Batch Processing

Layer Input

The backward concat layer accepts the input described below. Pass the Input ID as a parameter to the methods that provide input for your algorithm. For more details, see Algorithms.

Input ID

Input

inputGradient

Pointer to the tensor of size n1 x ... x np that stores input gradient G computed on the preceding layer. This input can be an object of any class derived from Tensor.

inputFromForward

Collection of data needed for the backward concat layer.

Element ID

Element

auxInputDimensions

Collection of integers that stores the sizes along the k-th dimension of the input tensors in the forward computation step: nk1, ..., nkt.

Layer Parameters

For common parameters of neural network layers, see Common Parameters.

In addition to the common parameters, the backward concat layer has the following parameters:

Parameter

Default Value

Description

concatDimension

0

Index of the dimension along which deconcatenation should be implemented.

Layer Output

The backward concat layer calculates the result described below. Pass the Result ID as a parameter to the methods that access the results of your algorithm. For more details, see Algorithms.

Result ID

Result

resultLayerData

Collection of tensors of size n1 x ... x nkj x ... x np that stores the result of the backward concat layer. This collection can contain objects of any class derived from Tensor.

Note

The gradient field stores a null pointer. All the computation results are stored in resultLayerData.

For more complete information about compiler optimizations, see our Optimization Notice.
Select sticky button color: 
Orange (only for download buttons)