Update pytorch to 2.1.2 + fix CI/linux CPU pytorch #2429
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Outdated suggestions cannot be applied.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.
I guess one benefit to taking so long for a major release is that we can stay up to date with pytorch and not feel bad about it.
Also, I realized after looking at https://pytorch.org/get-started/locally/ that the reason our CI was installing the cuda version of pytorch was because for some odd reason, linux uses the cuda version by default and the cpu is the one you have to specify the extra index url for. So I just was explicit for both and tada now it works and it takes 1/4 of the time to run our CI.
This also means that linux users were previously unable to install the cpu version of torch. Oops. What an odd choice...