Python for Finance Cookbook

How to do it...

Execute the following steps to run both grid search and randomized search on the decision tree pipeline we have created in the Implementing scikit-learn's pipelines recipe.

1. Import the libraries:

from sklearn.model_selection import (GridSearchCV, cross_val_score,                                      RandomizedSearchCV,                                      cross_validate,                                      StratifiedKFold)

2. Define a cross-validation scheme:

k_fold = StratifiedKFold(5, shuffle=True, random_state=42)

3. Evaluate the Pipeline using cross-validation:

cross_val_score(tree_pipeline, X_train, y_train, cv=k_fold)

This results in an array containing the estimator's default score (accuracy) values:

array([0.73193085, 0.735625, 0.71458333, 0.72395833, 0.72619296])

4. Add extra metrics to the cross-validation:

cross_validate(tree_pipeline, ...