Scikit-Learn GridSearchCV RANSACRegressor

Hyperparameter tuning is a crucial step in optimizing machine learning models for best performance. In this example, we’ll demonstrate how to use scikit-learn’s GridSearchCV to perform hyperparameter tuning for RANSACRegressor, a robust regression algorithm.

Grid search is a method for evaluating different combinations of model hyperparameters to find the best performing configuration. It exhaustively searches through a specified parameter grid, trains and evaluates the model for each combination using cross-validation, and selects the hyperparameters that yield the best performance metric.

RANSACRegressor is a robust regression model designed to fit a model to the inliers in the data, effectively ignoring the outliers. This makes it particularly useful for datasets with significant noise or outliers.

The key hyperparameters for RANSACRegressor include the minimum number of samples (min_samples) required to fit the model, the residual threshold (residual_threshold) for classifying a data point as an inlier, and the maximum number of iterations (max_trials) allowed for random sample selection.

from sklearn.datasets import make_regression
from sklearn.model_selection import train_test_split, GridSearchCV
from sklearn.linear_model import RANSACRegressor
from sklearn.metrics import mean_squared_error

# Generate synthetic regression dataset
X, y = make_regression(n_samples=1000, n_features=10, noise=10, 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 parameter grid
param_grid = {
    'min_samples': [0.1, 0.5, 0.9],
    'residual_threshold': [1.0, 5.0, 10.0],
    'max_trials': [100, 500, 1000]
}

# Perform grid search
grid_search = GridSearchCV(estimator=RANSACRegressor(random_state=42),
                           param_grid=param_grid,
                           cv=5,
                           scoring='neg_mean_squared_error')
grid_search.fit(X_train, y_train)

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

# Evaluate on test set
best_model = grid_search.best_estimator_
y_pred = best_model.predict(X_test)
test_mse = mean_squared_error(y_test, y_pred)
print(f"Test set mean squared error: {test_mse:.3f}")

Running the example gives an output like:

Best score: 95.573
Best parameters: {'max_trials': 100, 'min_samples': 0.9, 'residual_threshold': 1.0}
Test set mean squared error: 94.704

The steps are as follows:

1. Generate a synthetic regression dataset using make_regression.
2. Split the dataset into train and test sets using train_test_split.
3. Define the parameter grid with different values for min_samples, residual_threshold, and max_trials hyperparameters.
4. Perform grid search using GridSearchCV, specifying the RANSACRegressor model, parameter grid, 5-fold cross-validation, and negative mean squared error scoring metric.
5. Report the best cross-validation score and best set of hyperparameters found by grid search.
6. Evaluate the best model on the hold-out test set and report the mean squared error.

By using GridSearchCV, we can systematically explore different hyperparameter settings and identify the combination that maximizes the performance of our RANSACRegressor model. This automated approach simplifies the hyperparameter tuning process, ensuring we select the best configuration for robust regression tasks.