found by random search.
6. Evaluate the best model on the hold-out test set and report the R^2 score.

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 RANSACRegressor model.