Set CLIP LoRA Hook 🎭🅐🅓

Documentation

Class name: ADE_AttachLoraHookToCLIP
Category: Animate Diff 🎭🅐🅓/conditioning
Output node: False

This node is designed to attach a LoRA hook to a CLIP model, enhancing its functionality by integrating additional computational layers or modifications. It serves as a customization point for the CLIP model, allowing for tailored adjustments that can influence the model's behavior or outputs.

Input types

Required

clip
- The CLIP model to which the LoRA hook will be attached. This parameter is crucial as it determines the base model that will be modified.
- Comfy dtype: CLIP
- Python dtype: CLIP
lora_hook
- The LoRA hook to be attached to the CLIP model. This hook represents the modifications or enhancements to be applied, playing a key role in customizing the model's functionality.
- Comfy dtype: LORA_HOOK
- Python dtype: LoraHookGroup

Output types

hook_CLIP
- Comfy dtype: CLIP
- The modified CLIP model with the LoRA hook attached, representing the enhanced version of the original model.
- Python dtype: CLIPWithHooks

Usage tips

Infra type: CPU
Common nodes: unknown

Source code

class SetClipLoraHook:
    @classmethod
    def INPUT_TYPES(s):
        return {
            "required": {
                "clip": ("CLIP",),
                "lora_hook": ("LORA_HOOK",),
            }
        }
    
    RETURN_TYPES = ("CLIP",)
    RETURN_NAMES = ("hook_CLIP",)
    CATEGORY = "Animate Diff 🎭🅐🅓/conditioning"
    FUNCTION = "apply_lora_hook"

    def apply_lora_hook(self, clip: CLIP, lora_hook: LoraHookGroup):
        new_clip = CLIPWithHooks(clip)
        new_clip.set_desired_hooks(lora_hooks=lora_hook)
        return (new_clip, )

1.8 KiB Raw Permalink Blame History