Scikit-Learn RandomizedSearchCV CategoricalNB

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 CategoricalNB model, commonly used for classification tasks with categorical features.

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.

CategoricalNB is a variant of the Naive Bayes algorithm designed for categorical data. It estimates the likelihood of each category and combines these probabilities to make a classification decision.

Key hyperparameters for CategoricalNB include the additive smoothing parameter (alpha), which prevents zero probabilities in calculations, and fit_prior, a boolean that determines whether to learn class prior probabilities or not.

from sklearn.model_selection import train_test_split, RandomizedSearchCV
from sklearn.naive_bayes import CategoricalNB
from scipy.stats import uniform
import numpy as np

# Generate synthetic categorical dataset
np.random.seed(42)
X = np.random.randint(0, 5, size=(1000, 10))
y = np.random.randint(0, 2, size=(1000,))

# 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 = CategoricalNB()

# Define hyperparameter distribution
param_dist = {
    'alpha': uniform(loc=0, scale=1),
    'fit_prior': [True, False]
}

# Perform random search
random_search = RandomizedSearchCV(estimator=model,
                                   param_distributions=param_dist,
                                   n_iter=100,
                                   cv=5,
                                   scoring='accuracy',
                                   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_
accuracy = best_model.score(X_test, y_test)
print(f"Test set accuracy: {accuracy:.3f}")

Running the example gives an output like:

Best score: 0.486
Best parameters: {'alpha': 0.3745401188473625, 'fit_prior': True}
Test set accuracy: 0.500

The steps are as follows:

1. Generate a synthetic categorical classification dataset.
2. Split the dataset into train and test sets using train_test_split.
3. Define the CategoricalNB model and the hyperparameter distributions for alpha and fit_prior.
4. Perform random search using RandomizedSearchCV, specifying the CategoricalNB model, hyperparameter distributions, 100 iterations, 5-fold cross-validation, and accuracy 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 accuracy.

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 CategoricalNB model.