C++ API Reference for Intel® Data Analytics Acceleration Library 2018 Update 3

References | Namespaces | Classes | Enumerations

Contains classes for neural network fully-connected layer. More...

References

 Backward Fully-connected Layer
 Contains classes for backward fully-connected layer.
 
 Forward Fully-connected Layer
 Contains classes for the forward fully-connected layer.
 

Namespaces

 daal::algorithms::neural_networks::layers::fullyconnected
 Contains classes for neural network fully-connected layer.
 
 daal::algorithms::neural_networks::layers::fullyconnected::interface1
 Contains version 1.0 of Intel(R) Data Analytics Acceleration Library (Intel(R) DAAL) interface.
 

Classes

class  Batch< algorithmFPType, method >
 Computes the result of the forward and backward fully-connected layer of neural network in the batch processing mode. More...
 
class  Parameter
 Fully-connected layer parameters. More...
 

Enumerations

enum  Method { defaultDense = 0 }
 
enum  LayerDataId { auxData = 0, auxWeights = 1 }
 

Enumeration Type Documentation

enum LayerDataId

Available identifiers of results of the forward fully-connected layer and input objects for the backward fully-connected layer

Enumerator
auxData 

Data processed at the forward stage of the layer

auxWeights 

Weights used at the forward stage of the layer

enum Method

Available methods to compute forward and backward fully-connected layer

Enumerator
defaultDense 

Default: performance-oriented method.

For more complete information about compiler optimizations, see our Optimization Notice.