R Xgboost validation error as stopping metric - r

Question

I am using a train and validation dataset on an xgboost binary classification model.
params5 <- list(booster = "gbtree", objective = "binary:logistic",
eta=0.0001, gamma=0.5, max_depth=15, min_child_weight=1, subsample=0.6,
colsample_bytree=0.4,seed =2222)
xgb_MOD5 <- xgb.train (params = params5, data = dtrain, nrounds = 4000,
watchlist = list(validation = dvalid,train = dtrain),
print_every_n =30,early_stopping_rounds = 100
maximize = F ,serialize = TRUE)
It automatically picks the train error as stopping metric. This results in the model continuing to train while overfitting.
Multiple eval metrics are present. Will use train_error for early stopping.
Will train until train_error hasn't improved in 100 rounds.
How do I assign the validation error as stopping metric?

Answer 1

I do not use the R binding of xgboost and the R-package documentation is not specific about it. However, the python-API documentation (see the early_stopping_rounds argument documentation) has a relevant clarification on this issue:
Requires at least one item in evals. If there’s more than one, will use the last.
Here, evals is the list of samples on which metrics will be evaluated, i.e. it is analogous to your watchlist argument. So I would guess, it can be that you just need to swap the order of items in the list provided as that argument

Answer 2

Thanks #abhiieor for the solution. Adding to that from what I observed,when we use only the validation in watchlist:
xgb_MOD5 <- xgb.train (params = params5, data = dtrain, nrounds = 400,watchlist = list(validation = dvalid),
print_every_n =30,early_stopping_rounds = 100, maximize = F ,serialize = TRUE)
log results while it runs:
[1] validation-error:0.222037
Will train until validation_error hasn't improved in 100 rounds.
[31] validation-error:0.201712
[61] validation-error:0.201635
And if we want to see both the train error and validation error while it runs,
adding the validation as 2nd argument in the watch list did it while using validation error as the stopping metric .
xgb_MOD5 <- xgb.train (params = params5, data = dtrain, nrounds = 400,watchlist = list(train =dtrain,validation = dvalid),
print_every_n =30,early_stopping_rounds = 100, maximize = F ,serialize = TRUE)
[1] train-error:0.202131 validation-error:0.232341
Multiple eval metrics are present. Will use validation_error for early stopping.
Will train until validation_error hasn't improved in 100 rounds.
[31] train-error:0.174278 validation-error:0.202871
[61] train-error:0.173909 validation-error:0.202288