Nick Becker

### What is your question? Constructing a graph from a pandas edgelist requires using `source` and `destination` in cuGraph. In NetworkX, the arguments are `source` and `target`. Is there any...

Groupby Rank

2

Similar to https://github.com/dask/dask/issues/7095 , I'd also like to be able to use `rank` on groupby Series and DataFrame objects to create per-group rank columns. Today, I can do this in...

[BUG] Dask KNeighborsClassifier fit fails on cupy backed dask array

9

In the 2021-03-26 nightly (and perhaps earlier), `cuml.dask.neighbors.KNeighborsClassifier` fails during `fit` on dask array input if the arrays are backed by CuPy. ```python from distributed import Client, wait from dask_cuda...

[QST] Would using streams for async copying table / DataFrame columns speed up CPU/GPU transfers?

1

Many users and downstream applications rely on CPU/GPU transfers for interoperability, including in the zero code change interfaces for pandas, Polars, and Spark. Speed of CPU/GPU transfers can sometimes materially...

[FEA] Enable Polars GPU execution via global configuration (default options)

2

Polars provides a [Config](https://docs.pola.rs/py-polars/html/reference/config.html) object that can be used globally or as a context manager. Recent discussions have clarified that some users (and downstream libraries) would benefit from being able...