flux-style-shaping / comfy /comfy_types.py
multimodalart's picture
Squashing commit
4450790 verified
raw
history blame
909 Bytes
import torch
from typing import Callable, Protocol, TypedDict, Optional, List
class UnetApplyFunction(Protocol):
"""Function signature protocol on comfy.model_base.BaseModel.apply_model"""
def __call__(self, x: torch.Tensor, t: torch.Tensor, **kwargs) -> torch.Tensor:
pass
class UnetApplyConds(TypedDict):
"""Optional conditions for unet apply function."""
c_concat: Optional[torch.Tensor]
c_crossattn: Optional[torch.Tensor]
control: Optional[torch.Tensor]
transformer_options: Optional[dict]
class UnetParams(TypedDict):
# Tensor of shape [B, C, H, W]
input: torch.Tensor
# Tensor of shape [B]
timestep: torch.Tensor
c: UnetApplyConds
# List of [0, 1], [0], [1], ...
# 0 means conditional, 1 means conditional unconditional
cond_or_uncond: List[int]
UnetWrapperFunction = Callable[[UnetApplyFunction, UnetParams], torch.Tensor]