Update Dockerfile to enable CUDA pytorch #760
Merged
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Outdated suggestions cannot be applied.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.
With the current Docker build,
conda
defaults to installingpytorch
in CPU mode despite the use of pytorch and nvidia channels. This may be a change in howconda
handles package relationships. The below command adjustment passes in the specific version numbers, which ensures thatconda
recognizes the use ofpytorch-cuda
as related, and properly installs the CUDA version ofpytorch
.This can be validated after-the-fact with starting the container and running:
The current image returns an error, while after this PR, returns "True".
Note: This also updates CUDA in pytorch to 12.1 to better align to the CUDA docker image being used as the base.