Skip to content

Reproducible images by supplying latents to pipeline #247

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 8 commits into from
Aug 25, 2022
Merged

Reproducible images by supplying latents to pipeline #247

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
16c50f4
Accept latents as input for StableDiffusionPipeline.
pcuenca Aug 25, 2022
ef43f1a
Notebook to demonstrate reusable seeds (latents).
pcuenca Aug 25, 2022
b189cdd
More accurate type annotation
pcuenca Aug 25, 2022
fb6a05c
Review comments: move to device, raise instead of assert.
pcuenca Aug 25, 2022
71127e2
Actually commit the test notebook.
pcuenca Aug 25, 2022
d2d2c4a
Adapt notebook to Colab.
pcuenca Aug 25, 2022
fa3dfac
Update examples readme.
pcuenca Aug 25, 2022
7857678
Move notebook to personal repo.
pcuenca Aug 25, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 5 additions & 1 deletion examples/inference/readme.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,4 +47,8 @@ with autocast("cuda"):

images[0].save("fantasy_landscape.png")
```
You can also run this example on colab [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/patil-suraj/Notebooks/blob/master/image_2_image_using_diffusers.ipynb)
You can also run this example on colab [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/patil-suraj/Notebooks/blob/master/image_2_image_using_diffusers.ipynb)

## Tweak prompts reusing seeds and latents

You can generate your own latents to reproduce results, or tweak your prompt on a specific result you liked. [This notebook](stable-diffusion-seeds.ipynb) shows how to do it step by step. You can also run it in Google Colab [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/pcuenca/diffusers-examples/blob/main/notebooks/stable-diffusion-seeds.ipynb).
19 changes: 13 additions & 6 deletions src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,6 +46,7 @@ def __call__(
guidance_scale: Optional[float] = 7.5,
eta: Optional[float] = 0.0,
generator: Optional[torch.Generator] = None,
latents: Optional[torch.FloatTensor] = None,
output_type: Optional[str] = "pil",
**kwargs,
):
Expand Down Expand Up @@ -98,12 +99,18 @@ def __call__(
# to avoid doing two forward passes
text_embeddings = torch.cat([uncond_embeddings, text_embeddings])

# get the intial random noise
latents = torch.randn(
(batch_size, self.unet.in_channels, height // 8, width // 8),
generator=generator,
device=self.device,
)
# get the initial random noise unless the user supplied it
latents_shape = (batch_size, self.unet.in_channels, height // 8, width // 8)
if latents is None:
latents = torch.randn(
latents_shape,
generator=generator,
device=self.device,
)
else:
if latents.shape != latents_shape:
raise ValueError(f"Unexpected latents shape, got {latents.shape}, expected {latents_shape}")
latents = latents.to(self.device)

# set timesteps
accepts_offset = "offset" in set(inspect.signature(self.scheduler.set_timesteps).parameters.keys())
Expand Down