C++ API Reference for Intel® Data Analytics Acceleration Library 2020 Update 1

List of all members
Model Class Reference

Class representing the model of neural network. More...

Class Declaration

Deprecated:
This item will be removed in a future release.

Member Function Documentation

services::Status allocate ( const services::Collection< size_t > &  sampleSize,
const Parameter parameter = Parameter() 
)
inline

Allocates the buffers needed for the training using neural network

Parameters
[in]sampleSizeDimensionality of the batch for the input to the first layer
[in]parameterParameters of the training
Returns
Status of computations
Deprecated:
This item will be removed in a future release.
const layers::backward::LayerIfacePtr getBackwardLayer ( size_t  index) const
inline

Returns the backward stage of a layer with certain index in the network

Parameters
[in]indexIndex of the layer in the network
Returns
Backward stage of a layer with certain index in the network
Deprecated:
This item will be removed in a future release.
const BackwardLayersPtr getBackwardLayers ( ) const
inline

Returns list of backward layers

Returns
List of backward layers
Deprecated:
This item will be removed in a future release.
DAAL_DEPRECATED const services::ErrorCollection& getErrors ( ) const
inline

Returns the errors of the Model

Deprecated:
This item will be removed in a future release.
Returns
Collection of errors
Deprecated:
This item will be removed in a future release.
const layers::forward::LayerIfacePtr getForwardLayer ( size_t  index) const
inline

Returns the forward stage of a layer with certain index in the network

Parameters
[in]indexIndex of the layer in the network
Returns
Forward stage of a layer with certain index in the network
Deprecated:
This item will be removed in a future release.
DAAL_DEPRECATED const ForwardLayersPtr getForwardLayers ( ) const
inline

Returns list of forward layers

Returns
List of forward layers
Deprecated:
This item will be removed in a future release.
const prediction::ModelPtr getPredictionModel ( )
inline

Returns list of forward layers and their parameters organised in the prediction::Model

Returns
List of forward layers and their parameters organised in the prediction::Model
Deprecated:
This item will be removed in a future release.
DAAL_DEPRECATED algorithms::OptionalArgumentPtr getSolverOptionalArgument ( size_t  index)
inline

Return the OptionalArgument from the neural netowrk model that stores intermediate status of solver between epochs by index

Parameters
indexIndex in collection of required OptionalArgument
Returns
the OptionalArgument from the neural netowrk model that stores intermediate status of solver between epochs
Deprecated:
This item will be removed in a future release.
DAAL_DEPRECATED data_management::DataCollection getSolverOptionalArgumentCollection ( )
inline

Return the OptionalArgument from the neural netowrk model that stores intermediate status of solver between epochs

Returns
the OptionalArgument from the neural netowrk model that stores intermediate status of solver between epochs
Deprecated:
This item will be removed in a future release.
DAAL_DEPRECATED data_management::NumericTablePtr getWeightsAndBiases ( size_t  idx) const

Returns the weights and biases of the forward layer of neural network as numeric table

Parameters
[in]idxIndex of the backward layer
Returns
Weights and biases derivatives container
Deprecated:
This item will be removed in a future release.
DAAL_DEPRECATED data_management::NumericTablePtr getWeightsAndBiasesDerivatives ( ) const

Returns the weights and biases derivatives of all backward layers of neural network as numeric table

Returns
Weights and biases derivatives container
Deprecated:
This item will be removed in a future release.
DAAL_DEPRECATED data_management::NumericTablePtr getWeightsAndBiasesDerivatives ( size_t  idx) const

Returns the weights and biases derivatives of the backward layer of neural network as numeric table

Parameters
[in]idxIndex of the backward layer
Returns
Weights and biases derivatives container
Deprecated:
This item will be removed in a future release.
DAAL_DEPRECATED bool getWeightsAndBiasesStorageStatus ( ) const
inline

Returns weights and biases storage status

Returns
Weights and biases storage status. True if weights and biases of all layers stored in one numeric table. False otherwise.
Deprecated:
This item will be removed in a future release.
services::Status initialize ( const services::Collection< size_t > &  sampleSize,
const Topology topology,
const Parameter parameter = Parameter() 
)
inline

Initializes neural network

Parameters
[in]sampleSizeDimensionality of the batch for the input to the first layer
[in]topologyCollection of LayerDescriptor of every inserted layer
[in]parameterParameters of the training
Returns
Status of computations
Deprecated:
This item will be removed in a future release.
DAAL_DEPRECATED services::Status setErrors ( services::ErrorCollection &  errors)
inline

Sets the error collection to the Model

Parameters
[in]errorsCollection of errors
Deprecated:
This item will be removed in a future release.
Returns
Status of computations
Deprecated:
This item will be removed in a future release.
DAAL_DEPRECATED services::Status setSolverOptionalArgument ( const algorithms::OptionalArgumentPtr &  solverOptionalArgument,
size_t  index 
)
inline

Sets the OptionalArgument to neural netowrk model to store intermediate status of solver between epochs

Parameters
solverOptionalArgumentOptionalArgumentPtr to set in collection
indexIndex in collection of required OptionalArgument
Returns
Status of computations
Deprecated:
This item will be removed in a future release.
DAAL_DEPRECATED services::Status setSolverOptionalArgumentCollection ( const data_management::DataCollection &  solverOptionalArgumentCollection)
inline

Sets the OptionalArgument to neural netowrk model to store intermediate status of solver between epochs

Parameters
solverOptionalArgumentCollectionStructure to store intermediate status of solver
Returns
Status of computations
Deprecated:
This item will be removed in a future release.
DAAL_DEPRECATED services::Status setWeightsAndBiases ( size_t  idx,
const data_management::NumericTablePtr &  table 
)

Sets table containing weights and biases of one forward layer of neural network

Parameters
[in]idxIndex of the forward layer
[in]tableTable containing weights and biases of one forward layer of neural network
Returns
Status of computations
Deprecated:
This item will be removed in a future release.

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

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