Skip to content

Conversation

@lavinal712
Copy link
Contributor

@lavinal712 lavinal712 commented Mar 17, 2025

What does this PR do?

I found that the scripts/convert_vae_pt_to_diffusers.py script doesn't handle cases well where the encoder and decoder have attention, such as latent-diffusion/configs/autoencoder/autoencoder_kl_16x16x16.yaml at main · CompVis/latent-diffusion. Therefore, I made corresponding changes to support it.

This PR is bug fix for #11077

@sayakpaul

@HuggingFaceDocBuilderDev

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

@hlky
Copy link
Contributor

hlky commented Mar 17, 2025

@bot /style

@github-actions
Copy link
Contributor

Style fixes have been applied. View the workflow run here.

@lavinal712 lavinal712 changed the title Fix convert_vae_pt_to_diffusers bug [BUG] Fix convert_vae_pt_to_diffusers bug Mar 20, 2025
@lavinal712
Copy link
Contributor Author

@sayakpaul @DN6 Please take a look at this PR when you have a chance. Thanks.

@sayakpaul sayakpaul requested a review from DN6 April 8, 2025 06:07
Copy link
Contributor

@hlky hlky left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Apologies for the delay @lavinal712. LGTM, thanks!

@hlky
Copy link
Contributor

hlky commented Apr 10, 2025

Failing tests are unrelated.

@hlky hlky merged commit e121d0e into huggingface:main Apr 10, 2025
11 of 12 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants