Skip to content

add output type #13091

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

Open
asomoza wants to merge 1 commit into
base: modular-wan-more-pipelines
Choose a base branch
from
+12 −2
Open

add output type #13091

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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: 6 additions & 0 deletions src/diffusers/modular_pipelines/mellon_node_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -156,6 +156,12 @@ def _name_to_label(name: str) -> str:
"display": "slider",
"required_block_params": ["layers"],
},
"output_type": {
"label": "Output Type",
"type": "dropdown",
"default": "np",
"options": ["np", "pil", "pt"],
},
# ControlNet
"controlnet_conditioning_scale": {
"label": "Controlnet Conditioning Scale",
Expand Down
8 changes: 6 additions & 2 deletions src/diffusers/modular_pipelines/wan/decoders.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -56,7 +56,10 @@ def inputs(self) -> List[Tuple[str, Any]]:
required=True,
type_hint=torch.Tensor,
description="The denoised latents from the denoising step",
)
),
InputParam(
"output_type", default="np", type_hint=str, description="The output type of the decoded videos"
),
]

@property
Expand Down Expand Up @@ -87,7 +90,8 @@ def __call__(self, components, state: PipelineState) -> PipelineState:
latents = latents.to(vae_dtype)
block_state.videos = components.vae.decode(latents, return_dict=False)[0]

block_state.videos = components.video_processor.postprocess_video(block_state.videos, output_type="np")
output_type = getattr(block_state, "output_type", "np")
block_state.videos = components.video_processor.postprocess_video(block_state.videos, output_type=output_type)

self.set_block_state(state, block_state)

Expand Down
Loading