Bradley Dice
Bradley Dice
I met with some folks from CCCL and cuda-python earlier this week to discuss memory resources. The new CCCL features in https://github.com/NVIDIA/cccl/pull/2824 will change our strategy for owning memory resources....
Sure, that is fine.
I opened two PRs, #1625 and #1626, that should help with this.
Hi @Borda, thanks for this. However, we have been keeping isort because ruff does not support Cython code but isort does. See this discussion for more info: https://github.com/rapidsai/cudf/issues/14882#issuecomment-1910887340
You’re right. I forgot the rest of that discussion. @vyasr If you’re in support, I am fine with it.
@Borda I'm going to trigger tests to run -- we can hold off on merging the upstream for now, unless there are conflicts.
/ok to test
/ok to test
/ok to test