Developer Guide for Intel® Data Analytics Acceleration Library 2018 Update 2

Batch Processing

Layer Input

The backward split 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

inputGradientCollection

Collection of k tensors of size n1 x ... x np that stores the result of the forward split layer. This collection can contain objects of any class derived from Tensor.

Note

inputGradient stores a null pointer while inputGradientCollection stores all input tensors.

Layer Parameters

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

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

Parameter

Default Value

Description

nInputs

1

Number of input tensors.

Layer Output

The backward split 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

gradient

Tensor of size n1 x ... x np that stores result z of the backward split layer. This result can be an object of any class derived from Tensor.

Examples

C++: split_layer_dense_batch.cpp

Java*: SplitLayerDenseBatch.java

Python*: split_layer_dense_batch.py

See Also