Adam Li

FYI: @glemaitre and @d-kleine https://github.com/scikit-learn/scikit-learn/pull/28552

Adding to the meson-related discussion here, I was recently playing with the editable installs (https://github.com/mesonbuild/meson-python/issues/47) with meson and meson-python and it is nice. Some thoughts: I've also found using meson...

Some of the CI checks were skipped. Unsure if that is due to me

> Nice PR, thanks @adam2392 ! Thanks for the review! I think I addressed most of your comments and then left a few questions for follow-up

Thanks @OmarManzoor for the quick review :)

@tylerjereddy just to clarify: Is the issue that the `oob` predictions are occurring without joblib parallelization? Do you have a link to your fix, and are interested in submitting a...

@thomasjpfan just a gentle ping to see if this is something you might be interested and have time for review now?

Hi @thomasjpfan just checking in to see if you still have the bandwidth to check this PR? This will consolidate towards the support of missing-values in decision trees throughout the...

So the issue with @TomDLT's solution is that this is super slow if we want a separate radius per sample. For example, I only want to query the radius_neighbors of...

I think in terms of sequential experiments to run: 1. `RandomForestClassifier` in scikit-learn vs `RandomForestClassifier` in scikit-tree in just `n_samples` vs time to fit with n_jobs =1 vs n_jobs =...