Scikit-Learn GridSearchCV ComplementNB

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 ComplementNB, a variant of the Naive Bayes classifier particularly suited for imbalanced datasets.

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.

ComplementNB is a modification of the Multinomial Naive Bayes algorithm designed to handle imbalanced data more effectively. It is particularly useful for text classification tasks where the classes are not evenly distributed.

The key hyperparameters for ComplementNB include the smoothing parameter alpha, which helps to avoid zero probabilities by adding a constant value to the counts, and the norm parameter, which is a boolean flag indicating whether to normalize the weights of samples.

from sklearn.datasets import make_classification
from sklearn.model_selection import train_test_split, GridSearchCV
from sklearn.naive_bayes import ComplementNB
import numpy as np

# Generate synthetic imbalanced classification dataset
X, y = make_classification(n_samples=1000, n_features=20, n_informative=15, n_redundant=5, weights=[0.9, 0.1], random_state=42)
X = np.abs(X)

# 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 = {
    'alpha': [0.1, 0.5, 1.0, 2.0],
    'norm': [True, False]
}

# Perform grid search
grid_search = GridSearchCV(estimator=ComplementNB(),
                           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: 0.635
Best parameters: {'alpha': 2.0, 'norm': True}
Test set accuracy: 0.630

The steps are as follows:

1. Generate a synthetic imbalanced classification dataset using scikit-learn’s make_classification function.
2. Split the dataset into train and test sets using train_test_split.
3. Define the parameter grid with different values for alpha and norm hyperparameters.
4. Perform grid search using GridSearchCV, specifying the ComplementNB model, parameter grid, 5-fold cross-validation, and accuracy 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 accuracy.

By using GridSearchCV, we can efficiently find the optimal hyperparameters for ComplementNB, enhancing its performance on imbalanced classification tasks. This automated approach ensures we select the best configuration, saving time and improving model effectiveness.