-
Notifications
You must be signed in to change notification settings - Fork 197
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
Build and test with CUDA 12.5.1 #2378
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
These places are technically not "missed", they are accurate ... but I do think we should update them to CUDA 12.5, to help encourage the use of newer CUDA versions.
Lines 283 to 285 in ab5e128
```bash | |
# for CUDA 12.0 | |
mamba install -c rapidsai -c conda-forge -c nvidia raft-dask pylibraft cuda-version=12.0 |
Lines 291 to 293 in ab5e128
```bash | |
# for CUDA 12.0 | |
mamba install -c rapidsai -c conda-forge -c nvidia libraft libraft-headers cuda-version=12.0 |
Do we want to update these?
Edit: Noticed this line in |
You can safely change that. That's setting a default for a build argument that's always overridden. I'm going to propose removing that default (precisely for this reason of not confusing people) once |
After discussing offline, we decided to punt on RAFT ANN for now. This is in part as we understand this to benchmark related and it may involve more than updating the versions (like rerunning the benchmark and looking for changes). Have raised an issue for follow up: #2383 We can circle back once we know better how to proceed |
/merge |
Thanks Kyle and James! 🙏 |
This PR updates the latest CUDA build/test version 12.2.2 to 12.5.1.
Contributes to rapidsai/build-planning#73