fix: force eval mode for DAT models to resolve BatchNorm training mode compatibility#242
Open
paulthedev wants to merge 2 commits into
Open
fix: force eval mode for DAT models to resolve BatchNorm training mode compatibility#242paulthedev wants to merge 2 commits into
paulthedev wants to merge 2 commits into
Conversation
Author
|
Hi @TNTwise would love your support on this. AI usage disclosure: yes |
…e compatibility Models trained with neosr have BatchNorm layers that require batch_size > 1 when in training mode. This fix sets models to eval mode before inference, allowing neosr-trained DAT models (e.g., UltraSharpV2) to load correctly.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Problem
Models trained with neosr (e.g., 4x-UltraSharpV2) have BatchNorm layers in training mode that require
batch_size > 1. When loaded with batch_size=1 for testing, this causes:ValueError: Expected more than 1 value per channel when training, got input size torch.Size(1, 22, 1, 1)
Solution
Force
model.eval()after loading models via spandrel. This is a defensive fix that ensures any model with BatchNorm layers in training mode gets set to eval mode.Scope
This fix applies to PyTorch backend only.
Testing
Changes
backend/src/pytorch/UpscaleModelWrapper.py: Addmodel.eval()after spandrel model load