mlc-ai / mlc-ai/web-stable-diffusion

Exporting Stable diffusion 's VAE model Failed

Open
#25 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Jupyter Notebook
Stars
3.7k
Forks
238
PR merge metrics
No merged PRs in 30d

Description

I'm using torch 2.1.0.dev20230425+cpu and diffuser 0.16  to build stable diffusion v1_5, But I got the following error:
assert ( AssertionError: Unsupported function type scaled_dot_product_attention

I print symbolic traced graph , and I found vae module is using torch.c.scaled_dot_procuct_attention op

scaled_dot_product_attention = torch._C._nn.scaled_dot_product_attention(permute, permute_1, permute_2, dropout_p = 0.0, is_causal = False); permute_1 = permute_2 = None
I can attach the code which can make this happend
output.zip

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the attached output.zip and reproduce the export using torch 2.1.0.dev20230425+cpu and diffuser 0.16. Inspect the symbolic traced VAE graph where scaled_dot_product_attention appears, then determine how the exporter should handle this unsupported function type. Done means exporting the Stable Diffusion v1_5 VAE completes without the assertion.

Written by the indexing model from the issue text.

Assessment

Tech stack
pytorch
Domain
compilers, machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
30/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.