Skip to content

AudioDiffusionPipeline - fix encode method after config changes #3114

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
patrickvonplaten merged 3 commits into from
Apr 27, 2023
Merged

AudioDiffusionPipeline - fix encode method after config changes #3114

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
e534900
config fixes
teticio Apr 15, 2023
3f24e15
deprecate get_input_dims
teticio Apr 22, 2023
ae92c9b
Merge branch 'huggingface:main' into main
teticio Apr 22, 2023
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
19 changes: 1 addition & 18 deletions src/diffusers/pipelines/audio_diffusion/pipeline_audio_diffusion.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -51,21 +51,6 @@ def __init__(
super().__init__()
self.register_modules(unet=unet, scheduler=scheduler, mel=mel, vqvae=vqvae)

def get_input_dims(self) -> Tuple:
"""Returns dimension of input image

Returns:
`Tuple`: (height, width)
"""
input_module = self.vqvae if self.vqvae is not None else self.unet
# For backwards compatibility
sample_size = (
(input_module.config.sample_size, input_module.config.sample_size)
if type(input_module.config.sample_size) == int
else input_module.config.sample_size
)
return sample_size

def get_default_steps(self) -> int:
"""Returns default number of steps recommended for inference

Expand Down Expand Up @@ -123,8 +108,6 @@ def __call__(
# For backwards compatibility
if type(self.unet.config.sample_size) == int:
self.unet.config.sample_size = (self.unet.config.sample_size, self.unet.config.sample_size)
input_dims = self.get_input_dims()
self.mel.set_resolution(x_res=input_dims[1], y_res=input_dims[0])
if noise is None:
noise = randn_tensor(
(
Expand Down Expand Up @@ -234,7 +217,7 @@ def encode(self, images: List[Image.Image], steps: int = 50) -> np.ndarray:
sample = torch.Tensor(sample).to(self.device)

for t in self.progress_bar(torch.flip(self.scheduler.timesteps, (0,))):
prev_timestep = t - self.scheduler.num_train_timesteps // self.scheduler.num_inference_steps
prev_timestep = t - self.scheduler.config.num_train_timesteps // self.scheduler.num_inference_steps
alpha_prod_t = self.scheduler.alphas_cumprod[t]
alpha_prod_t_prev = (
self.scheduler.alphas_cumprod[prev_timestep]
Expand Down