SVM 交叉验证中每个标签的准确性？

如何解决SVM 交叉验证中每个标签的准确性？

我有以下 df

         text     count     daytime        label
   I think...        4      morning          pos
You should...        3    afternoon          neg
    Better...        7      evening          neu

我尝试使用 ColumnTransform 仅预处理 text 列

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.compose import ColumnTransformer

transformer = ColumnTransformer([
    ('vectorizer',TfidfVectorizer(ngram_range=(1,1)),'text')
],remainder='passthrough')

然后我使用以下代码分别应用 count 和 daytime

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.compose import ColumnTransformer
from sklearn.preprocessing import StandardScaler
from sklearn.preprocessing import OneHotEncoder

transformer = ColumnTransformer([
    ('vectorizer','text'),('scaler',StandardScaler(),['count']),('enc',OneHotEncoder(),['daytime'])
],remainder='passthrough')

X_transformed = transformer.fit_transform(X)

之后，我使用 shufflesplit 对其进行交叉验证并使用 SVM 对其进行训练

from sklearn.svm import SVC
from sklearn.model_selection import ShuffleSplit

model = SVC(kernel='linear')
cv = ShuffleSplit(n_splits=10,test_size=0.2,random_state=69)

之后，我使用 cross_val_score 应用 mdel

scores = cross_val_score(model,X_vec,y,cv=cv,scoring='accuracy')

输出是每次迭代的准确性

[0.33150734 0.08022311 0.03531764]

是否可以知道每个标签的分类报告？

喜欢这个输出示例

   precision    recall  f1-score   support

          pos       1.00      0.48      0.65       702
          neg       0.78      1.00      0.88      1276
          neu       0.78      1.00      0.88      1276

     accuracy                           0.82      1978
    macro avg       0.89      0.74      0.76      1978
 weighted avg       0.86      0.82      0.80      1978

我试过了

print(classification_report(y,scores))

按照建议。但它给出了像

这样的错误

Found input variables with inconsistent numbers of samples: [10761,10]