Closed NickCH-K closed 3 years ago
Thanks Nick. I'm actually not planning on writing a machine learning package. I'm working on a tutorial SJ paper, demonstrating lots of tricks that can be done with rcall
. Your package seems very promising though! Why not working further on it and getting it on Stata Journal?
Hmm, I hadn't considered the possibility of it going to SJ, maybe that's enough incentive to expand it. And good to know, thanks!
welcome! And I can tell that the topic is highly of interest of SJ! cheers.
I have already done some expansions and will be doing more! Thanks for the nudge.
excellent news! And if you think your package can also be introduced in a tutorial paper, we could have some collaboration chances, which I welcome! my email is haghish@uio.no if that idea sounds interesting to you.
I imagine you've already seen it with your rcall GitHub search in the slides, but I have a package MLRtime which is also designed for running machine learning methods from Stata using
rcall
.ranger::ranger()
andgrf::causal_forest()
are implemented. I'm not currently planning to expand it, though. So feel free to nab any code you'd like if you want to incorporate those two.