search

microsoft / LightGBM

A fast, distributed, high performance gradient boosting (GBT, GBDT, GBRT, GBM or MART) framework based on decision tree algorithms, used for ranking, classification and many other machine learning tasks.
https://lightgbm.readthedocs.io/en/latest/
MIT License
16.61k stars 3.83k forks source link

Update macOS install instructions #782

Closed fulldecent closed 6 years ago

fulldecent commented 7 years ago

This regards https://github.com/Microsoft/LightGBM/wiki/Installation-Guide

I cannot edit the file, so creating an issue.

Just want to note that the normal brew install gcc worked for me. This removes about two hours from the installation process for macOS as currently documented in this file.

guolinke commented 7 years ago

@wxchan @henry0312 Does using brew install gcc have openmp support ?

henry0312 commented 7 years ago

Yes, it have OpenMP, but I don't know whether it works fine. How can I check simply? https://computing.llnl.gov/tutorials/openMP/samples/C/omp_hello.c works at least.

guolinke commented 7 years ago

so why we need the --without-multilib ?

henry0312 commented 7 years ago

I don't know the details, but there was a bug in the previous gcc. cf. https://github.com/Homebrew/homebrew-core/commit/bc4ce49717a77a0debdb2c3b8b2542c87ab58daa.

And I don't know whether the bug are solved.

guolinke commented 7 years ago

@wxchan can you help to confirm this ?

thierrygosselin commented 7 years ago

Hi LightGBM team, I'm a new user. I've been working with XGBoost for imputations and starting to try LightGBM.

I have better doc on how to upgrade GCC and CLANG with OpenMP support for macOS. Both approaches are homebrew and macport free. Additionally they take less than 5 min to do ... yep.

Can someone point me to the link where I could edit the latest doc on LightGBM ?

Best Thierry

Laurae2 commented 7 years ago

https://github.com/Microsoft/LightGBM/blob/master/docs/Installation-Guide.rst

and if there are extra requirements for R: https://github.com/Microsoft/LightGBM/blob/master/R-package/README.md