使用Voting Classifier集成KNN与决策树结合Pipeline的参数配置问题

参数网格的命名规则错了。Pipeline里嵌套的分类器参数得按步骤层级指定：你的VotingClassifier是Pipeline中名为regressor的步骤，所以它内部的KNN和随机森林参数必须加上regressor__前缀，否则GridSearch找不到对应的参数项。

import numpy as np
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler
from sklearn.ensemble import VotingClassifier, RandomForestClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.model_selection import train_test_split, GridSearchCV

# 步骤定义保留原逻辑
steps = [('scaler', StandardScaler()),
        ('regressor', VotingClassifier(estimators=[
        ('knn', KNeighborsClassifier()), 
        ('clf', RandomForestClassifier())],voting='soft'))]
pipeline = Pipeline(steps)

# 关键修正：给所有子分类器参数加上regressor__前缀
parameters = [
    {'regressor__knn__n_neighbors': np.arange(1, 50)},
    {
        'regressor__clf__n_estimators': [10, 20, 30],
        'regressor__clf__criterion': ['gini', 'entropy'],
        'regressor__clf__max_features': [5, 10, 15],
        'regressor__clf__max_depth': ['auto', 'log2', 'sqrt', None]
    }
]

# 后续训练逻辑不变
X_train, X_test, y_train, y_test = train_test_split(X, y.values.ravel(),
test_size=0.3, random_state=65)
cv = GridSearchCV(pipeline, param_grid=parameters)
cv.fit(X_train, y_train)
y_pred = cv.predict(X_test)

如果以后再遇到类似参数找不到的问题，直接打印Pipeline的所有可用参数就能明确正确命名：

print(pipeline.get_params().keys())

运行后会看到像regressor__knn__n_neighbors、regressor__clf__n_estimators这类完整参数名，照着用就行。

内容的提问来源于stack exchange，提问作者Hamza Abbasi