API Reference - PoolingLayers

Constructors

FastAveragePooling1D


PoolingLayers.FastAveragePooling1D{tensor: tensor, kernelDimensionSize: number, strideDimensionSize: number}: AutomaticDifferentiationTensor

Parameters:

  • tensor: The tensor that will be used as inputs.

  • kernelDimensionSize: The dimension size for the kernel. [Default: 2]

  • strideDimensionSize: The dimension size for the stride. [Default: 1]

Returns:

  • AutomaticDifferentiationTensor: The automatic differentiation tensor that is created as a result of calling this function.

FastAveragePooling2D


PoolingLayers.FastAveragePooling2D{tensor: tensor,  kernelDimensionSize: {number}, strideDimensionSizeArray: {number}}: AutomaticDifferentiationTensor

Parameters:

  • tensor: The tensor that will be used as inputs.

  • kernelDimensionSizeArray: The dimension size for the stride. The index of the array represents the dimension and the value represents the size for that particular dimension. [Default: {2, 2}]

  • strideDimensionSizeArray: The dimension size for the stride. The index of the array represents the dimension and the value represents the size for that particular dimension. [Default: {1, 1}]

Returns:

  • AutomaticDifferentiationTensor: The automatic differentiation tensor that is created as a result of calling this function.

FastAveragePooling3D


PoolingLayers.FastAveragePooling3D{tensor: tensor,  kernelDimensionSizeArray: {number}, strideDimensionSizeArray: {number}}: AutomaticDifferentiationTensor

Parameters:

  • tensor: The tensor that will be used as inputs.

  • kernelDimensionSizeArray: The dimension size for the stride. The index of the array represents the dimension and the value represents the size for that particular dimension. [Default: {2, 2, 2}]

  • strideDimensionSizeArray: The dimension size for the stride. The index of the array represents the dimension and the value represents the size for that particular dimension. [Default: {1, 1, 1}]

Returns:

  • AutomaticDifferentiationTensor: The automatic differentiation tensor that is created as a result of calling this function.

FastMaximumPooling1D


PoolingLayers.FastMaximumPooling1D{tensor: tensor, kernelDimensionSize: number, strideDimensionSize: number}: AutomaticDifferentiationTensor

Parameters:

  • tensor: The tensor that will be used as inputs.

  • kernelDimensionSize: The dimension size for the kernel. [Default: 2]

  • strideDimensionSize: The dimension size for the stride. [Default: 1]

Returns:

  • AutomaticDifferentiationTensor: The automatic differentiation tensor that is created as a result of calling this function.

FastMaximumPooling2D


PoolingLayers.FastMaximumPooling2D{tensor: tensor,  kernelDimensionSize: {number}, strideDimensionSizeArray: {number}}: AutomaticDifferentiationTensor

Parameters:

  • tensor: The tensor that will be used as inputs.

  • kernelDimensionSizeArray: The dimension size for the stride. The index of the array represents the dimension and the value represents the size for that particular dimension. [Default: {2, 2}]

  • strideDimensionSizeArray: The dimension size for the stride. The index of the array represents the dimension and the value represents the size for that particular dimension. [Default: {1, 1}]

Returns:

  • AutomaticDifferentiationTensor: The automatic differentiation tensor that is created as a result of calling this function.

FastMaximumPooling3D


PoolingLayers.FastMaximumPooling3D{tensor: tensor,  kernelDimensionSizeArray: {number}, strideDimensionSizeArray: {number}}: AutomaticDifferentiationTensor

Parameters:

  • tensor: The tensor that will be used as inputs.

  • kernelDimensionSizeArray: The dimension size for the stride. The index of the array represents the dimension and the value represents the size for that particular dimension. [Default: {2, 2, 2}]

  • strideDimensionSizeArray: The dimension size for the stride. The index of the array represents the dimension and the value represents the size for that particular dimension. [Default: {1, 1, 1}]

Returns:

  • AutomaticDifferentiationTensor: The automatic differentiation tensor that is created as a result of calling this function.

FastMinimumPooling1D


PoolingLayers.FastMinimumPooling1D{tensor: tensor, kernelDimensionSize: number, strideDimensionSize: number}: AutomaticDifferentiationTensor

Parameters:

  • tensor: The tensor that will be used as inputs.

