search

RGF-team / rgf

Home repository for the Regularized Greedy Forest (RGF) library. It includes original implementation from the paper and multithreaded one written in C++, along with various language-specific wrappers.
378 stars 58 forks source link

New release #276

Closed StrikerRUS closed 5 years ago

StrikerRUS commented 5 years ago

I suppose it's time to release a new version with the support of warm start.

@fukatani Please release new Python version, and then @mlampros please upload to CRAN new R version.

fukatani commented 5 years ago

Thanks! I pushed tag 3.5.0.

mlampros commented 5 years ago

Should I now upload the new R version to CRAN based on the pushed 3.5.0 tag?

StrikerRUS commented 5 years ago

@fukatani Thank you!

@mlampros I think you should wait till @fukatani upload new release to PyPI. @fukatani Please ping @mlampros when it'll be done.

fukatani commented 5 years ago

Hi @mlampros pypi 3.5.0 version is uploaded

StrikerRUS commented 5 years ago

@fukatani Thanks a lot! @mlampros Please don't forget to merge recent changelogs into one under 1.0.6 version as we discussed here https://github.com/RGF-team/rgf/pull/273#discussion_r239602571

mlampros commented 5 years ago

Besides the init_model parameter and the save_model method (version 1.0.5.2) is there something else that should be added to the R package that I'm not aware of?

StrikerRUS commented 5 years ago

According to the history starting from Commits on Sep 28, 2018 | R package update to CRAN version 1.0.5 (#250) | a7c0a5b I don't see anything related to R-package that isn't listed in the changelog.

mlampros commented 5 years ago

thanks StrikerRUS. I'll proceed with the submission to CRAN and once accepted, I'll upload the new version to the rgf repository (pull request)

StrikerRUS commented 5 years ago

@mlampros Gracias!

mlampros commented 5 years ago

The R-package passed the CRAN tests. I opened pull request 277. Now it remains to add a few test cases for the new features ( ping to @jameslamb )

StrikerRUS commented 5 years ago

@fukatani Don't you know what is this?!

image

StrikerRUS commented 5 years ago

BTW, shields.io version is OK: PyPI Version

[![PyPI Version](https://img.shields.io/pypi/v/rgf_python.svg)]
fukatani commented 5 years ago

I have no idea :astonished:

jameslamb commented 5 years ago

@mlampros good morning! Yes we tried addressing #269 but found it was really hard to build RGF R package and run the tests. I think on previous PRs I'd tricked myself into thinking I had built and run it successfully, when in reality I never got past the skip_test_if_no_module checks.

So basically if your Python setup isn't correct the R tests just silently skip and can trick you into thinking the package is building fine.

I'm committed to coming back to it though!

StrikerRUS commented 5 years ago

@fukatani Don't you mind to replace the badge with working one?

fukatani commented 5 years ago

@StrikerRUS No, I don't.