John Zobolas

Also I somehow can't view any "Construction" code on the website, e.g. this is a [blank grey box](https://mlr3pipelines.mlr-org.com/reference/mlr_pipeops_pca.html), but can copy the code!

PCA issue fixed in 9e6b03c38807c9657597fbb03d1f8fb687a48b28. The pkgdown issue is more general so probably you would want to put it someplace else?

Hi! I wanted to add some more tests and I stumbled upon some things that I need to clear up first: 1. When I ran `test_pkg('xfun')` from the RStudio console,...

I have started writing some tests in a separate branch - @yihui would you mind giving me access to push it upstream at some point? (I will send a PR...

Hi Raphael, Thanks for your speedy response! Regarding the "touchy subject" definitely wrong phrasing - more like a confusing matter to me since I've heard different opinions about R packages...

That's a nice idea with the `$model` property! Desing-wise, should it be part of the `measure` class or the `prediction` class (like `$score` is)? Wouldn't it be a problem if...

Hi! Any updates from the `mlr3` team regarding this issue? I think the proper solution here would be to allow `MeasureSurv` to return measures at multiple time-points simultaneously because right...

Hi Raphael, Great to find another person who has found survival SVMs unstable. I wouldn't recommend this learner to anyone unless hyperparameters are [hand-picked](https://miro.medium.com/max/1136/0*xMWaKlmLBHcB4AfP.jpg) and no proper tuning is applied...

Hi @Liripo, I have not. Thumping up the original question is I guess the only way to get attention

1) Many other survival learners don't have the first predict type as it should (i.e. first element in the list) - easy to fix 2) The warning has a small...