我在尝试使用LinearSVC和OneVsRestClassifier计算SVM模型中的每个预测概率时遇到了错误

AttributeError: 'LinearSVC' object has no attribute 'predict_proba'

尝试的代码如下：

model = Pipeline([('vectorizer', CountVectorizer(ngram_range=(1,2))),    ('tfidf', TfidfTransformer(use_idf=True)),    ('clf', OneVsRestClassifier(LinearSVC(class_weight="balanced")))])model.fit(X_train, y_train)y_train.shapepred = model.predict(X_test)probas = model.predict_proba(X_test)

还尝试了以下方法：

from nltk.classify.scikitlearn import SklearnClassifierfrom sklearn.svm import SVCLinearSVC_classifier = SklearnClassifier(SVC(kernel='linear',probability=True))prob_1 = LinearSVC_classifier.predict_proba(X_test)

但仍然得到错误AttributeError: 'SklearnClassifier' object has no attribute 'predict_proba'

请提供解决建议。

回答：

使用您的线性SVM模型时：

from sklearn.calibration import CalibratedClassifierCVfrom sklearn.feature_extraction.text import TfidfVectorizerfrom sklearn.pipeline import FeatureUnion, make_pipelinefrom sklearn.svm import LinearSVCword_vectorizer = TfidfVectorizer(ngram_range=(1, 2))features = FeatureUnion([('words', word_vectorizer), ])calibrated_svc = CalibratedClassifierCV(LinearSVC(), method='sigmoid', cv=3)pipeline = make_pipeline(features, calibrated_svc)pipeline.fit(train_x, train_y)predicted = pipeline.predict_proba(test_x)

或者使用逻辑回归模型：

from sklearn.feature_extraction.text import TfidfVectorizerfrom sklearn.pipeline import FeatureUnion, make_pipelinefrom sklearn.linear_model import LogisticRegressionword_vectorizer = TfidfVectorizer(ngram_range=(1, 2))features = FeatureUnion([('words', word_vectorizer), ])pipeline = make_pipeline(features, LogisticRegression())pipeline.fit(train_x, train_y)predicted = pipeline.predict_proba(test_x)