{{ message }}
Repository navigation
Conversation
KronosPredictor moved the tokenizer/model to device but never called .eval(), so PyTorch's default training=True mode stayed active during inference. This kept dropout enabled inside F.scaled_dot_product_attention, which raises "NotImplementedError: scaled_dot_product_attention for MPS does not support dropout" on Apple Silicon (MPS backend). CUDA silently tolerates dropout in SDPA, which is why this went unnoticed there. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
KronosPredictor.__init__moves the tokenizer/model to device but never calls.eval(), leaving them in PyTorch's defaulttraining=Truemode during inference.F.scaled_dot_product_attention, which raisesNotImplementedError: scaled_dot_product_attention for MPS does not support dropouton Apple Silicon (MPS backend). CUDA silently tolerates dropout in SDPA, so this went unnoticed there.Fix
Call
.eval()on both the tokenizer and model right after moving them to device.Test plan
examples/prediction_example.pyon macOS (MPS backend) — previously crashed with the error above, now completes successfully end-to-end.🤖 Generated with Claude Code