Does GridSearchCV store all the scores for all parameter combinations?
Yes it does, exactly as it is stated in the docs:
grid_scores_
: list of named tuplesContains scores for all parameter combinations in param_grid. Each entry corresponds to one parameter setting. Each named tuple has the attributes:
parameters
, a dict of parameter settingsmean_validation_score
, the mean score over the cross-validation foldscv_validation_scores
, the list of scores for each fold
allscores=model.cv_results_['mean_test_score']
print(allscores)