Repository navigation
Compute AnyFlow latent interpolation in FP32 - #3300
Merged
Merged
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
BF16 and FP16 latent interpolation can round nearby finite-difference points onto the same or unevenly spaced latent values, producing an incorrect MeanFlow derivative even for an exact velocity field.
Compute the sigma path in at least FP32 while preserving FP64 inputs and leaving the original latent/noise tensors unchanged. Regression tests cover low-precision inputs across three epsilon values and verify zero loss and gradient for an exact velocity solution. Place the regressions in the top-level test directory so normal unittest discovery runs them.
Validation:
.venv/bin/python -m unittest -v -f tests.test_anyflow_interpolation_precision tests.helpers.distillation.test_anyflow_distiller— 65 tests passed..venv/bin/python -m unittest discover -v -f tests/— 7,109 tests completed successfully, with 323 skips.