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 ComplementNB
model, which is particularly useful for text classification tasks with imbalanced datasets.
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.
ComplementNB
is a variant of the Naive Bayes classifier designed to handle imbalanced datasets effectively. It computes model parameters using the complement of each class, providing better performance on imbalanced data.
Key hyperparameters for ComplementNB
include the smoothing parameter alpha
, which adjusts the smoothing applied to the model to handle zero probabilities, and norm
, which determines whether to apply normalization to the data.
from sklearn.datasets import make_classification
from sklearn.model_selection import train_test_split, RandomizedSearchCV
from sklearn.naive_bayes import ComplementNB
from scipy.stats import uniform
import numpy as np
# Generate synthetic binary classification dataset
X, y = make_classification(n_samples=1000, n_features=20, n_informative=10, n_redundant=10, 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 the model
model = ComplementNB()
# Define hyperparameter distribution
param_dist = {
'alpha': uniform(loc=0, scale=1),
'norm': [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: 0.762
Best parameters: {'alpha': 0.1834347898661638, 'norm': False}
Test set accuracy: 0.760
The steps are as follows:
- Generate a synthetic binary classification dataset using scikit-learn’s
make_classification
function. - Split the dataset into train and test sets using
train_test_split
. - Define the
ComplementNB
model. - Specify the hyperparameter distributions for
alpha
(usinguniform
) andnorm
. - Use
RandomizedSearchCV
to perform random search with 100 iterations, 5-fold cross-validation, and accuracy as the scoring metric. - Report the best cross-validation score and best set of hyperparameters found by random search.
- 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 ComplementNB
model.