Skip to content

Conversation

@topepo
Copy link
Contributor

@topepo topepo commented Nov 22, 2025

Closes #28

See dmlc/xgboost#11430

There is no version requirement for xgboost in the description file. Once the new version of xgboost is on CRAN, we can revert c2f87f9 and send xrf to CRAN.

Main changes here:

  • xrf requires R >= 4.3 due to xgboost requirement. 😢
  • Use xgb.train() instead of xgboost().
  • Helper function to populate the xgb arguments.
  • Archive of previous fit to verify backwards compatibility for previously fit models.
  • Small changes to summary.xrf().

@topepo topepo marked this pull request as ready for review November 22, 2025 09:47
@topepo
Copy link
Contributor Author

topepo commented Dec 2, 2025

I think XGBoost has sent their new version to CRAN. Would you like me to do anything different for either of these comments?

@holub008
Copy link
Owner

holub008 commented Dec 3, 2025

Thank you for the explanations, we are good! I got the failing build alert from CRAN this afternoon, so let's get it out the door. If you add in the xgb version requirement > 3.1.2.1 and cleanup the dev xgb install, I'll merge this in, do some final testing, and publish to CRAN.

I'm so appreciative of your work @topepo !

@topepo
Copy link
Contributor Author

topepo commented Dec 4, 2025

I made those updates. Good luck 🤞

@holub008 holub008 merged commit 0a46938 into holub008:master Dec 4, 2025
9 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

New xgboost version

2 participants