Update pipeline_stable_diffusion_instruct_pix2pix.py to use pytorch_xla and from single safetensor.

[chaowenguo]

What does this PR do?

Fixes # (issue)

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you read the contributor guideline?
• Did you read our philosophy doc (important for complex PRs)?
• Was this discussed/approved via a GitHub issue or the forum? Please add a link to it if that's the case.
• Did you make sure to update the documentation with your changes? Here are the
  documentation guidelines, and
  here are tips on formatting docstrings.
• Did you write any new necessary tests?

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.

@hlky

[hlky]

Thank you for the contribution @chaowenguo! I've added instruct-pix2pix to single_file_utils to complete from_single_file support.

Example

from diffusers import StableDiffusionInstructPix2PixPipeline

pipe = StableDiffusionInstructPix2PixPipeline.from_single_file("https://huggingface.co/timbrooks/instruct-pix2pix/blob/main/instruct-pix2pix-00-22000.safetensors")

[hlky]

cc @yiyixuxu We've had a few PRs for pytorch_xla support recently, should we create an issue to track pipelines missing pytorch_xla support and open it to the community? Edit: Created #10357

[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.

[chaowenguo]

@hlky could you actually merge the source？

[hlky]

Hi @chaowenguo we'll merge this after the single file changes get another review. Thank you for your patience.

[chaowenguo]

I also add support for pytorch_xla and from single safetensors to TextToVideoZeroPipeline. please review it @hlky

[chaowenguo]

I also add support for pytorch_xla to rerender_a_video.py. please review it @hlky

[hlky]

Thanks @chaowenguo, pytorch_xla changes for rerender_a_video look good.

Gentle ping @a-r-r-o-w @sayakpaul to take a look at the instruct-pix2pix from_single_file changes

[sayakpaul]

Thanks for your changes! Could we restrict the changes to InstructPix2Pix for this PR?

[chaowenguo]

I also add support for pytorch_xla to rerender_a_video.py. please review it @hlky. but @sayakpaul just want just the change to InstructPix2Pix. but now i am using render a video as a production, i need to install the render a video to 10 different compute node and want the change to be merge in main branch as soon as possible. should i open another pr to deal with render a video first？ @hlky

[sayakpaul]

Requesting you to send over a separate PR for render a video and keeping this PR dedicated to InstructPix2Pix.

[hlky]

@sayakpaul The PR was originally just XLA and from_single_file for instruct-pix2pix, I completed from_single_file support myself a797301 which is why I requested further review. The additional XLA changes were added while @chaowenguo has been waiting for review, as we are aiming to complete XLA support and the changes are standard for each pipeline I felt it was ok to add on to this PR.

[sayakpaul]

Oh okay. I didn't have the full background. In that case, I would like to review once again.

@chaowenguo sorry about the mistake on my part. Could you please reopen the PR? We will review and merge ASAP.

[chaowenguo]

well i already delete the repository. i can not fork the same repository twice. now i fork t

…

On Fri, Jan 3, 2025, 7:36 AM Sayak Paul ***@***.***> wrote: Oh okay. I didn't have the full background. In that case, I would like to review once again. @chaowenguo <https://github.com/chaowenguo> sorry about the mistake on my part. Could you please reopen the PR? We will review and merge ASAP. — Reply to this email directly, view it on GitHub <#10355 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BNIO5VUWNSUQE5YZZF2Y7BD2I2U6ZAVCNFSM6AAAAABUCUMYVGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKNRZGQYTSMBRGY> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[chaowenguo]

well i already delete the repository. i can not fork the same repository twice. now i fork to deal with render a video first. i need it for production.

…

On Fri, Jan 3, 2025, 7:49 AM chaowen guo ***@***.***> wrote: well i already delete the repository. i can not fork the same repository twice. now i fork t On Fri, Jan 3, 2025, 7:36 AM Sayak Paul ***@***.***> wrote: > Oh okay. I didn't have the full background. In that case, I would like to > review once again. > > @chaowenguo <https://github.com/chaowenguo> sorry about the mistake on > my part. Could you please reopen the PR? We will review and merge ASAP. > > — > Reply to this email directly, view it on GitHub > <#10355 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/BNIO5VUWNSUQE5YZZF2Y7BD2I2U6ZAVCNFSM6AAAAABUCUMYVGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKNRZGQYTSMBRGY> > . > You are receiving this because you were mentioned.Message ID: > ***@***.***> >

[sayakpaul]

Thanks and sincere apologies. @hlky I know this is inconvenient (sorry once again) but could we manually recover the other changes from this PR and make sure to get them in?