API Reference - AutomaticDifferentiationTensors

Constructors

new()

AutomaticDifferentiationTensors.new{tensor: tensor, PartialFirstDerivativeFunction: function, inputTensorArray: {tensor}}: AutomaticDifferentiationTensor

Parameters:

tensor: The tensor that is stored inside the automatic differentiation tensor.
PartialFirstDerivativeFunction: The function ths is involved in creating the automatic differentiation tensor.
inputTensorArray: An array containing the tensors that are involved in creating the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

createTensor()

AutomaticDifferentiationTensors.createTensor{dimensionSizeArray: {number}, allValues: number}: AutomaticDifferentiationTensor

Parameters:

dimensionSizeArray: The dimension size array for the automatic differentiation tensor.
allValues: The values to be set for the automatic differentiation tensor.

createRandomNormalTensor()

AutomaticDifferentiationTensors.createRandomNormalTensor{dimensionSizeArray: {number}, mean: number, standardDeviation: number}: AutomaticDifferentiationTensor

Parameters:

dimensionSizeArray: The dimension size array for the automatic differentiation tensor.
mean: The mean for the generated values.
standardDeviation: The standard deviation for the generated values.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

createRandomUniformTensor()

AutomaticDifferentiationTensors.createRandomNormalTensor{dimensionSizeArray: {number}, minimumValue: number, maximumValue: number}: AutomaticDifferentiationTensor

Parameters:

dimensionSizeArray: The dimension size array for the automatic differentiation tensor.
minimumValue: The minimum value for the generated values.
maximumValue: The maximum value for the generated values.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

radian()

AutomaticDifferentiationTensors.radian{tensor: tensor}: AutomaticDifferentiationTensor

Parameters:

tensor: The tensor that is used by the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

degree()

AutomaticDifferentiationTensors.degree{tensor: tensor}: AutomaticDifferentiationTensor

Parameters:

tensor: The tensor that is used by the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

sin()

AutomaticDifferentiationTensors.sin{tensor: tensor}: AutomaticDifferentiationTensor

Parameters:

tensor: The tensor that is used by the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

cos()

AutomaticDifferentiationTensors.cos{tensor: tensor}: AutomaticDifferentiationTensor

Parameters:

tensor: The tensor that is used by the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

tan()

AutomaticDifferentiationTensors.tan{tensor: tensor}: AutomaticDifferentiationTensor

Parameters:

tensor: The tensor that is used by the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

exponent()

AutomaticDifferentiationTensors.exponent{tensor: tensor}: AutomaticDifferentiationTensor

Parameters:

tensor: The tensor that is used by the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

logarithm()

AutomaticDifferentiationTensors.logarithm{numberTensor: tensor, baseTensor: tensor}: AutomaticDifferentiationTensor

Parameters:

numberTensor: The number tensor that is used by the automatic differentiation tensor.
baseTensor: The base tensor that is used by the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

clamp()

AutomaticDifferentiationTensors.clamp{tensor: tensor}: AutomaticDifferentiationTensor

Parameters:

tensor: The tensor that is used by the automatic differentiation tensor.
upperBondTensor: The upper bound tensor that is stored inside the automatic differentiation tensor.
lowerBondTensor: The lower bound tensor that is stored inside the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

maximum()

AutomaticDifferentiationTensors.maximum{...: tensor}: AutomaticDifferentiationTensor

Parameters:

tensor: The tensor that is used by the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

minimum()

AutomaticDifferentiationTensors.minimum{...: tensor}: AutomaticDifferentiationTensor

Parameters:

tensor: The tensor that is used by the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

add()

AutomaticDifferentiationTensors.add{...: tensor}: AutomaticDifferentiationTensor

Parameters:

tensor: The tensor that is used by the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

subtract()

AutomaticDifferentiationTensors.subtract{...: tensor}: AutomaticDifferentiationTensor

Parameters:

tensor: The tensor that is used by the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

multiply()

AutomaticDifferentiationTensors.multiply{...: tensor}: AutomaticDifferentiationTensor

Parameters:

tensor: The tensor that is used by the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

divide()

AutomaticDifferentiationTensors.divide{...: tensor}: AutomaticDifferentiationTensor

Parameters:

tensor: The tensor that is used by the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

concatenate()

AutomaticDifferentiationTensors.concatenate{...: tensor, dimension: number}: AutomaticDifferentiationTensor

Parameters:

tensor: The tensor that is used by the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

Arithmetic Functions

findMinimumValue()

AutomaticDifferentiationTensors:findMinimumValue{}: AutomaticDifferentiationTensor

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

findMaximumValue()

AutomaticDifferentiationTensors:findMaximumValue{}: AutomaticDifferentiationTensor

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

sum()

AutomaticDifferentiationTensors:sum{dimension: number}: AutomaticDifferentiationTensor

Parameters:

dimension: The dimension of calculating the sum along an axis. Can be empty. [Default: None]

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

unaryMinus()


