-
Notifications
You must be signed in to change notification settings - Fork 915
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[FEA] SUM_OF_SQUARES
aggregation for groupby in libcudf
#4667
Comments
Adding SUM_OF_SQUARES in hash groupby causes ptxas compiler segfault. (similar to crash faced by @davidwendt) cudf/cpp/src/groupby/hash/groupby.cu Line 74 in 6828e2c
|
If this does not happen in CUDA 11.x, then you can add it with an include guard to workaround the compiler issue. |
closes #4667 - Add hash groupby SUM_OF_SQUARES support - considering ptxas 10.2 crash, it's enabled for floating types only in unit tests. Authors: - Karthikeyan (@karthikeyann) Approvers: - Conor Hoekstra (@codereport) - Keith Kraus (@kkraus14) - Jake Hemstad (@jrhemstad) URL: #7362
Even with "closes" keyword in PR #7362 , this issue is not closed automatically! |
My understanding is that this only works if the PR's branch matches the default branch and our default branch is still 0.18. |
It would be great if libcudf groupby could provide a
SUM_OF_SQUARES
aggregation.Additional context
While Pandas does not have a
sum_of_squares
aggregation explicitly, it looks like thedask_cudf
implementation ofgroupby.std
could benefit from asum_of_squares
aggregation greatly (cc: @quasiben).The text was updated successfully, but these errors were encountered: