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 LabelPropagation, a semi-supervised learning 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.
LabelPropagation is a semi-supervised learning algorithm that propagates labels through the dataset based on a similarity graph. It leverages both labeled and unlabeled data to improve classification accuracy.
The key hyperparameters for LabelPropagation include the kernel coefficient gamma
, which influences the shape of the decision boundary; max_iter
, which sets the maximum number of iterations for the algorithm; and kernel
, which specifies the type of kernel used (e.g., ‘knn’ or ‘rbf’).
from sklearn.datasets import make_classification
from sklearn.model_selection import train_test_split, GridSearchCV
from sklearn.semi_supervised import LabelPropagation
# Generate synthetic dataset with partially labeled data
X, y = make_classification(n_samples=1000, n_features=10, n_informative=5, n_redundant=5, random_state=42)
y[::10] = -1 # Set 10% of labels to -1 to simulate unlabeled data
# 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 = {
'gamma': [0.1, 1, 10],
'max_iter': [100, 200, 300],
'kernel': ['knn', 'rbf']
}
# Perform grid search
grid_search = GridSearchCV(estimator=LabelPropagation(),
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: 0.848
Best parameters: {'gamma': 1, 'kernel': 'rbf', 'max_iter': 100}
Test set accuracy: 0.860
The steps are as follows:
- Generate a synthetic dataset with partially labeled data using scikit-learn’s
make_classification
function. - Split the dataset into train and test sets using
train_test_split
. - Define the parameter grid with different values for
gamma
,max_iter
, andkernel
hyperparameters. - Perform grid search using
GridSearchCV
, specifying theLabelPropagation
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 efficiently explore different hyperparameter settings and find the combination that maximizes the performance of our LabelPropagation model. This automated approach saves time and effort compared to manual hyperparameter tuning and helps ensure we select the best configuration for our semi-supervised learning model.