RandomizedSearchCV

E97068

RandomizedSearchCV is a scikit-learn tool that performs hyperparameter optimization by randomly sampling parameter combinations and evaluating them via cross-validation.

Jump to: Statements Referenced by

Statements (42)

Predicate Object
instanceOf hyperparameter optimization tool
model selection utility
scikit-learn class
advantage can find good configurations with fewer evaluations than grid search
explores large hyperparameter spaces efficiently
canOptimize any estimator with fit method
definedInModule sklearn.model_selection
differsFrom GridSearchCV by using random sampling instead of exhaustive search
documentationURL https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.RandomizedSearchCV.html
hasAttribute best_estimator_
best_params_
best_score_
cv_results_
n_splits_
hasParameter cv
error_score
estimator
iid
n_iter
n_jobs
param_distributions
pre_dispatch
random_state
refit
return_train_score
scoring
verbose
inheritsFrom BaseSearchCV
introducedFor model selection in scikit-learn
language Python
license BSD license (through scikit-learn)
output fitted estimator with best found hyperparameters
partOf scikit-learn
performs hyperparameter optimization
requires parameter distributions or lists in param_distributions
samples parameter combinations at random
similarTo GridSearchCV
supports multiple scoring metrics via scoring parameter
parallel computation via n_jobs
randomized hyperparameter search
typicalUseCase tuning machine learning model hyperparameters
uses cross-validation

Referenced by (1)

Full triples — surface form annotated when it differs from this entity's canonical label.

scikit-learn hasConcept RandomizedSearchCV