mlr3learners
mlr3learners copied to clipboard
mlr-org
Recommended learners for mlr3
The test sets are added to the watchlist and early stopping is performed. ```r learner = lrn("classif.xgboost", early_stopping_rounds = 10, early_stopping_set = "test", nrounds = 100, eval_metric = "error") rr...
https://mlr3extralearners.mlr-org.com/articles/learners/test_overview.html prints the following warning ``` ## Warning in as.data.table.list(x, keep.rownames = keep.rownames, check.names ## = check.names, : Item 2 has 85 rows but longest item has 87; recycled with...
[CRAN](https://cran.r-project.org/web/packages/h2o/h2o.pdf)
Using `predcontrib = TRUE` to get Shapley values with the xgboost classifier learner causes errors because of mlr3 expecting a different prediction format. With `predcontrib = TRUE` xgboost returns a...
Requested via private mail and also here for mlr: https://stackoverflow.com/questions/60110684/is-there-a-way-to-set-the-base-margin-for-xgboost-in-the-mlr-framework This could just be a `p_fct()` specifying the right column.
RegrKm SE prediction with nugget currently predicts the epistemic uncertainty (uncertainty of the mean prediction) + aleatoric uncertainty (nugget SE, uncertainty that the model sees as random error). E.g. sampling...
Before making this request, make sure that 1. The learner is not maintained in a third party repository (listed in the [Table of all additional Learners](https://mlr3learners.mlr-org.com/dev/articles/learners/additional-learners.html)), and 2. There is...
It would be nice to have some functions like: ```R install_mlr3_learner("classif.ranger") -> `Learner` install_mlr3_learners(c("classif.ranger", "classif.featureless")) -> `list` ``` Which on first run can automatically add the relevant code to install...
xgboost performance + intrinsic interpretability = interesting addition in times of interpretable ML demand for trust
