C++ API Reference for Intel® Data Analytics Acceleration Library 2020 Update 1
Provides methods for the backward loss layer in the batch processing mode. More...
algorithmFPType | Data type to use in intermediate computations for the backward loss layer, double or float |
|
inline |
Default constructor
Constructs a backward loss layer by copying input objects and parameters of another backward loss layer in the batch processing mode
[in] | other | Algorithm to use as the source to initialize the input objects and parameters of the layer |
|
inline |
Returns a pointer to a newly allocated backward loss layer with a copy of the input objects and parameters for this backward loss layer in the batch processing mode
For more complete information about compiler optimizations, see our Optimization Notice.