James Lamb

Thanks for using LightGBM! Without a [minimal, reproducible example](https://stackoverflow.com/help/minimal-reproducible-example), maintainers cannot help much with what you're asking. You haven't told us what you mean by "very slow", and you haven't...

@zyxue thanks for coming back to this pull request! As you work on it, please do not rebase + force push. Use merge commits instead (e.g. `git merge master`). Overwriting...

> Is https://github.com/microsoft/LightGBM/blob/master/include/LightGBM/utils/json11.h expected to be the same as https://raw.githubusercontent.com/dropbox/json11/master/json11.hpp ? That is where the `json11` code in LightGBM is originally from, but we consider it "vendored in", meaning that...

As I said, please wait until another maintainer offers their opinion. While you wait on that, you can try working through the other suggestions like https://github.com/microsoft/LightGBM/pull/4802#discussion_r786419637.

Due to a lack of activity on this pull request, we've decided to move forward with a separate PR for this feature in #5424, so I'm going to close this....

> Anything new? Similar to the comments I just left in https://github.com/microsoft/LightGBM/issues/5214#issuecomment-1127108914...without a reproducible example, or at least sample code + full logs, there isn't much that maintainers can do...

@jmoralez ah ok, the point about `predict()` makes sense. But to be honest, I don't understand how the new approach using `Sys.setenv(LGBM_DEFAULT_NUM_THREADS = "2")` at the top of `testthat.R` is...

Thanks very much for taking the time to write this up and share it @AlbertoEAF ! I'm interested in considering it, especially if it can be done in a way...

Hi @sofiavlachou28 , thanks very much for using LightGBM! I'd be happy to help you, but we need a little more information. 1. What version of `lightgbm` are you using...

Thanks for providing the dataset @sofiavlachou28 . Looking at the data, I have some observations and a suggestion. The `label` for a learning-to-rank problem is expected to be a "relevance...