[LRNRQ] Add <OPLS(-DA)> from package <ropls>

[abadgerw]

Algorithm

OPLS(-DA)

Package

ropls

Supported types

• [ X ] classif
• clust #289
• dens
• [ X ] regr
• surv

I have checked that this is not already implemented in

• [ X ] mlr3
• [ X ] mlr3learners
• [ X ] mlr3extralearners
• [ X ] Other core packages (e.g. mlr3proba, mlr3keras)

Why do I think this is a useful learner?

OPLS-DA is commonly used in omic studies, especially metabolomics, and would be useful to integrate with all the other amazing pipeline features of mlr3.

Further Optional Comments

(e.g. if this algorithm is already implemented under a different learner then please explain why this additional implementation is useful)

[sebffischer]

Hey, thanks for your interest in mlr3!
Do you want to give the implementation a try?
You can read about how to add learners to mlr3 here

[abadgerw]

Thanks, @sebffischer! I unfortunately don't have the background or time at the moment to learn how to implement. Therefore, it would be much appreciated if you or your team could do so.

[abadgerw]

Hope all is well! I just wanted to circle back on the potential of getting this implemented?

[sebffischer]

Sorry, I just haven't had the time to work on this.
There might be someone else soon who will work on implementing learners and we can start with your request, but I am not sure when exactly this will happen (Probably around this month or next)

[abadgerw]

@sebffischer Hope you had a nice thanksgiving! I just wanted to see if you had an updated timeline for this?

In addition, I had opened an issue for mlr3measures: mlr-org/mlr3measures#40 but haven't received a reply. Is there a different repository I should post this request under?

[sebffischer]

Ah, sorry for the missing update. The person that I had in mind was now given other tasks. Also I myself don't have the time to work on this still ... If you really need this, I suggest you read this tutorial and use the create_learner() function from mlr3extralearners to implement this yourself (it is really not that difficult)! I will be available for all questions and will review the pull request as well :)