Get all dependencies from conda #21
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Also, specify the regular `cupy` package from PyPI rather than `cupy-cuda12x`, to avoid httomo dragging in the `cupy-cuda12x` dependency when doing `pip install httomo-backends` (which causes issues in the CUDA container used for the IRIS CI).
yousefmoazzam
force-pushed
the
cupy-from-conda
branch
2 times, most recently
from
a2c933c to
9d24ca8
Compare
Some notes on this may be useful: - It was found that specifying `astra-toolbox` as a conda dependency in addition to `cupy` was able to make the GPU detectable in the CUDA container used in the IRIS CI job. However, the first job was sometimes such that the GPU wasn't detected and needed to be rerun, but then the second job would have the GPU be detected, for some reason. - Specifying all required packages as conda dependencies somehow got rid of the need to rerun the job, where the first job seemingly was consistent in having the GPU be detected. The reason(s) for this behaviour are unknown, but this may be a good enough workaround to enable `httomo` in the following PR to install and use `httomo-backends` without issues in the IRIS CI: DiamondLightSource/httomo#532.
yousefmoazzam
force-pushed
the
cupy-from-conda
branch
from
9d24ca8 to
f707df0
Compare
yousefmoazzam
changed the title
cupy dependency from conda
|
See the message body of f707df0 for more details on why all dependencies were taken from conda |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The
cupy-cuda12xdependency is dragged in by httomo when doingpip install httomo-backendsin the IRIS CI workflow, and it causes issues with the GPU on IRIS nodes not being detectable, see the following failed workflow: https://github.com/DiamondLightSource/httomo/actions/runs/12694511591/job/35384845408.So, this PR is opting to get
cupyfrom conda (we already gettomopyfrom conda anyway), and see if this allows the httomo IRIS CI to pass in this PR where the methods database info have been removed from httomo and uses httomo-backends for that info instead: DiamondLightSource/httomo#532