Interface Delta

Local estimate of the derivative of the input data along the selected axis. Outputs the derivative of the input features along the window.

Example

var deltaBuilder = MachineLearning.createDeltaBuilder();
var delta = deltaBuilder.setNumFeatures(numFeatures).setWindowSize(windowSize).build();
interface Delta {
    maxTensorSize: number;
    getTypeName(): string;
    isOfType(type: string): boolean;
    isSame(other: ScriptObject): boolean;
    process(inTensor: Float32Array, inShape: vec3, outTensor: Float32Array): vec3;
}

Hierarchy (view full)

Properties

maxTensorSize

Methods

getTypeName isOfType isSame process

Properties

ReadonlymaxTensorSize

maxTensorSize: number

Maximum amount of features (channels).

Methods

getTypeName

isOfType

  • isOfType(type): boolean

  • Returns true if the object matches or derives from the passed in type.

    Parameters

    • type: string

    Returns boolean

isSame

  • isSame(other): boolean

  • Returns true if this object is the same as other. Useful for checking if two references point to the same thing.

    Parameters

    Returns boolean

process

  • process(inTensor, inShape, outTensor): vec3

  • Process inTensor with shape inShape and writes the result to the outTensor and returns the shape of outTensor.

    Parameters

    • inTensor: Float32Array
    • inShape: vec3
    • outTensor: Float32Array

    Returns vec3

Settings

Member Visibility

On This Page

Properties
Methods