如何从一个已拟合的GridSearchCV中提取最佳管道，以便将其传递给cross_val_predict？
直接传递已拟合的GridSearchCV对象会导致cross_val_predict再次运行整个网格搜索，我只想让最佳管道接受cross_val_predict的评估。
我的自包含代码如下：
from sklearn.datasets import fetch_20newsgroupsfrom sklearn.feature_extraction.text import TfidfVectorizerfrom sklearn.svm import SVCfrom sklearn.multiclass import OneVsRestClassifierfrom sklearn.pipeline import Pipelinefrom sklearn.grid_search import GridSearchCVfrom sklearn.model_selection import cross_val_predictfrom sklearn.model_selection import StratifiedKFoldfrom sklearn import metrics# fetch data datanewsgroups = fetch_20newsgroups(remove=('headers', 'footers', 'quotes'), categories=['comp.graphics', 'rec.sport.baseball', 'sci.med'])X = newsgroups.datay = newsgroups.target# setup and run GridSearchCVwordvect = TfidfVectorizer(analyzer='word', lowercase=True)classifier = OneVsRestClassifier(SVC(kernel='linear', class_weight='balanced'))pipeline = Pipeline([('vect', wordvect), ('classifier', classifier)])scoring = 'f1_weighted'parameters = { 'vect__min_df': [1, 2], 'vect__max_df': [0.8, 0.9], 'classifier__estimator__C': [0.1, 1, 10] }gs_clf = GridSearchCV(pipeline, parameters, n_jobs=8, scoring=scoring, verbose=1)gs_clf = gs_clf.fit(X, y)### outputs: Fitting 3 folds for each of 12 candidates, totalling 36 fits# manually extract the best models from the grid search to re-build the pipelinebest_clf = gs_clf.best_estimator_.named_steps['classifier']best_vectorizer = gs_clf.best_estimator_.named_steps['vect']best_pipeline = Pipeline([('best_vectorizer', best_vectorizer), ('classifier', best_clf)])# passing gs_clf here would run the grind search again inside cross_val_predicty_predicted = cross_val_predict(pipeline, X, y)print(metrics.classification_report(y, y_predicted, digits=3))我目前的做法是从best_estimator_手动重建管道。但我的管道通常有更多步骤，比如SVD或PCA，有时我会添加或删除步骤并重新运行网格搜索来探索数据。然后在手动重建管道时，这一步总是需要重复，这很容易出错。
有没有办法直接从已拟合的GridSearchCV中提取最佳管道，以便将其传递给cross_val_predict？
回答：
y_predicted = cross_val_predict(gs_clf.best_estimator_, X, y)有效并返回：
Fitting 3 folds for each of 12 candidates, totalling 36 fits[Parallel(n_jobs=4)]: Done 36 out of 36 | elapsed: 43.6s finished precision recall f1-score support 0 0.920 0.911 0.916 584 1 0.894 0.943 0.918 597 2 0.929 0.887 0.908 594avg / total 0.914 0.914 0.914 1775[编辑] 当我再次尝试代码时，简单地传递pipeline（原始管道），它返回了相同的结果（传递best_pipeline也是如此）。所以你可能可以直接使用管道本身，但我对此不是100%确定。
