CODEWITHSUNDEEP
parametersscikit-learngrid-search
kilojoules
kilojoules

Reputation: 10083

Understanding sklearn grid_search

I am struggling to understand how the grid_search class works. I want to find the best max_depth parameter which I can use with the RandomForestClassifier. I specified the possible options I wanted the search to run through, and I expected the module to output the "best fitting" max_depth option. 

from sklearn.datasets import load_iris
from sklearn.ensemble import RandomForestClassifier
from sklearn import grid_search

iris= load_iris()    
forest_parameters = {'max_depth': [1,2,3,4]}
forest = RandomForestClassifier()
explorer = grid_search.GridSearchCV(forest, forest_parameters)
explorer.fit(iris['data'], iris['target'])

I expected my explorer grid search module to return with the best max_depth parameter, given the set of possible options [1,2,3,4]. Why is the default value of None still being used? How can I use grid_search to find "best-fitting" parameters?

Out[13]: 
GridSearchCV(cv=None, error_score='raise',
       estimator=RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',
       ---> max_depth=None, max_features='auto', max_leaf_nodes=None,
            min_samples_leaf=1, min_samples_split=2,
            min_weight_fraction_leaf=0.0, n_estimators=10, n_jobs=1,
            oob_score=False, random_state=None, verbose=0,
            warm_start=False),
       fit_params={}, iid=True, loss_func=None, n_jobs=1,
       param_grid={'max_depth': [1, 2, 3, 4]}, pre_dispatch='2*n_jobs',
       refit=True, score_func=None, scoring=None, verbose=0)

Upvotes: 1

Views: 863

Answers (1)

David Maust
David Maust

Reputation: 8270

Those are only the parameters the grid search was invoked with. To identify to the best parameters use explorer.best_params_, or you could find the estimator using explorer.best_estimator_, provided refit is enabled.

Upvotes: 2

Related Questions