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 Bernoulli Naive Bayes (BernoulliNB), a variation of the Naive Bayes classifier for binary data.
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.
BernoulliNB is a variation of the Naive Bayes classifier for binary/boolean features. It is particularly useful for text classification tasks where the input features are word occurrences or frequencies.
The key hyperparameters for BernoulliNB include alpha
(smoothing parameter), which controls the level of smoothing applied to prevent zero probabilities; binarize
(threshold for binarizing the features), which determines the threshold at which the features are considered as 1 or 0; and fit_prior
(whether to learn class priors), which decides if the model should learn class prior probabilities or not.
from sklearn.datasets import make_classification
from sklearn.model_selection import train_test_split, GridSearchCV
from sklearn.naive_bayes import BernoulliNB
# Generate synthetic binary classification dataset
X, y = make_classification(n_samples=1000, n_features=20, n_informative=10, n_redundant=10, random_state=42)
# Binarize the dataset to create binary features
X = (X > 0).astype(int)
# 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],
'binarize': [0.0, 0.5, 1.0],
'fit_prior': [True, False]
}
# Perform grid search
grid_search = GridSearchCV(estimator=BernoulliNB(),
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: 0.750
Best parameters: {'alpha': 0.5, 'binarize': 0.0, 'fit_prior': True}
Test set accuracy: 0.790
The steps are as follows:
- Generate a synthetic binary classification dataset using
make_classification
. - Binarize the dataset to create binary features.
- Split the dataset into train and test sets using
train_test_split
. - Define the parameter grid with different values for
alpha
,binarize
, andfit_prior
hyperparameters. - Perform grid search using
GridSearchCV
, specifying theBernoulliNB
model, parameter grid, 5-fold cross-validation, and accuracy scoring metric. - Report the best cross-validation score and best set of hyperparameters found by grid search.
- Evaluate the best model on the hold-out test set and report the accuracy.
By using GridSearchCV
, we can easily 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 BernoulliNB model.