Added a --use_cuda_bfloat16 option to the boltz predict command. #264
Conversation
This file contains hidden or 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
… saves memory allowing prediction of structures with 40% more residues without running out of GPU memory on Linux or falling back to CPU computation on Windows. I tested on a half-dozen PDB structures and did not see any loss of accuracy.
|
There is a spelling error, "use_cuda_bloat16: bool = False" |
|
Thanks for pointing out the typo. I have fixed it in this pull request. I fixed it a week ago but forgot to update this fork because I decided to make another boltz fork specifically for the version of Boltz to use in ChimeraX and did my testing and fix on that other fork and forgot to fix this one. |
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Outdated suggestions cannot be applied.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.
Fixes #263. This allows predicting structures with 40% more residues without running out of GPU memory. I tested on a half-dozen PDB structures and did not see any loss of accuracy.