利用lightgbm进行顺序向前选择算法

Question

# 分离特征和目标变量
y = '23h结果'
X= [x for x in data.columns if x not in [y, '23h结果']]

X=pd.DataFrame(X)
# 定义LightGBM模型
params = {'boosting_type': 'gbdt',
          'objective': 'regression',
          'metric': ('mae','mse','r2'),# 默认使用MAE作为评价指标
          'num_leaves': 31,
          'learning_rate': 0.05,
          'feature_fraction': 0.9}

# 顺序向前选择特征
selected_features = []
score_list = []
for i in range(len(X.columns)):
    scores = []
    for feature in X.columns:
        if feature not in selected_features:
            features = selected_features + [feature]
            X_selected = X[features]
            # 划分训练集和测试集
            X_train, X_test, y_train, y_test = train_test_split(X_selected, y, test_size=0.2, random_state=42)
            lgb_train = lgb.Dataset(X_train, y_train)
            lgb_test = lgb.Dataset(X_test, y_test, reference=lgb_train)
            # 训练模型并预测
            gbm = lgb.train(params, lgb_train, num_boost_round=1000, valid_sets=lgb_test, early_stopping_rounds=10,
                            verbose_eval=False)
            y_pred = gbm.predict(X_test)

运行上述代码时，出现了如下错误

Traceback (most recent call last):
  File "C:/Users/86187/PycharmProjects/pythonProject10/23h预测.py", line 49, in <module>
    X_train, X_test, y_train, y_test = train_test_split(X_selected, y, test_size=0.2, random_state=42)
  File "D:\py\anaconda3\lib\site-packages\sklearn\model_selection\_split.py", line 2559, in train_test_split
    arrays = indexable(*arrays)
  File "D:\py\anaconda3\lib\site-packages\sklearn\utils\validation.py", line 443, in indexable
    check_consistent_length(*result)
  File "D:\py\anaconda3\lib\site-packages\sklearn\utils\validation.py", line 397, in check_consistent_length
    raise ValueError(
ValueError: Found input variables with inconsistent numbers of samples: [18, 5]

具体解决方是什么？

Answer 1

根据报错信息，输入变量的样本数量不一致，X_selected的样本数量为18，而y的数量为5。可能是由于在分离特征和目标变量时，没有对y进行正确的处理。应该修改为：

# 分离特征和目标变量
y = data['23h结果']
X = data.drop(['23h结果'], axis=1)