반응형
이번엔 그리드 탐색을 이용한 준비 단계의 옵션을 자동 탐색하는 방법을 알아보겠습니다.
param_grid = [
('preparation__num_pipeline__imputer__strategy' : ['mean', 'median', 'most_frequent'],
'feature_selection__k' : list(range(1,len(feature_importances) + 1 ))}
]
grid_search_prep=GridSearchCV(prepare_select_and_predict_pipeline,param_grid, cv=5, scoring='neg_mean_squared_error', verbose=2, n_jobs=1)
grid_search_prep.fit(housing, housing_labels)
이렇게 코드를 작성해주시고 한참을 기다리시면 됩니다..
Fitting 5 folds for each of 48 candidates, totalling 240 fits
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=mean
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=mean, total= 14.9s
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=mean
[Parallel(n_jobs=1)]: Done 1 out of 1 | elapsed: 22.8s remaining: 0.0s
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=mean, total= 15.5s
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=mean, total= 14.9s
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=mean, total= 15.0s
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=mean, total= 14.8s
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=median, total= 14.8s
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=median, total= 15.9s
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=median, total= 15.3s
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=median, total= 14.7s
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=median, total= 15.2s
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.0s
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=most_frequent, total= 15.6s
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=most_frequent, total= 15.8s
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=most_frequent, total= 15.5s
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=1, preparation__num_pipeline__imputer__strategy=most_frequent, total= 15.4s
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=mean, total= 14.8s
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=mean, total= 15.0s
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=mean, total= 15.3s
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=mean, total= 15.0s
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=mean, total= 16.0s
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=median, total= 16.4s
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=median, total= 16.9s
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=median, total= 16.9s
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=median, total= 17.2s
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=median, total= 15.9s
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.4s
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=most_frequent, total= 17.0s
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.3s
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.1s
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=2, preparation__num_pipeline__imputer__strategy=most_frequent, total= 15.8s
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=mean, total= 16.1s
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=mean, total= 14.8s
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=mean, total= 14.6s
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=mean, total= 14.9s
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=mean, total= 14.9s
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=median, total= 15.5s
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=median, total= 14.9s
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=median, total= 14.8s
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=median, total= 15.1s
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=median, total= 14.9s
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=most_frequent, total= 15.9s
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=most_frequent, total= 15.9s
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=most_frequent, total= 15.7s
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=most_frequent, total= 15.8s
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=3, preparation__num_pipeline__imputer__strategy=most_frequent, total= 15.8s
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=mean, total= 15.6s
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=mean, total= 15.3s
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=mean, total= 15.6s
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=mean, total= 15.7s
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=mean, total= 15.3s
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=median, total= 15.6s
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=median, total= 15.3s
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=median, total= 15.6s
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=median, total= 15.6s
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=median, total= 15.5s
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=most_frequent, total= 17.7s
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=most_frequent, total= 17.3s
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.4s
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.6s
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=4, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.4s
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=mean, total= 15.6s
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=mean, total= 16.6s
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=mean, total= 16.7s
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=mean, total= 17.0s
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=mean, total= 15.6s
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=median, total= 16.2s
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=median, total= 16.0s
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=median, total= 15.5s
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=median, total= 15.6s
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=median, total= 15.4s
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.3s
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.4s
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.3s
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.6s
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=5, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.4s
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=mean, total= 15.7s
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=mean, total= 16.5s
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=mean, total= 16.0s
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=mean, total= 16.5s
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=mean, total= 18.0s
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=median, total= 18.3s
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=median, total= 19.3s
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=median, total= 19.2s
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=median, total= 17.9s
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=median, total= 18.2s
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=most_frequent, total= 17.5s
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.4s
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.6s
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.2s
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=6, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.1s
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=mean, total= 15.8s
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=mean, total= 15.7s
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=mean, total= 16.2s
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=mean, total= 16.0s
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=mean, total= 15.7s
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=median, total= 16.6s
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=median, total= 15.9s
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=median, total= 15.9s
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=median, total= 15.5s
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=median, total= 15.6s
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=most_frequent, total= 17.3s
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.6s
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=most_frequent, total= 17.1s
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=most_frequent, total= 16.6s
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=7, preparation__num_pipeline__imputer__strategy=most_frequent, total= 17.2s
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=mean, total= 17.8s
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=mean, total= 17.2s
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=mean, total= 19.0s
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=mean, total= 18.2s
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=mean, total= 19.3s
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=median, total= 18.4s
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=median, total= 17.4s
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=median, total= 18.6s
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=median, total= 18.8s
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=median, total= 18.4s
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=most_frequent, total= 18.8s
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=most_frequent, total= 18.3s
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=most_frequent, total= 19.4s
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=most_frequent, total= 18.1s
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=8, preparation__num_pipeline__imputer__strategy=most_frequent, total= 19.1s
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=mean, total= 22.6s
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=mean, total= 20.5s
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=mean, total= 22.1s
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=mean, total= 22.3s
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=mean, total= 21.0s
