Fix bug with cache_vae_outputs and SDXL
#25
Merged
Conversation
This file contains 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
…cache loading code for all use cases.
| # The text encoder output may have been cached and included in the data_batch. If not, we calculate it here. | ||
| if "prompt_embeds" in data_batch: | ||
| prompt_embeds = data_batch["prompt_embeds"] | ||
| pooled_prompt_embeds = data_batch["pooled_prompt_embeds"] |
There was a problem hiding this comment.
I assume this will always exist when prompt_embeds does
There was a problem hiding this comment.
Correct. If "pooled_prompt_embeds" wasn't set we'd want to raise an Exception.
brandonrising
approved these changes
Aug 30, 2023
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Outdated suggestions cannot be applied.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.
Prior to this fix, when
cache_vae_outputswas enabled when training with SDXL, an exception would be raised becauseoriginal_size_hwandcrop_top_left_yx. This change caches those fields along with the vae_outputs.cache_vae_outputs=True,cache_text_encoder_outputs=Truecache_vae_outputs=False,cache_text_encoder_outputs=Falsecache_vae_outputs=True,cache_text_encoder_outputs=Truecache_vae_outputs=False,cache_text_encoder_outputs=False