bonsai
bonsai copied to clipboard
aorsf classification and regression engines
Adds support for aorsf
(#73).
I am uploading aorsf
version 0.1.4 to CRAN and I don't expect any problems, but I totally understand if you'd prefer we hold off on this PR until the new version is settled on CRAN.
bonsai
is an awesome package. Thanks for creating it!
Wonderful, this is a great start!
I just approved running GHA—the failures with snap changes are unrelated to this PR and you are welcome to ignore.
Could you add some tests along the lines of those you put together for tidymodels/censored#211? Specifically interested in whether we can reproduce the same results using the aorsf interface and parsnip interface.
Another next step here would be to add sections on the newly supported modes to parsnip's documentation. You put together the basics in tidymodels/parsnip#828; a follow-up PR would add # Translation from parsnip to the original package (mode)
sections for mode
s regression
and classification
to man/rmd/rand_forest_aorsf.Rmd
and clarify any language in that documentation suggesting that the engine can only be used for survival analysis.
If you're interesting in testing this against our CI, I've just approved workflows. They will fail as-is, but merging upstream should resolve any issues not due to this PR. :)
I'm assuming I should stay hands-off until a companion PR is made to parsnip, but feel free to holler if you'd appreciate support.
Thank you! I have updated the parsnip docs and opened https://github.com/tidymodels/parsnip/pull/1120
I merged the upstream main branch in and all tests passed locally, with 2 warnings.
Sorry, I know I have been pushing things and opening PRs either during weekends or late on Fridays. My work schedule is a little off and I don't have any expectation whatsoever to hear back outside of normal working hours.
Thanks for all of your work here! Really excited about these additions. I think these PRs are in a good spot to head to main
and will look towards a CRAN release of bonsai once the aorsf version that follows up on 0.1.4 is on CRAN.
Awesome! Thank you so much for working with me on this.
👋, @simonpcouch, just wanted to let you know aorsf
version 0.1.5 is on its way to CRAN. It should be stable in the next few days.