Scikit-Learn RandomizedSearchCV SVR

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 support vector regression (SVR) 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.

Support vector regression (SVR) is a regression algorithm that fits a model to predict continuous outcomes using support vectors. It aims to minimize prediction error within a margin of tolerance.

Key hyperparameters for SVR include the regularization parameter C, which controls model complexity and helps prevent overfitting; the kernel type (linear, poly, rbf, sigmoid), which determines the function used to map data into higher-dimensional space; and the epsilon parameter epsilon, which defines the margin of tolerance within which no penalty is given for errors.

from sklearn.datasets import make_regression
from sklearn.model_selection import train_test_split, RandomizedSearchCV
from sklearn.svm import SVR
from scipy.stats import uniform

# Generate synthetic regression dataset
X, y = make_regression(n_samples=100, n_features=10, 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 = SVR()

# Define hyperparameter distribution
param_dist = {
    'C': uniform(loc=0.1, scale=9.9),
    'kernel': ['linear', 'poly', 'rbf', 'sigmoid'],
    'epsilon': uniform(loc=0.01, scale=0.09)
}

# 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: -0.018
Best parameters: {'C': 8.103133746352965, 'epsilon': 0.03741523922560336, 'kernel': 'linear'}
Test set mean squared error: 1.000

The steps are as follows:

1. Generate a synthetic regression dataset using scikit-learn’s make_regression function.
2. Split the dataset into training and testing sets using train_test_split.
3. Define the SVR model and the hyperparameter distribution with different values for C, kernel, and epsilon.
4. Perform random search using RandomizedSearchCV, specifying the SVR model, hyperparameter distribution, 100 iterations, 5-fold cross-validation, and negative mean squared error as the scoring metric.
5. Report the best cross-validation score and best set of hyperparameters found by random search.
6. 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 SVR model.