Philip Hyunsu Cho

Thanks for raising the issue. Would you be interested in submitting a pull request to fix it?

> This is necessary since the default_left is changed from boolean to integer. How necessary is this? Was the default_left changed to improve performance?

Let's discuss whether `num_roots` should be kept or not. I think it is not documented and experimental and should be deprecated. Already, I decided not to support it when I...

@trivialfis Why is `XGDMatrixCreateFromCSC` deprecated? It is used by the R wrapper. @tqchen Got it

Just found out that ``TreeParam.max_depth`` is never set and never used: https://github.com/dmlc/xgboost/blob/3d81c48d3fa9a2a2ceb0d7388d3ac1436af7d8ee/include/xgboost/tree_model.h#L34 This field is different from the ``TrainParam.max_depth``, which is given at https://github.com/dmlc/xgboost/blob/42bf90eb8fbcda2fac494c43bc326c7ca1d254d0/src/tree/param.h#L29 Every occurrence of ``max_depth`` refers to...

@trivialfis > I'm only tracking the features that's already decided to be deprecated here, so that we can have a time frame for removing those code Got it. I'll file...

@pseudotensor I cannot reproduce the problem on my end. I ran this script ```python import numpy as np from cuml.ensemble import RandomForestClassifier from cuml.datasets import make_classification X, y = make_classification(n_samples=1000,...

The warning is thrown from https://github.com/rapidsai/cuml/blob/5b36ced22a70ee86c3153311efdc0cd7f0272101/python/cuml/ensemble/randomforestclassifier.pyx#L343-L344 This line of code is calling a static class method of `TreeliteModel` class, and somehow the class has become `None` at the time of...

The same issue still persists in 24.08 nightly Docker.