DetailerPipe -> BasicPipe¶
Documentation¶
- Class name:
DetailerPipeToBasicPipe
- Category:
ImpactPack/Pipe
- Output node:
False
This node transforms a detailer pipe configuration into two basic pipe configurations, one for the base model and another for the refiner model. It abstracts the complexity of handling different components within a detailer pipe and simplifies it into more manageable basic pipe structures.
Input types¶
Required¶
detailer_pipe
- The detailer pipe input is a comprehensive configuration that includes models, clips, VAEs, and conditioning information for both the base and refiner setups. It serves as the source from which the basic pipe configurations are derived.
- Comfy dtype:
DETAILER_PIPE
- Python dtype:
Tuple[torch.nn.Module, torch.nn.Module, torch.nn.Module, str, str, NoneType, NoneType, NoneType, NoneType, NoneType, torch.nn.Module, torch.nn.Module, str, str]
Output types¶
base_basic_pipe
- Comfy dtype:
BASIC_PIPE
- The base basic pipe configuration, derived from the input detailer pipe, containing the model, clip, VAE, and conditioning information for the base setup.
- Python dtype:
Tuple[torch.nn.Module, torch.nn.Module, torch.nn.Module, str, str]
- Comfy dtype:
refiner_basic_pipe
- Comfy dtype:
BASIC_PIPE
- The refiner basic pipe configuration, also derived from the input detailer pipe, but focusing on the refiner model's setup including its model, clip, VAE, and conditioning information.
- Python dtype:
Tuple[torch.nn.Module, torch.nn.Module, torch.nn.Module, str, str]
- Comfy dtype:
Usage tips¶
- Infra type:
CPU
- Common nodes:
Source code¶
class DetailerPipeToBasicPipe:
@classmethod
def INPUT_TYPES(s):
return {"required": {"detailer_pipe": ("DETAILER_PIPE",), }}
RETURN_TYPES = ("BASIC_PIPE", "BASIC_PIPE")
RETURN_NAMES = ("base_basic_pipe", "refiner_basic_pipe")
FUNCTION = "doit"
CATEGORY = "ImpactPack/Pipe"
def doit(self, detailer_pipe):
model, clip, vae, positive, negative, _, _, _, _, _, refiner_model, refiner_clip, refiner_positive, refiner_negative = detailer_pipe
pipe = model, clip, vae, positive, negative
refiner_pipe = refiner_model, refiner_clip, vae, refiner_positive, refiner_negative
return (pipe, refiner_pipe)