New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Roadmap] 1.3.2 Patch release #6571
Comments
@hcho3 All merged now. |
@trivialfis Great! Why don't you give a shot at making the patch release? Use #6543 as an example. |
Thanks for the suggestion, let me do that tomorrow. |
@hcho3 @hetong007 I created a release page. Next would be submitting packages to pip and cran. |
@trivialfis I uploaded this branch onto R-hub, and it still reports error: https://builder.r-hub.io/status/xgboost_1.3.2.1.tar.gz-f47cc590f1434dddb4c4b89cd02703dc This error is from R-hub with platform |
Em .. failing to install |
Looked a little bit closer:
|
I don't know if there's a way to retrieve that log. |
Got it passed: https://builder.r-hub.io/status/xgboost_1.3.2.1.tar.gz-72aa615847ba65333ecfea61c676b7b4 Will open a PR. |
I patched this PR and submitted it to R-hub again, but got an error in installation: https://artifacts.r-hub.io/xgboost_1.3.2.1.tar.gz-5c8a82e1e66188afea79817c398575fd/xgboost.Rcheck/00install.out |
|
My configuration on r-hub is:
I can submit again to see if this is a false alarm, will update this thread if ends up with a different result. |
I hope that we don't have to bisect it on R Hub .... |
I tried the old 1.2 release, which is also failing due to type check (removed in 1.3): https://artifacts.r-hub.io/xgboost_1.2.0.1.tar.gz-1fa6b58859caf51810b85692f5e870c6/xgboost.Rcheck/00install.out |
Seems the current CRAN error is not the same as what we have on r-hub: https://www.r-project.org/nosvn/R.check/r-patched-solaris-x86/xgboost-00install.html |
@hetong007 That was fixed in #6578 |
At latest 1.3 branch, all compiler related workarounds/fixes are applied. But the linker issue you found #6571 (comment) is still there, and I'm not sure how to remove it. The |
@hetong007 Can we just submit the package? |
Well if a fix is not easily accessible, I can try to submit. That's at least an action trying to patch. |
@hetong007 Thanks! Please keep me in the loop if there's any insight on the linker error. I wonder if it's reproducible on CRAN. Since of the 2 Solaris configurations on R Hub, the one that without oracle developer studio is passing. |
@hcho3 All wheels are uploaded to pypi. Many thanks for the help! |
We are looking to release a patch release 1.3.2.
This issue will remain open until January 11, 2021.
@dmlc/xgboost-committer
Pull requests:
data
toX
inpredict_proba
. #6555 Fix compatibility with newer scikit-learn.best_ntree_limit
for multi-class. #6569 Fix wrongbest_ntree_limit
in multi-class.best_ntree_limit
for dart and gblinear. #6579 Fixbest_ntree_limit
for linear and dart.The text was updated successfully, but these errors were encountered: