-
When creating a graph learner that has as input a task with tf columns, the `data_prototype` that is saved in the learner's state after training contains the `arg` and `value` vectors as well as the e…
-
Machine learning packages such as scikit-learn and mlr3 allow mtry/max_features type hyperparameters to be supplied (and tuned) as a proportion of available features ('max_features' in sklearn and 'mt…
-
Goal: Compare XGBoost with a simple multi layer perceptron in a simple benchmark experiment.
1. We need to define the tasks:
Use three simple small tasks from OpenML: https://mlr3book.mlr-org.c…
-
_Quelques arguments en vrac pour utiliser `R6` (orienté-objet encapsulé) plutôt que `S3`_
## ➕
- Standardiser la codebase (`mlr3`, que nous manipulons beaucoup, est implémenté en `R6`) ce qui fac…
-
We should probably implement a calibration plot as it seems to be requested and as it existed in mlr2:
https://stackoverflow.com/questions/66819169/how-to-draw-a-calibration-plot-of-a-binary-classi…
-
I am having issues with nested `future` calls.
Here I'm using `mlr3` and `furrr`, both using `future` package. When using them both together I'm having issues to have the parallelized calls to term…
-
Hey, we will release a new mlr3 version soon. Currently, we get with `mlr-org/mlr3@encapsulate` (next main):
```
* using log directory ‘/home/ubuntu/mlr3/revdep/checks/mlr3resampling/new/mlr3resa…
-
Hi mlr3proba team,
Some recent recommendations on evaluation/validation of survival prediction models call for estimation of individuals' observed outcome probabilities and the model predicted pro…
-
Based on our discussion, the next to-do list is `Tuning spaces `
> To-Do list from last meeting:
> - benchmark comparsion facetting (waiting for review)
> - ROC plots (waiting for review)
> - Tu…
-
First, thank you for providing this great dataset!
For the mlr3 project (www.github.com/mlr-org/mlr3) we've been replacing all instances of the problematic `iris` dataset with `penguins`. However o…