fix: initialize torch.compile per thread - #1495
Merged
Merged
Conversation
torch._dynamo.config overrides are thread-local. The existing call in checkpointing_util runs in the main thread and is invisible to the training thread spawned by the UI. This caused compiled optimizers (e.g. AdamW_adv with compiled_optimizer=True) to hit the default recompile_limit of 8 and abort with FailOnRecompileLimitHit when training models with more than 8 distinct parameter shapes. Fix: call init_compile() from GenericTrainer.__init__, which runs in whichever thread/process owns training (UI thread, CLI main thread, or torch.multiprocessing.spawn subprocess for multi-GPU). Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Merged
3 tasks
BitcrushedHeart
pushed a commit
to BitcrushedHeart/OneTrainer
that referenced
this pull request
Jun 20, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
torch._dynamo.config overrides are thread-local. The existing call in checkpointing_util runs in the main thread and is invisible to the training thread spawned by the UI. This caused compiled optimizers (e.g. AdamW_adv with compiled_optimizer=True) to hit the default recompile_limit of 8 and abort with FailOnRecompileLimitHit when training models with more than 8 distinct parameter shapes.
Fix: call init_compile() from GenericTrainer.init, which runs in whichever thread/process owns training (UI thread, CLI main thread, or torch.multiprocessing.spawn subprocess for multi-GPU).