Dante Gama Dessavre

Builds on top of #275

**Describe the bug** When using cuVS on an ARM machine, at the time of deallocating `cdef` classes like `cuvs.neighbors.ivf_flat.ivf_flat.Index` there is a `TypeError` that occurs when closing a Python session...

closes #6025

Capturing a task to simplify and improve our python input processing code, based on suggestions by @jakirkham like this one: https://github.com/rapidsai/cuml/pull/5954#pullrequestreview-2186917109

[BUG] cudf.pandas causes issues with sklearn Random Forest in cuML's pytests

7

While debugging the last few tests that left as bugs of the integration of cuML and cudf.pandas, found out that enabling cudf.pandas causes crashes on Scikit-learn Random Forest models. I'm...

NCCL 2.22.3.1 in `conda-forge` was not available for CUDA < 11.8 until [yesterday](https://github.com/conda-forge/nccl-feedstock/pull/111), which was reflected in cuML's CI by [failing all CUDA 11.4](https://github.com/rapidsai/cuml/actions/runs/10131183934) jobs until today. But RAFT's CUDA...

Make `get_param_names` a class method on single GPU estimators to match Scikit-learn closer

3

Small difference between our estimators and Scikit-learn is that `get_param_names` are a classmethod in sklearn, and not in ours. This can make a few corner cases fail for using our...

We need to update documentation of cuvs-bench docs being added/modified in #715 when containers are uploaded to NGC at release.