Bradley Dice
Bradley Dice
@gpucibot merge
This should be worked on. Without a comparator for tables/columns and scalars, lifetime management of a column derived from the scalar gets very awkward. https://github.com/rapidsai/cudf/blob/ee15d1c10b51169abe26b1fa052ae568296696d5/cpp/src/lists/contains.cu#L251-L268 I commented here: https://github.com/rapidsai/cudf/pull/10548/files#r946174668 >...
> LGTM! Does the CI setup test both configurations? Otherwise we might miss the failure detection breaking silently No, we don't have a way to test the configuration with S3...
@gpucibot merge
@Shafi2016 What kind of GPU do you have? I see `GeForce ...` in the output of `nvidia-smi`. It sounds like a previous issue concluded that Pascal GPUs (which would include...
This is pretty cool! Some parts feel a bit overengineered and break from some pytest writing conventions specifically through the use of "assert" injection. From the test logic alone and...
@vyasr @galipremsagar I tagged you both for review because of your experience with the copyright scripts and pre-commit. Thanks!
> Is this good to go? I’d like a final approval from @vyasr to make sure there wasn’t anything else he wanted to address. Then please feel free to merge!
A small consideration that might tip the balance: using underscores makes it map more directly to valid Python identifiers, e.g. ```python # Using underscores in the keys: frame_count = metadata["frame_count"]...
@michaelmaalouly This error is probably specific to your data, and might be an issue with numerical precision (those have come up a few times in the Voronoi code and are...