AutomaticDifferentiationTensors:unaryMinus{}: AutomaticDifferentiationTensor

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

absolute()


AutomaticDifferentiationTensors:absolute{}: AutomaticDifferentiationTensor

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

power()

AutomaticDifferentiationTensors:power{otherTensor: tensor}: AutomaticDifferentiationTensor

Parameters:

otherTensor: The tensor to be used as an exponent.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

dotProduct()

AutomaticDifferentiationTensors:dotProduct{otherTensor: tensor}: AutomaticDifferentiationTensor

Parameters:

otherTensor: The tensor to be used for dot product operation.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

transpose()

AutomaticDifferentiationTensors:transpose{dimensionArray: {number}}: AutomaticDifferentiationTensor

Parameters:

dimensionArray: An array containing the dimensions to transpose the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

flatten()

AutomaticDifferentiationTensors:flatten{dimensionArray: {number}}: AutomaticDifferentiationTensor

Parameters:

dimensionArray: An array containing the dimensions to flatten the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

reshape()

AutomaticDifferentiationTensors:reshape{dimensionSizeArray: {number}}: AutomaticDifferentiationTensor

Parameters:

dimensionSizeArray: An array containing the dimension sizes to reshape the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

permute()

AutomaticDifferentiationTensors:permute{dimensionArray: {number}}: AutomaticDifferentiationTensor

Parameters:

dimensionArray: An array containing the dimensions to permute the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

mean()

AutomaticDifferentiationTensors:mean{dimension: number}: AutomaticDifferentiationTensor

Parameters:

dimension: The dimension of calculating the mean along an axis. Can be empty. [Default: None]

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

standardDeviation()

AutomaticDifferentiationTensors:standardDeviation{dimension: number}: AutomaticDifferentiationTensor

Parameters:

dimension: The dimension of calculating the standard deviation along an axis. Can be empty. [Default: None]

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

zScoreNormalization()

AutomaticDifferentiationTensors:zScoreNormalization{dimension: number}: AutomaticDifferentiationTensor

Parameters:

dimension: The dimension of calculating the z-score normalization along an axis. Can be empty. [Default: None]

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

expandDimensionSizes()

AutomaticDifferentiationTensors:expandDimensionSizes{targetDimensionSizeArray: {number}}: AutomaticDifferentiationTensor

Parameters:

targetDimensionSizeArray: The target dimension sizes to add to the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

expandNumberOfDimensions()

AutomaticDifferentiationTensors:expandNumberOfDimensions{dimensionSizeToAddArray: {number}}: AutomaticDifferentiationTensor

Parameters:

dimensionSizeToAddArray: The dimension size to add to the automatic differentiation tensor.

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

Non-Arithmetic Functions


AutomaticDifferentiationTensors:differentiate{firstDerivativeTensor: tensor}

Parameters:

firstDerivativeTensor: The tensor to be used for calculating chain rule first derivative tensors.

copy()


AutomaticDifferentiationTensors:copy{}: AutomaticDifferentiationTensor

Returns:

AutomaticDifferentiationTensor: The generated automatic differentiation tensor object.

getTensor()


AutomaticDifferentiationTensors:getTensor{doNotDeepCopy: boolean}: tensor

Parameters:

doNotDeepCopy: Set whether or not to deep copy the tensor.

Returns:

tensor: The tensor that is stored inside the automatic differentiation tensor.

setTensor()

AutomaticDifferentiationTensors:setTensor{tensor: tensor, doNotDeepCopy: boolean}

Parameters:

tensor: The tensor that is will be stored inside the automatic differentiation tensor.
doNotDeepCopy: Set whether or not to deep copy the tensor.

getTotalFirstDerivativeTensor()

AutomaticDifferentiationTensors:getTotalFirstDerivativeTensor{doNotDeepCopy: boolean}: tensor

Parameters:

doNotDeepCopy: Set whether or not to deep copy the total first derivative tensor.

Returns:

tensor: The total first derivative tensor that is stored inside the automatic differentiation tensor.

setTotalFirstDerivativeTensor()

AutomaticDifferentiationTensors:setTotalFirstDerivativeTensor{tensor: tensor, doNotDeepCopy: boolean}

Parameters:

tensor: The total first derivative tensor that is will be stored inside the automatic differentiation tensor.
doNotDeepCopy: Set whether or not to deep copy the total first derivative tensor.

destroy()

AutomaticDifferentiationTensors:destroy{areDescendantsDestroyed: boolean, destroyFirstInputTensor: boolean}

Parameters:

areDescendantsDestroyed: Set whether or not to destroy descendants of the automatic differentiation tensor.
destroyFirstInputTensor: Set whether or not to destroy the very first tensors that are used as inputs.

isAutomaticDifferentiationTensor()


AutomaticDifferentiationTensors:isAutomaticDifferentiationTensor{}: boolean

Returns:

isAutomaticDifferentiationTensor: A boolean that indicates if the object is an automatic differentiation tensor object.