Dante Gama Dessavre

Seen in multiple CI runs, there seems to be a very minor mismatch in a ANN test: ```python =========================== short test summary info ============================ FAILED test_nearest_neighbors.py::test_ivfpq_pred[8-512-4000-False-4-32-8] - assert + where...

Fix `cudf.pandas` failure on `test_convert_matrix_order_cuml_array`

3

Error came from the fact that pandas and cudf convert to numpy by default with different order. Towards fixing #5876

This failure was just a testing failure, expectint identical pointers of actual dataframes, as opposed to wrapped objects. Contributes to fixing #5876 cc @betatim

cc @vyasr

Closes #4477 Adds the capability for all estimators to accept any dtype by converting the inputs if needed to. Currently, for most estimators, this means converting to float32. Todo: -...

Currently we test accuracy of results in general, but there are cases like #5946 where we return correct values with the wrong dtype, breaking interoperability. This task encompasses adding a...

Avoid extra memory copy when using cp.concatenate in cuml.dask kmeans

1

Partial solution for #5936 Issue was that concatenating when having a single array per worker was causing a memory copy (not sure if always, but often enough). This PR avoids...

Closes #5918 Correctly gathers the labels of all workers together for the `labels_` attribute of the cuml.dask KMeans estimator.

cuML has both hard and soft dependencies, but given the complex structure of imports in the library, it is easy for something to make a soft dependency hard. This task...

Adding conda environments and a few extra files needed for the upcoming cuvs-bench. cc @tarang-jain @cjnolet could use a look at the correct version of packages like hnsw