Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

modify script to use objective and eval_metric

Browse Source
This commit is contained in:
hetong007 2015-05-30 15:48:57 -07:00
parent 27e4cbb215
commit 36031d9a36
3 changed files with 10 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
R-package/demo/cross_validation.R
View File

@ -40,10 +40,10 @@ evalerror <- function(preds, dtrain) {
return(list(metric = "error", value = err)) return(list(metric = "error", value = err))
} }
param <- list(max.depth=2,eta=1,silent=1) param <- list(max.depth=2,eta=1,silent=1,
objective = logregobj, eval_metric = evalerror)
# train with customized objective # train with customized objective
xgb.cv(param, dtrain, nround, nfold = 5, xgb.cv(param, dtrain, nround, nfold = 5)
obj = logregobj, feval=evalerror)
# do cross validation with prediction values for each fold # do cross validation with prediction values for each fold
res <- xgb.cv(param, dtrain, nround, nfold=5, prediction = TRUE) res <- xgb.cv(param, dtrain, nround, nfold=5, prediction = TRUE)

3
R-package/demo/custom_objective.R
View File

@ -61,4 +61,5 @@ logregobjattr <- function(preds, dtrain) {
print ('start training with user customized objective, with additional attributes in DMatrix') print ('start training with user customized objective, with additional attributes in DMatrix')
# training with customized objective, we can also do step by step training # training with customized objective, we can also do step by step training
# simply look at xgboost.py's implementation of train # simply look at xgboost.py's implementation of train
bst <- xgb.train(param, dtrain, num_round, watchlist, logregobjattr, evalerror) bst <- xgb.train(param, dtrain, num_round, watchlist,
objective=logregobj, eval_metric=evalerror)

9
R-package/demo/early_stopping.R
View File

@ -31,9 +31,10 @@ evalerror <- function(preds, dtrain) {
return(list(metric = "error", value = err)) return(list(metric = "error", value = err))
} }
print ('start training with early Stopping setting') print ('start training with early Stopping setting')
# training with customized objective, we can also do step by step training
# simply look at xgboost.py's implementation of train bst <- xgb.train(param, dtrain, num_round, watchlist,
bst <- xgb.train(param, dtrain, num_round, watchlist, logregobj, evalerror, maximize = FALSE, objective = logregobj, eval_metric = evalerror, maximize = FALSE,
early.stop.round = 3) early.stop.round = 3)
bst <- xgb.cv(param, dtrain, num_round, nfold=5, obj=logregobj, feval = evalerror, bst <- xgb.cv(param, dtrain, num_round, nfold = 5,
objective = logregobj, eval_metric = evalerror,
maximize = FALSE, early.stop.round = 3) maximize = FALSE, early.stop.round = 3)