pandas基于特定列创建交叉验证

import pandas as pd
from sklearn.model_selection import KFold

# 读取数据集
data = pd.read_csv('data.csv')

# 选择特征列和目标列
features = data[['feature1', 'feature2', 'feature3']]
target = data['target']

# 创建交叉验证对象
kfold = KFold(n_splits=5, random_state=42)

# 进行交叉验证
for train_index, test_index in kfold.split(data):
    # 获取训练集和测试集
    train_features, test_features = features.iloc[train_index], features.iloc[test_index]
    train_target, test_target = target.iloc[train_index], target.iloc[test_index]
    
    # 在训练集上训练模型
    model.fit(train_features, train_target)
    
    # 在测试集上评估模型
    score = model.score(test_features, test_target)
    
    # 打印模型评估结果
    print("Model score: ", score)