如何使用sklearn Pipeline & FeatureUnion选择多个(数字和文本)列进行文本分类？

from sklearn.pipeline import Pipeline, FeatureUnion
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.feature_extraction import DictVectorizer
from sklearn.svm import SVC

# 定义数字列和文本列的列名
numeric_columns = ['numeric_col1', 'numeric_col2']
text_columns = ['text_col1', 'text_col2']

# 定义数字列的特征提取器
numeric_transformer = Pipeline(steps=[
    # 可以根据需要添加其他预处理步骤，比如缺失值处理、标准化等
])

# 定义文本列的特征提取器
text_transformer = Pipeline(steps=[
    ('tfidf', TfidfVectorizer()),
    # 可以根据需要添加其他预处理步骤，比如去除停用词、词干提取等
])

# 使用FeatureUnion同时处理数字列和文本列的特征提取
preprocessor = FeatureUnion(transformer_list=[
    ('numeric_features', numeric_transformer),
    ('text_features', text_transformer),
])

# 定义分类器
classifier = SVC()

# 构建Pipeline
pipeline = Pipeline(steps=[
    ('preprocessor', preprocessor),
    ('classifier', classifier),
])

# 训练和预测
pipeline.fit(X_train, y_train)
y_pred = pipeline.predict(X_test)