Skip to content

Create Sigma Schedule 🎭🅐🅓

Documentation

  • Class name: ADE_SigmaSchedule
  • Category: Animate Diff 🎭🅐🅓/sample settings/sigma schedule
  • Output node: False

The ADE_SigmaSchedule node is designed to generate a sigma schedule based on a given beta schedule. It abstracts the complexity of sigma schedule creation, offering a straightforward way to obtain a sigma schedule tailored to specific model sampling types and configurations.

Input types

Required

  • beta_schedule
    • Specifies the beta schedule to be used for generating the sigma schedule. This parameter is crucial as it determines the base configuration from which the sigma schedule will be derived.
    • Comfy dtype: COMBO[STRING]
    • Python dtype: BetaSchedules.ALIAS_ACTIVE_LIST

Output types

  • sigma_schedule
    • Comfy dtype: SIGMA_SCHEDULE
    • Outputs a sigma schedule object, which is essential for defining the progression of noise levels in diffusion-based generative models.
    • Python dtype: SigmaSchedule (custom type from the animatediff package)

Usage tips

  • Infra type: CPU
  • Common nodes: unknown

Source code

class SigmaScheduleNode:
    @classmethod
    def INPUT_TYPES(s):
        return {
            "required": {
                "beta_schedule": (BetaSchedules.ALIAS_ACTIVE_LIST,),
            }
        }

    RETURN_TYPES = ("SIGMA_SCHEDULE",)
    CATEGORY = "Animate Diff 🎭🅐🅓/sample settings/sigma schedule"
    FUNCTION = "get_sigma_schedule"

    def get_sigma_schedule(self, beta_schedule: str):
        model_type = ModelSamplingType.from_alias(ModelSamplingType.EPS)
        new_model_sampling = BetaSchedules._to_model_sampling(alias=beta_schedule,
                                                              model_type=model_type)
        return (SigmaSchedule(model_sampling=new_model_sampling, model_type=model_type),)