  • kernelDimensionSize: The dimension size for the kernel. [Default: 2]

  • strideDimensionSize: The dimension size for the stride. [Default: 1]

Returns:

  • AutomaticDifferentiationTensor: The automatic differentiation tensor that is created as a result of calling this function.

FastMinimumPooling2D


PoolingLayers.FastMinimumPooling2D{tensor: tensor,  kernelDimensionSize: {number}, strideDimensionSizeArray: {number}}: AutomaticDifferentiationTensor

Parameters:

  • tensor: The tensor that will be used as inputs.

  • kernelDimensionSizeArray: The dimension size for the stride. The index of the array represents the dimension and the value represents the size for that particular dimension. [Default: {2, 2}]

  • strideDimensionSizeArray: The dimension size for the stride. The index of the array represents the dimension and the value represents the size for that particular dimension. [Default: {1, 1}]

Returns:

  • AutomaticDifferentiationTensor: The automatic differentiation tensor that is created as a result of calling this function.

FastMinimumPooling3D


PoolingLayers.FastMinimumPooling3D{tensor: tensor,  kernelDimensionSizeArray: {number}, strideDimensionSizeArray: {number}}: AutomaticDifferentiationTensor

Parameters:

  • tensor: The tensor that will be used as inputs.

  • kernelDimensionSizeArray: The dimension size for the stride. The index of the array represents the dimension and the value represents the size for that particular dimension. [Default: {2, 2, 2}]

  • strideDimensionSizeArray: The dimension size for the stride. The index of the array represents the dimension and the value represents the size for that particular dimension. [Default: {1, 1, 1}]

Returns:

  • AutomaticDifferentiationTensor: The automatic differentiation tensor that is created as a result of calling this function.

FastMaximumUnpoolingPooling1D


PoolingLayers.FastMaximumUnpoolingPooling1D{tensor: tensor, kernelDimensionSize: number, strideDimensionSize: number, unpoolingMethod: string}: AutomaticDifferentiationTensor

Parameters:

  • tensor: The tensor that will be used as inputs.

  • kernelDimensionSize: The dimension size for the kernel. [Default: 2]

  • strideDimensionSize: The dimension size for the stride. [Default: 1]

  • unpoolingMethod: The unpooling method that determines how the transformed tensor is generated. Available options are:

    • NearestNeighbour (Default)

    • BedOfNails

Returns:

  • AutomaticDifferentiationTensor: The automatic differentiation tensor that is created as a result of calling this function.

FastMaximumUnpoolingPooling2D


PoolingLayers.FastMaximumUnpoolingPooling2D{tensor: tensor,  kernelDimensionSize: {number}, strideDimensionSizeArray: {number}: unpoolingMethod: string}: AutomaticDifferentiationTensor

Parameters:

  • tensor: The tensor that will be used as inputs.

  • kernelDimensionSizeArray: The dimension size for the stride. The index of the array represents the dimension and the value represents the size for that particular dimension. [Default: {2, 2}]

  • strideDimensionSizeArray: The dimension size for the stride. The index of the array represents the dimension and the value represents the size for that particular dimension. [Default: {1, 1}]

  • unpoolingMethod: The unpooling method that determines how the transformed tensor is generated. Available options are:

    • NearestNeighbour (Default)

    • BedOfNails

Returns:

  • AutomaticDifferentiationTensor: The automatic differentiation tensor that is created as a result of calling this function.

FastMaximumUnpoolingPooling3D


PoolingLayers.FastMaximumUnpoolingPooling3D{tensor: tensor,  kernelDimensionSizeArray: {number}, strideDimensionSizeArray: {number}, unpoolingMethod: string}: AutomaticDifferentiationTensor

Parameters:

  • tensor: The tensor that will be used as inputs.

  • kernelDimensionSizeArray: The dimension size for the stride. The index of the array represents the dimension and the value represents the size for that particular dimension. [Default: {2, 2, 2}]

  • strideDimensionSizeArray: The dimension size for the stride. The index of the array represents the dimension and the value represents the size for that particular dimension. [Default: {1, 1, 1}]

  • unpoolingMethod: The unpooling method that determines how the transformed tensor is generated. Available options are:

    • NearestNeighbour (Default)

    • BedOfNails

Returns:

  • AutomaticDifferentiationTensor: The automatic differentiation tensor that is created as a result of calling this function.