After we are done with #19 we need to test this structure and implement nested grid search in sklearn-like object. best_estimator_ and best_params_ should return lists of estimators and best_params.
Methods fit and predict use mean of all best models with best params.
After we are done with #19 we need to test this structure and implement nested grid search in sklearn-like object. best_estimator_ and best_params_ should return lists of estimators and best_params.
Methods fit and predict use mean of all best models with best params.