Scikit-Learn GridSearchCV NuSVC

Hyperparameter tuning is essential for 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 NuSVC, a variant of the Support Vector Classification 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.

NuSVC is a version of Support Vector Classification that provides control over the number of support vectors used in the model. It is useful for both binary and multiclass classification tasks.

The key hyperparameters for NuSVC include nu, which is an upper bound on the fraction of margin errors and a lower bound on the fraction of support vectors; kernel, which determines the type of decision boundary; and gamma, which defines the influence of individual data points.

from sklearn.datasets import make_classification
from sklearn.model_selection import train_test_split, GridSearchCV
from sklearn.svm import NuSVC

# Generate synthetic dataset for binary classification
X, y = make_classification(n_samples=1000, n_features=20, n_informative=10, n_redundant=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 = {
    'nu': [0.1, 0.3, 0.5],
    'kernel': ['linear', 'rbf'],
    'gamma': ['scale', 'auto']
}

# Perform grid search
grid_search = GridSearchCV(estimator=NuSVC(random_state=42),
                           param_grid=param_grid,
                           cv=5,
                           scoring='accuracy')
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_
accuracy = best_model.score(X_test, y_test)
print(f"Test set accuracy: {accuracy:.3f}")

Running the example gives an output like:

Best score: 0.940
Best parameters: {'gamma': 'scale', 'kernel': 'rbf', 'nu': 0.3}
Test set accuracy: 0.950

The steps are as follows:

1. Generate a synthetic dataset for binary classification using make_classification.
2. Split the dataset into training and testing sets using train_test_split.
3. Define the parameter grid for nu, kernel, and gamma hyperparameters.
4. Use GridSearchCV to perform the grid search with NuSVC, using 5-fold cross-validation and accuracy as the scoring metric.
5. Report the best cross-validation score and corresponding hyperparameters.
6. Evaluate the best model on the hold-out test set and report the accuracy.

By utilizing GridSearchCV, the optimal hyperparameters for NuSVC can be identified, ensuring improved model performance. This streamlined approach aids in efficiently tuning hyperparameters, saving time and effort.