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

Namespaces | Enumerations
daal::algorithms::neural_networks::training Namespace Reference

Contains classes for training the model of the neural network.

Namespaces

 interface1
 Contains version 1.0 of Intel(R) Data Analytics Acceleration Library (Intel(R) DAAL) interface.
 

Enumerations

enum  Method { defaultDense = 0, feedforwardDense = 0 }
 Computation methods for the neural network model based training. More...
 
enum  InputId { data, groundTruth }
 Available identifiers of input objects for the neural network model based training. More...
 
enum  InputCollectionId { groundTruthCollection = lastInputId + 1 }
 Available identifiers of input collection objects for the neural network model based training. More...
 
enum  Step1LocalInputId { inputModel = lastInputCollectionId + 1 }
 Available identifiers of input objects for the neural network model based training. More...
 
enum  Step2MasterInputId { partialResults }
 Partial results from the previous steps in the distributed processing mode required by the second distributed step of the algorithm. More...
 
enum  Step1LocalPartialResultId
 Available identifiers of partial results of the neural network training algorithm required by the first distributed step. More...
 
enum  Step2MasterPartialResultId
 Available identifiers of partial results of the neural network training algorithm equired by the second distributed step. More...
 
enum  ResultId { model = 0 }
 Available identifiers of result of the neural network model based training. More...
 

Enumeration Type Documentation

enum Method

Enumerator
defaultDense 

Default: Feedforward neural network

feedforwardDense 

Feedforward neural network

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