Skip to content
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

Temporarily allow for increase in RAFT nightly failures #2601

Closed
divyegala opened this issue Mar 11, 2025 · 0 comments · Fixed by #2607
Closed

Temporarily allow for increase in RAFT nightly failures #2601

divyegala opened this issue Mar 11, 2025 · 0 comments · Fixed by #2607

Comments

@divyegala
Copy link
Member

There have been certain spectral partitioning bugs that have popped up in CUDA 11.4 nightlies. While they are being investigated at the moment in #2600, we are approaching burndown quickly and need certain PRs to go in to unblock downstream work.

Specifically in cuML (mostly blocked on #2586):

  1. Follow-up PR to simplify (and make more robust) the parameter handling in cuml UMAP cpp. This would be removing duplicate parameters, and making graph_degree set to n_neighbors.
  2. Removing the now unnecessary slicing
  3. Some customer-facing testing for performance and memory usage.
    I4. mproving cuml UMAP's default behavior to better dispatch build algo based on data size and input type. This isn't really blocked on this upstream work, but would be nice to test in concert with it.
rapids-bot bot pushed a commit that referenced this issue Mar 11, 2025
Tracking issue: #2601

Temporarily allow for PRs to be unblocked while we face failures in nightlies.

Authors:
  - Divye Gala (https://github.com/divyegala)

Approvers:
  - Bradley Dice (https://github.com/bdice)

URL: #2602
@rapids-bot rapids-bot bot closed this as completed in b94de48 Mar 17, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

1 participant