Skip to content
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

instruct pix2pix pipeline: remove sigma scaling when computing classifier free guidance #7006

Merged
merged 2 commits into from
Mar 11, 2024
Merged

instruct pix2pix pipeline: remove sigma scaling when computing classifier free guidance #7006

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
a8e6548
remove sigma scaling when computing cfg
erliding Feb 18, 2024
7ed5f54
Merge branch 'main' into update-instructp2p-pipe
sayakpaul Mar 4, 2024
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
20 changes: 0 additions & 20 deletions src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_instruct_pix2pix.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -323,8 +323,6 @@ def __call__(
batch_size = prompt_embeds.shape[0]

device = self._execution_device
# check if scheduler is in sigmas space
scheduler_is_in_sigma_space = hasattr(self.scheduler, "sigmas")

# 2. Encode input prompt
prompt_embeds = self._encode_prompt(
Expand Down Expand Up @@ -411,15 +409,6 @@ def __call__(
return_dict=False,
)[0]

# Hack:
# For karras style schedulers the model does classifer free guidance using the
# predicted_original_sample instead of the noise_pred. So we need to compute the
# predicted_original_sample here if we are using a karras style scheduler.
if scheduler_is_in_sigma_space:
step_index = (self.scheduler.timesteps == t).nonzero()[0].item()
sigma = self.scheduler.sigmas[step_index]
noise_pred = latent_model_input - sigma * noise_pred

# perform guidance
if self.do_classifier_free_guidance:
noise_pred_text, noise_pred_image, noise_pred_uncond = noise_pred.chunk(3)
Expand All @@ -429,15 +418,6 @@ def __call__(
+ self.image_guidance_scale * (noise_pred_image - noise_pred_uncond)
)

# Hack:
# For karras style schedulers the model does classifer free guidance using the
# predicted_original_sample instead of the noise_pred. But the scheduler.step function
# expects the noise_pred and computes the predicted_original_sample internally. So we
# need to overwrite the noise_pred here such that the value of the computed
# predicted_original_sample is correct.
if scheduler_is_in_sigma_space:
noise_pred = (noise_pred - latents) / (-sigma)

# compute the previous noisy sample x_t -> x_t-1
latents = self.scheduler.step(noise_pred, t, latents, **extra_step_kwargs, return_dict=False)[0]

Expand Down
20 changes: 0 additions & 20 deletions src/diffusers/pipelines/stable_diffusion_xl/pipeline_stable_diffusion_xl_instruct_pix2pix.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -774,8 +774,6 @@ def __call__(
# of the Imagen paper: https://arxiv.org/pdf/2205.11487.pdf . `guidance_scale = 1`
# corresponds to doing no classifier free guidance.
do_classifier_free_guidance = guidance_scale > 1.0 and image_guidance_scale >= 1.0
# check if scheduler is in sigmas space
scheduler_is_in_sigma_space = hasattr(self.scheduler, "sigmas")

# 3. Encode input prompt
text_encoder_lora_scale = (
Expand Down Expand Up @@ -906,15 +904,6 @@ def __call__(
return_dict=False,
)[0]

# Hack:
# For karras style schedulers the model does classifer free guidance using the
# predicted_original_sample instead of the noise_pred. So we need to compute the
# predicted_original_sample here if we are using a karras style scheduler.
if scheduler_is_in_sigma_space:
step_index = (self.scheduler.timesteps == t).nonzero()[0].item()
sigma = self.scheduler.sigmas[step_index]
noise_pred = latent_model_input - sigma * noise_pred

# perform guidance
if do_classifier_free_guidance:
noise_pred_text, noise_pred_image, noise_pred_uncond = noise_pred.chunk(3)
Expand All @@ -928,15 +917,6 @@ def __call__(
# Based on 3.4. in https://arxiv.org/pdf/2305.08891.pdf
noise_pred = rescale_noise_cfg(noise_pred, noise_pred_text, guidance_rescale=guidance_rescale)

# Hack:
# For karras style schedulers the model does classifer free guidance using the
# predicted_original_sample instead of the noise_pred. But the scheduler.step function
# expects the noise_pred and computes the predicted_original_sample internally. So we
# need to overwrite the noise_pred here such that the value of the computed
# predicted_original_sample is correct.
if scheduler_is_in_sigma_space:
noise_pred = (noise_pred - latents) / (-sigma)

# compute the previous noisy sample x_t -> x_t-1
latents = self.scheduler.step(noise_pred, t, latents, **extra_step_kwargs, return_dict=False)[0]

Expand Down
Loading