C++ API Reference for Intel® Data Analytics Acceleration Library 2019 Update 5

References | Namespaces | Classes | Enumerations
Softmax Cross-entropy Layer

Contains classes for softmax cross-entropy layer. More...

References

 Backward Softmax Cross-entropy Layer
 Contains classes for the backward softmax cross-entropy layer.
 
 Forward Softmax Cross-entropy Layer
 Contains classes for the forward softmax cross-entropy layer.
 

Namespaces

 daal::algorithms::neural_networks::layers::loss::softmax_cross
 Contains classes for softmax cross-entropy layer.
 
 daal::algorithms::neural_networks::layers::loss::softmax_cross::interface1
 Contains version 1.0 of Intel(R) Data Analytics Acceleration Library (Intel(R) DAAL) interface.
 

Classes

class  Batch< algorithmFPType, method >
 Provides methods for the softmax cross-entropy layer in the batch processing mode. More...
 
class  Parameter
 Parameters for the softmax cross-entropy layer. More...
 

Enumerations

enum  Method { defaultDense = 0 }
 Computation methods for the softmax cross-entropy layer. More...
 
enum  LayerDataId { auxProbabilities = layers::lastLayerInputLayout + 1, auxGroundTruth }
 Identifiers of input objects for the backward softmax cross-entropy layer and results for the forward softmax cross-entropy layer. More...
 

Enumeration Type Documentation

enum LayerDataId

Enumerator
auxProbabilities 

Tensor that stores probabilities for the forward softmax cross-entropy layer

auxGroundTruth 

Tensor that stores ground truth data for the forward softmax cross-entropy layer

enum Method

Enumerator
defaultDense 

Default: performance-oriented method

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