search

mlr-org / mlr3learners

Recommended learners for mlr3
https://mlr3learners.mlr-org.com
GNU Lesser General Public License v3.0
89 stars 14 forks source link

Explicitly set eval_metric in xgboost #218

Closed mllg closed 2 years ago

mllg commented 2 years ago

Fixes #210 .

mllg commented 2 years ago

@be-marc do you have an idea why the param tests fail? 

Failure (test_paramtest_classif.xgboost.R:63:3): predict classif.xgboost
`ParamTest` is not TRUE

`actual` is a list
`expected` is a logical vector (TRUE)

Missing parameters in mlr3 param set:
- iterationrange
- strict_shape

I cannot find iterationrange or strict_shape in our package nor on the website :/

be-marc commented 2 years ago

iterationrange and strict_shape are parameters of xgboost:::predict.xgb.Booster in the latest CRAN version (08.11.2021). Our tests seem to work again.