1.3 KiB
1.3 KiB
tags
| tags | |||
|---|---|---|---|
|
FloatToNumber
Documentation
- Class name:
CM_FloatToNumber - Category:
math/conversion - Output node:
False
This node is designed for converting floating-point values to a more general number type, facilitating operations across different numeric types by standardizing them to a common format.
Input types
Required
a- The floating-point value to be converted into a general number type. This conversion enables seamless integration and manipulation within a broader range of mathematical operations.
- Comfy dtype:
FLOAT - Python dtype:
float
Output types
number- Comfy dtype:
NUMBER - The result of converting the input floating-point value to a general number type, allowing for its use in a wide array of numerical computations and operations.
- Python dtype:
float
- Comfy dtype:
Usage tips
- Infra type:
CPU - Common nodes: unknown
Source code
class FloatToNumber:
@classmethod
def INPUT_TYPES(cls) -> Mapping[str, Any]:
return {"required": {"a": ("FLOAT", {"default": 0.0})}}
RETURN_TYPES = ("NUMBER",)
FUNCTION = "op"
CATEGORY = "math/conversion"
def op(self, a: float) -> tuple[number]:
return (a,)