Handle CPU devices without DataParallel - #8
Draft
fulopjoz wants to merge 2 commits into
Draft
Conversation
Owner
Author
|
Recreating as upstream PR against CDDLeiden/DrugEx |
Normalize string and torch.device inputs, preserve the CPU sentinel, and add portable regression coverage for CPU and mocked CUDA wrapping.
Route all nine generator and explorer wrapping sites through the shared device-aware helper so CPU execution never passes the -1 sentinel to DataParallel.
fulopjoz
marked this pull request as draft
July 29, 2026 14:34
fulopjoz
force-pushed
the
fix/cpu-sentinel-dataparallel
branch
from
July 29, 2026 14:34
04d4ffd to
b52b41a
Compare
fulopjoz
changed the title
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What changed
Root cause
CPU execution stores (-1,) as a sentinel GPU tuple. Several training paths passed that tuple to DataParallel, which is invalid and can crash even when CPU execution was explicitly requested.
Compatibility
CUDA paths still construct DataParallel with the requested device IDs. CPU paths use the module directly. Public model constructors now accept both cpu strings and torch.device objects.
Validation