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=median, total= 21.5s
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=median, total= 26.7s
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=median, total= 21.7s
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=median, total= 23.1s
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=median, total= 23.0s
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=most_frequent, total= 23.4s
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=most_frequent, total= 22.7s
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=most_frequent, total= 22.5s
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=most_frequent, total= 21.8s
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=9, preparation__num_pipeline__imputer__strategy=most_frequent, total= 23.0s
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=mean, total= 23.2s
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=mean, total= 24.5s
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=mean, total= 28.6s
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=mean, total= 26.0s
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=mean, total= 24.8s
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=median, total= 23.6s
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=median, total= 27.0s
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=median, total= 25.2s
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=median, total= 25.4s
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=median, total= 24.2s
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=most_frequent, total= 24.8s
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=most_frequent, total= 25.5s
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=most_frequent, total= 24.6s
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=most_frequent, total= 26.6s
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=10, preparation__num_pipeline__imputer__strategy=most_frequent, total= 27.0s
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=mean, total= 29.7s
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=mean, total= 27.2s
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=mean, total= 27.4s
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=mean, total= 29.1s
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=mean, total= 28.0s
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=median, total= 25.3s
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=median, total= 25.1s
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=median, total= 25.5s
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=median, total= 29.7s
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=median, total= 28.8s
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=most_frequent, total= 31.3s
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=most_frequent, total= 26.6s
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=most_frequent, total= 27.2s
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=most_frequent, total= 28.8s
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=11, preparation__num_pipeline__imputer__strategy=most_frequent, total= 33.4s
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=mean, total= 30.5s
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=mean, total= 30.4s
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=mean, total= 31.7s
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=mean, total= 30.8s
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=mean, total= 31.8s
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=median, total= 29.3s
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=median, total= 29.4s
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=median, total= 31.3s
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=median, total= 30.1s
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=median, total= 31.0s
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=most_frequent, total= 28.6s
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=most_frequent, total= 30.7s
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=most_frequent, total= 35.1s
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=most_frequent, total= 29.5s
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=12, preparation__num_pipeline__imputer__strategy=most_frequent, total= 30.7s
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=mean, total= 35.3s
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=mean, total= 33.5s
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=mean, total= 34.3s
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=mean, total= 32.2s
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=mean, total= 27.7s
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=median, total= 30.3s
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=median, total= 34.3s
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=median, total= 35.4s
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=median, total= 34.4s
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=median, total= 31.4s
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=most_frequent, total= 30.4s
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=most_frequent, total= 34.5s
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=most_frequent, total= 35.6s
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=most_frequent, total= 35.0s
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=13, preparation__num_pipeline__imputer__strategy=most_frequent, total= 31.7s
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=mean, total= 28.6s
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=mean, total= 32.9s
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=mean, total= 33.1s
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=mean, total= 33.5s
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=mean, total= 31.5s
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=median, total= 33.4s
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=median, total= 34.7s
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=median, total= 33.7s
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=median, total= 34.0s
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=median, total= 31.0s
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=most_frequent, total= 34.7s
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=most_frequent, total= 30.8s
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=most_frequent, total= 32.6s
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=most_frequent, total= 35.9s
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=14, preparation__num_pipeline__imputer__strategy=most_frequent, total= 42.1s
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=mean, total= 34.8s
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=mean, total= 33.7s
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=mean, total= 35.7s
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=mean, total= 29.7s
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=mean, total= 32.8s
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=median, total= 30.3s
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=median, total= 34.3s
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=median, total= 36.3s
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=median, total= 36.3s
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=median, total= 35.7s
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=most_frequent, total= 38.5s
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=most_frequent, total= 38.8s
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=most_frequent, total= 39.1s
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=most_frequent, total= 33.5s
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=15, preparation__num_pipeline__imputer__strategy=most_frequent, total= 42.4s
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=mean, total= 37.1s
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=mean, total= 39.4s
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=mean, total= 32.3s
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=mean, total= 35.3s
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=mean
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=mean, total= 30.2s
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=median, total= 31.6s
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=median, total= 34.7s
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=median, total= 33.1s
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=median, total= 30.0s
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=median
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=median, total= 33.7s
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=most_frequent, total= 32.1s
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=most_frequent, total= 35.6s
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=most_frequent, total= 33.6s
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=most_frequent, total= 33.8s
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=most_frequent
[CV] feature_selection__k=16, preparation__num_pipeline__imputer__strategy=most_frequent, total= 36.4s
[Parallel(n_jobs=1)]: Done 240 out of 240 | elapsed: 125.6min finished
무려 2시간;; 이 지난 후, grid_search_prep.best_params_로 최상의 Imputer 정책이 무엇인지 알아봅니다.
그 결과, 최적의 Strategy는 Most_frequent 최빈값이 나왔고,
거의 모든 특성 (16개 특성 중 15개)이 유용하다는 결과를 얻을 수 있었습니다.
마지막 특성인 ISLAND는 잡음만이 추가될 뿐이라는 점을 알 수 있습니다.
블로그
출처
이 글의 상당 부분은 [핸즈온 머신러닝, 한빛미디어/오렐리앙 제롱/박해선] 서적을 참고하였습니다.
나머지는 부수적인 함수나 메서드에 대해 부족한 설명을 적어두었습니다.
학습용으로 포스팅 하는 것이기 때문에 복제보다는 머신러닝에 관심이 있다면 구매해보시길 추천합니다.
도움이 되셨다면 로그인 없이 가능한
아래 하트♥공감 버튼을 꾹 눌러주세요!
반응형
'## 오래된 게시글 (미관리) ## > Python (Linux)' 카테고리의 다른 글
| 22. Python - 다중 분류 (0) | 2018.11.30 |
|---|---|
| 21. Python - MNIST를 이용한 분류 기초와 성능 평가 지표 (0) | 2018.11.28 |
| 19. Python - 전체 데이터 준비와 예측을 하나의 파이프라인으로 만들기 (0) | 2018.11.28 |
| 18. Python - 중요한 특성을 선택하는 변환기를 파이프라인에 추가하기 (0) | 2018.11.28 |
| 17. Python - 그리드 탐색을 랜덤 탐색으로 바꾸기 (2) | 2018.11.28 |
