GridSearchCV:XGBRegressionor中的参数错误

Question

我使用GridSearchCV查找XGB模型中的最佳参数，下面是我的代码

# Xgboost
grid_xgb = {
    'booster': 'gbtree',
    'objective': 'binary:logistic',
    'subsample': [0.6,0.7,0.8,0.9],
    'colsample_bytree': [0.6,0.7,0.8,0.9],
    'eta': [0.05,0.1,0.2,0.3],
    'max_depth': [3,5,7],
    'seed': [2021,2022],
    'eval_metric': 'logloss'
}

model_xgb = xgb.XGBRegressor()
search_xgb = GridSearchCV(estimator = model_xgb, 
                         param_grid = grid_xgb, 
                         cv = 5,
                         n_jobs = -1,
                         verbose = 2)

但我得到一个错误信息：

 Parameter values for parameter (booster) need to be a sequence(but not a string) or np.ndarray

我检查了XGB文档，它说

“助推器[默认值=gbtree]

要使用的助推器。可以是gbtree、gblinear或dart；gbtree和dart使用基于树的模型，而gblinear使用线性函数。”

所以，我不知道为什么会发生错误，也不知道该怎么解决

Answer 1

如scikit-learn documentation中所述，不同的参数值需要作为列表传递给GridSearchCV，这意味着booster、objective和eval_metric也需要用方括号括起来：

grid_xgb = {
    'booster': ['gbtree'],
    'objective': ['binary:logistic'],
    'subsample': [0.6,0.7,0.8,0.9],
    'colsample_bytree': [0.6,0.7,0.8,0.9],
    'eta': [0.05,0.1,0.2,0.3],
    'max_depth': [3,5,7],
    'seed': [2021,2022],
    'eval_metric': ['logloss']
}