KJNodes/noisegenerated

Float To Sigmas(FloatToSigmas)

From a technical implementation perspective, this node accepts a floating-point number input in the form of a Python list.

Float To Sigmas

SIGMAS
float_list
0
KJNodes

The Float To Sigmas node is a utility node for noise scheduling. Its core function is to convert a single floating-point value into a SIGMAS tensor specifically used by diffusion models in ComfyUI. This conversion process serves as a crucial bridge connecting a custom noise parameter with standard sampler nodes.

Node Function

From a technical implementation perspective, this node receives an input in the form of a single floating-point number. Internally, it uses PyTorch's torch.tensor method to convert the input value into a 32-bit floating-point tensor. The format of this tensor is fully compatible with the SIGMAS input format required by sampler nodes like KSampler in ComfyUI, enabling seamless data integration.

Node Parameter Description - Float To Sigmas

Control Parameters (Parameters)

Parameter NameData TypeRequiredDefault ValueValue Range/OptionsDescription
float_listFLOATYes0.0-Input parameter float_list: A single floating-point number used to create the SIGMAS tensor. This parameter accepts any float value.

Output

Parameter NameData TypeDescription
SIGMASSIGMASOutput parameter SIGMAS: The converted SIGMAS tensor, which can be directly used by sampler nodes.

Usage Scenarios

In practical workflows, this node is very useful when you need a constant or precisely defined noise intensity value that can be used as a SIGMAS input. For example, you can first use a Primitive node to set a custom floating-point value, then use this node to convert it into SIGMAS, and finally input it into the sigmas port of a KSampler to execute noise scheduling with a fixed step.

The Float To Sigmas node is from the ComfyUI-KJNodes node package.

Comments

Sign in with GitHub to join the discussion.

Loading comments…