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

List of all members
Batch< algorithmFPType, method > Class Template Reference

Provides methods for neural network model-based training in the batch processing mode. More...

Class Declaration

template<typename algorithmFPType = DAAL_ALGORITHM_FP_TYPE, Method method = defaultDense>
class daal::algorithms::neural_networks::training::interface1::Batch< algorithmFPType, method >

Template Parameters
algorithmFPTypeData type to use in intermediate computations for neural network model-based training, double or float
methodNeural network training method, training::Method
Enumerations
References

Constructor & Destructor Documentation

Batch ( services::SharedPtr< optimization_solver::iterative_solver::Batch< algorithmFPType, method > >  optimizationSolver_)
inline

Default constructor

Batch ( const Batch< algorithmFPType, method > &  other)
inline

Constructs neural network by copying input objects and parameters of another neural network

Parameters
[in]otherAn algorithm to be used as the source to initialize the input objects and parameters of the algorithm

Member Function Documentation

services::SharedPtr<Batch<algorithmFPType, method> > clone ( ) const
inline

Returns a pointer to the newly allocated neural network with a copy of input objects and parameters of this neural network

Returns
Pointer to the newly allocated layer
virtual int getMethod ( ) const
inlinevirtual

Returns method of the algorithm

Returns
Method of the algorithm
ResultPtr getResult ( )
inline

Returns the structure that contains the results of the neural network algorithm

Returns
Structure that contains the results of the neural network algorithm
services::Status initialize ( const services::Collection< size_t > &  sampleSize,
const training::Topology &  topology 
)
inline

Initializes the neural network topology

Parameters
[in]sampleSizeDimensionality of the batch for the input to the first layer
[in]topologyNeural network topology
Returns
Status of computations
services::Status setResult ( const ResultPtr &  res)
inline

Register user-allocated memory to store the results of the neural network algorithm

Parameters
[in]resStructure to store the results of the neural network algorithm
Returns
Status of computations

Member Data Documentation

InputType input

Input data structure

ParameterType parameter

Training parameters


The documentation for this class was generated from the following file:

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