-
In mlr we maybe have wrong default values, for example in the `?mboost::boost_control` documentation it says that `center = TRUE` is the default. In our mlr param set, the default is `FALSE`, see:
`…
-
Just for confirmation.If it is, which part of source code reveales this?By the way, what is the usage of `loss` agrument in `Family` object. Thanks in advance!
-
I would like to see support for quantile regression in mlr.
I have added a first cut in this branch:
https://github.com/ktcorby/mlr/tree/quantile_regression
This includes two new base learners…
-
Look in fixDataForLearnner.R.
Status quo:
- this function is used to "enhance" learner capabilities.
- it can do 3 things:
-- restore.levels in features. so if factor levels are dropped from fea…
-
I'm finding that `CatBoostRegressor` isn't minimising the mean squared error for my problem, I can trivially improve the mse by adding / subtracting a constant.
i.e.
```
offsets = np.arange(-15…
-
Using the attached data `example_data.csv`, I get the following error:
```
> example_data summary(example_data)
target x1 x2
Min. : 53.75 Min. :…
-
``` r
library(FDboost)
#> Loading required package: mboost
#> Loading required package: parallel
#> Loading required package: stabs
#> This is mboost 2.9-1. See 'package?mboost' and 'news(packa…
-
Using the following example (highly whittled down from data in my actual application), I get a warning about a failed call to `min` inside the training:
```
> data.frame(target = c(46, 42, 42, 39,…
-
Hi Prof. Toby,
I have been running IntervalRegression for the neuroblastoma dataset. It has been running for long. Please find the code attached. How i can resolve this issue?
[intervalCV.r.zip](h…
-
I'm really not understanding the connection to the parallel package here. Further documentation/examples would be highly appreciated.
Especially from the following (and only) example i'm really havi…
hkosm updated
5 years ago