Sample Spaceship Titanic 🚀 scikit-learn logistic regression pipeline with GridSearch hyperparameter tuning ⚙️

logregPipeline.py

from sklearn.preprocessing import PolynomialFeatures, StandardScaler
from sklearn.feature_selection import SelectPercentile, f_classif
from sklearn.pipeline import make_pipeline
from sklearn.linear_model import LogisticRegression

pipe_lgrg = make_pipeline(
    PolynomialFeatures(include_bias=False),
    StandardScaler(),
    SelectPercentile(score_func=f_classif),
    LogisticRegression(max_iter=2000, random_state=22)
)

param_grid_lgrg = {
    'polynomialfeatures__degree': [2],
    'logisticregression__C': [0.1],
    'selectpercentile__percentile': [45]
}

grid_search_lgrg = GridSearchCV(estimator=pipe_lgrg, param_grid=param_grid_lgrg, cv=5, scoring='accuracy', n_jobs=-1)

grid_search_lgrg.fit(X_train, y_train)

print("Best Params:", grid_search_lgrg.best_params_)

accuracy_train = grid_search_lgrg.best_estimator_.score(X_train, y_train)
print("Accuracy on train set:", accuracy_train)
accuracy_test = grid_search_lgrg.best_estimator_.score(X_test, y_test)
print("Accuracy on test set:", accuracy_test)