-
Hi,
Pipe operator `mlr3pipelines::PipeOpTextVectorizer` is painfully slow in comparison with `quanteda::dfm()`:
````
library(mlr3)
library(mlr3learners)
library(mlr3pipelines)
library(quante…
-
seems to depend on https://github.com/mlr-org/mlr3/issues/470 which is closed now
mb706 updated
1 month ago
-
If so, it is not documented here:
https://github.com/mlr-org/mlr3/blob/d72be13637058ebabf00dc169ca51087db55b74d/R/Learner.R#L62-L64
If not, then the AutoTuner should be modified:
https://github…
mb706 updated
1 month ago
-
We might wanna be able to access validation split performance
-
It is not possible to build a stacked learner using mlr3 pipelines where the feature union uses a different task type than the final prediction. The following example is following the mlr3 book and st…
-
### Description
In `surv.aorsf`, when the `leaf_min_events` parameter is tuned, then allowed values depend on the number of events in the respective task (or subset of the task used for training). Th…
-
#271 solution currently needs to compare data to make sure clashing columns are not different. Once https://github.com/mlr-org/mlr3/issues/476 has an answer we can speed up this test.
mb706 updated
1 month ago
-
-
Currently mlr supports observation weights for a task. In situations when the observation weights could change in resampling training instances (time series) it would be beneficial to extend the Resam…
-
Loading e.g. a `Task` that was generated with an old mlr3-version using `readRDS()` will give errors like this: https://github.com/mb706/autoxgboost3/issues/2#issuecomment-596969950 . Maybe there is a…
mb706 updated
1 month ago