带有XGBoost错误的贝叶斯优化：“某些尾随字符无法解析：'Inf'”

Question

我正在尝试使用R中的另一个数据集来复制此方法。

https://www.kaggle.com/btyuhas/bayesian-optimization-with-xgboost

我的数据缺少值，但没有Inf值。但是，它给了我错误信息：

xgb.iter.update（fd $ bst，fd $ dtrain，迭代-1，obj）中的错误：   无法解析某些结尾字符：'Inf'

计时停在：0.39 0.19 0.58

我在线搜索，但找不到相关答案。

谢谢！

dtrain = xgb.DMatrix(data = as.matrix(x_train), label=y_train)
dtest = xgb.DMatrix(as.matrix(x_test))


xgb_evaluate = function(max_depth, gamma, colsample_bytree){
  params = list(max_depth = max_depth,
                subsample = 0.8, 
                eta = 0.1,
                gamma = gamma,
                colsample_bytree = 0.3,
                eval_metric = "rmse")

  # Used around 1000 boosting rounds in the full model
  cv_result = xgb.cv(params, dtrain, nround=100, nfold=3)    

  # Bayesian optimization only knows how to maximize, not minimize, so return the negative RMSE
  return (-1.0 *tail(xgboostModelCV$evaluation_log$test_rmse_mean, 1))
}


xgb_bo = BayesianOptimization(xgb_evaluate, 
                              bounds = list(max_depth = c(4L, 6L),
                                            gamma = c(0, 1), 
                                            colsample_bytree = c(0.3, 0.9)), 
                              n_iter = 20)