调整XGboost参数在R中

时间:2015-11-27 03:10:26

标签: r r-caret xgboost

我正在尝试使用R中的插入符号包调整参数,但得到

Error in train.default(x = as.matrix(df_train %>% select(-c(Response,  : 
  The tuning parameter grid should have columns nrounds, lambda, alpha 

每当我尝试训练模型时,即使列nrounds,lambda和alpha都在那里。

library(caret)
library(xgboost)
library(readr)
library(dplyr)
library(tidyr)

 xgb_grid_1 <- expand.grid(
  nrounds= 2400,
  eta=c(0.01,0.001,0.0001),
  lambda = 1,
  alpha =0
)

xgb_trcontrol <- trainControl(
  method="cv",
  number = 5,
  verboseIter = TRUE,
  returnData=FALSE,
  returnResamp = "all",
  allowParallel = TRUE,

)

xgb_train_1 <- train(
  x = as.matrix(df_train %>% select(-c(Response, Id))),
  y= df_train$Response,
 trControl = xgb_trcontrol,
 tuneGrid = xgb_grid_1,
 method="xgbLinear"
)

1 个答案:

答案 0 :(得分:7)

问题在于你的xgb_grid_1。如果删除行eta,它将起作用。

xgboost和插入符号的标准调整选项是“nrounds”,“lambda”和“alpha”。不是eta。使用modelLookup函数查看可用的模型参数。如果你也想使用eta,你必须创建自己的插入符号模型,以便在调优中使用这个额外的参数。

modelLookup("xgbLinear")
      model parameter                 label forReg forClass probModel
1 xgbLinear   nrounds # Boosting Iterations   TRUE     TRUE      TRUE
2 xgbLinear    lambda     L2 Regularization   TRUE     TRUE      TRUE
3 xgbLinear     alpha     L2 Regularization   TRUE     TRUE      TRUE