Skip to content

Lerp 📅🅕🅝

Documentation

  • Class name: Lerp
  • Category: FizzNodes 📅🅕🅝/WaveNodes
  • Output node: False

The Lerp node linearly interpolates between two values based on a given strength and the current frame within a specified number of images. It's designed to smoothly transition or blend values across a sequence, making it useful for animations or gradual changes.

Input types

Required

  • num_Images
    • Specifies the total number of images in the sequence. It determines the interpolation step size, affecting the smoothness of the transition.
    • Comfy dtype: FLOAT
    • Python dtype: float
  • strength
    • Defines the interpolation strength. It influences the range of output values, affecting how significantly the output changes across frames.
    • Comfy dtype: FLOAT
    • Python dtype: float
  • current_frame
    • Indicates the current frame number in the sequence. It's used to calculate the specific interpolation value for that frame, determining the output at each step.
    • Comfy dtype: INT
    • Python dtype: int

Output types

  • float
    • Comfy dtype: FLOAT
    • The floating-point result of the linear interpolation, representing the interpolated value at the current frame.
    • Python dtype: float
  • int
    • Comfy dtype: INT
    • The integer representation of the interpolated value, providing a discretized output for scenarios requiring whole numbers.
    • Python dtype: int

Usage tips

  • Infra type: CPU
  • Common nodes: unknown

Source code

class Lerp:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"num_Images": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 9999.0, "step": 1.0}),
                             "strength": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
                             "current_frame": ("INT", {"default": 1.0, "min": 0.0, "max": 9999, "step": 1.0}),
                             }}
    RETURN_TYPES = ("FLOAT", "INT",)
    FUNCTION = "lerp"

    CATEGORY = "FizzNodes 📅🅕🅝/WaveNodes"

    def lerp(self, num_Images, strength, current_frame):
        step = strength/num_Images
        output = strength - (step * current_frame)
        return (output, int(output),)