C++ API Reference for Intel® Data Analytics Acceleration Library 2020 Update 1
Parameter base class for the iterative solver algorithm More...
Parameter | ( | const sum_of_functions::interface1::BatchPtr & | function_, |
size_t | nIterations_ = 100 , |
||
double | accuracyThreshold_ = 1.0e-05 , |
||
bool | optionalResultReq_ = false , |
||
size_t | batchSize_ = 1 |
||
) |
Constructs the parameter base class of the iterative solver algorithm
[in] | function_ | Objective function represented as sum of functions |
[in] | nIterations_ | Maximal number of iterations of the algorithm |
[in] | accuracyThreshold_ | Accuracy of the algorithm. The algorithm terminates when this accuracy is achieved |
[in] | optionalResultReq_ | Flag indicating if algorithm-specific result data generation is needed |
[in] | batchSize_ | Batch size |
|
virtual |
Checks the correctness of the parameter
Reimplemented in Parameter< momentum >, Parameter< miniBatch >, Parameter< defaultDense >, Parameter, Parameter, Parameter, and BaseParameter.
double accuracyThreshold |
Accuracy of the algorithm. The algorithm terminates when this accuracy is achieved
size_t batchSize |
Number of batch indices to compute the stochastic gradient. If batchSize is equal to the number of terms in objective function then no random sampling is performed, and all terms are used to calculate the gradient. This parameter is ignored if batchIndices is provided.
sum_of_functions::interface1::BatchPtr function |
Objective function represented as sum of functions
size_t nIterations |
Maximal number of iterations of the algorithm
bool optionalResultRequired |
Indicates whether optional result is required
For more complete information about compiler optimizations, see our Optimization Notice.