RegionalPrompt

Documentation

• Class name: RegionalPrompt
• Category: ImpactPack/Regional
• Output node: False

The RegionalPrompt node is designed to generate regional prompts based on a given mask and an advanced sampler. It focuses on creating specific prompts that are tailored to particular regions of an image, enhancing the customization and precision of image generation tasks.

Input types

Required

• mask
  • The mask parameter specifies the area of interest within an image for which the regional prompt will be generated. It plays a crucial role in defining the scope and focus of the prompt generation process.
  • Comfy dtype: MASK
  • Python dtype: torch.Tensor
• advanced_sampler
  • The advanced_sampler parameter is utilized to apply sophisticated sampling techniques for generating the regional prompt. It significantly influences the quality and characteristics of the generated prompt.
  • Comfy dtype: KSAMPLER_ADVANCED
  • Python dtype: KSAMPLER_ADVANCED

Output types

• regional_prompts
  • Comfy dtype: REGIONAL_PROMPTS
  • This output consists of regional prompts generated based on the specified mask and advanced sampler. It's essential for subsequent image generation or manipulation tasks.
  • Python dtype: List[str]

Usage tips

• Infra type: CPU
• Common nodes: unknown

Source code

class RegionalPrompt:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {
                     "mask": ("MASK", ),
                     "advanced_sampler": ("KSAMPLER_ADVANCED", ),
                     },
                }

    RETURN_TYPES = ("REGIONAL_PROMPTS", )
    FUNCTION = "doit"

    CATEGORY = "ImpactPack/Regional"

    def doit(self, mask, advanced_sampler):
        regional_prompt = core.REGIONAL_PROMPT(mask, advanced_sampler)
        return ([regional_prompt], )