GridSearchCV
E97067
GridSearchCV is a scikit-learn tool that systematically searches over specified hyperparameter values using cross-validation to find the best-performing model configuration.
Statements (51)
| Predicate | Object |
|---|---|
| instanceOf |
hyperparameter optimization tool
ⓘ
model selection utility ⓘ scikit-learn class ⓘ |
| acceptsParameter |
cv
ⓘ
error_score ⓘ estimator ⓘ iid ⓘ n_jobs ⓘ param_grid ⓘ pre_dispatch ⓘ refit ⓘ return_train_score ⓘ scoring ⓘ verbose ⓘ |
| compatibleWith | any scikit-learn estimator with fit method ⓘ |
| definedInModule | sklearn.model_selection ⓘ |
| documentedAt | https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.GridSearchCV.html ⓘ |
| hasAttribute |
best_estimator_
ⓘ
best_index_ ⓘ best_params_ ⓘ best_score_ ⓘ cv_results_ ⓘ n_splits_ ⓘ scorer_ ⓘ |
| hasMethod |
fit
ⓘ
get_params ⓘ predict ⓘ score ⓘ set_params ⓘ |
| inheritsFrom | BaseSearchCV ⓘ |
| introducedInLibrary |
scikit-learn
ⓘ
surface form:
scikit-learn 0.16 or earlier
|
| parallelization | uses joblib for parallel computation ⓘ |
| parameterType |
cv can be cross-validation splitter
ⓘ
cv can be int ⓘ n_jobs can be -1 for using all processors ⓘ param_grid can be dict ⓘ param_grid can be list of dicts ⓘ scoring can be callable ⓘ scoring can be string ⓘ |
| partOf | scikit-learn ⓘ |
| primaryPurpose |
hyperparameter tuning
ⓘ
model selection ⓘ |
| refitBehavior | refits best_estimator_ on full training data when refit=True ⓘ |
| searchStrategy | exhaustive grid search ⓘ |
| selectionCriterion | maximizes scoring metric on validation folds ⓘ |
| supports |
classification
ⓘ
clustering if estimator supports scoring ⓘ regression ⓘ |
| supportsLanguage | Python ⓘ |
| usesTechnique | cross-validation ⓘ |
| writtenInLanguage | Python ⓘ |
Referenced by (2)
Full triples — surface form annotated when it differs from this entity's canonical label.