-
Notifications
You must be signed in to change notification settings - Fork 4
Merge conda environment.yml files #30
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
Conversation
Consolidate two conda specification files into one environment.yml file. Notable changes are: - Staying on Python 3.11 instead of Python 3.12 - Drop torchvision/torchaudio as not used in this repo - Using `nvidia-dali-python` from conda-forge instead of `nvidia-dali-cuda120` from PyPI - Using nightly `kvikIO=25.04.00a` instead of https://github.com/TomAugspurger/kvikio-zarr-v3 - Using `xarray>=2025.3.0` containing pydata/xarray#10078
jakirkham
left a comment
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.
Thanks Wei! 🙏
Had a few questions/comments below
| dependencies: | ||
| - cupy~=13.3.0 | ||
| - dask-jobqueue~=0.9.0 | ||
| - rapidsai-nightly::kvikio>=25.04.00a |
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.
FWIW KvikIO 25.04 should be released later this week
ref: https://docs.rapids.ai/maintainers/
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.
Yes, I'm keeping a close eye on that release, and will update this once it's out!
| - jupyterlab~=4.3.5 | ||
| - nvidia-dali-python~=1.45.0 | ||
| - python=3.11 | ||
| - nvtx~=0.2.11 |
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.
Am curious about the context for the nvtx version constraint
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.
Probably ok to unpin, I just applied a ~= constraint for consistency in this environment.yml file
| - xarray>=2025.3.0 | ||
| - zarr~=3.0.3 | ||
| - pip: | ||
| - nvidia-nvcomp-cu12 |
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.
FWIW we have nvcomp packages on conda-forge
Though those are C++ only atm. If you need the Python bits, please let us know with a new issue on that feedstock
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.
Yep, that would be nice! Asked over at conda-forge/nvcomp-feedstock#21
Consolidate three conda specification files into one environment.yml file. Notable changes are:
nvidia-dali-pythonfrom conda-forge instead ofnvidia-dali-cuda120from PyPIkvikIO=25.04.00ainstead of https://github.com/TomAugspurger/kvikio-zarr-v3xarray>=2025.3.0containing Generalize lazy backend indexing a little more pydata/xarray#10078See also #28, part of #29