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

List of all members

Parameter class for LBFGS algorithm More...

Class Declaration

Deprecated:
This item will be removed in a future release.
struct DAAL_EXPORT Parameter : public optimization_solver::iterative_solver::interface1::Parameter
{
DAAL_DEPRECATED Parameter(sum_of_functions::interface1::BatchPtr function = sum_of_functions::interface1::BatchPtr(),
size_t nIterations = 100, double accuracyThreshold = 1.0e-5,
size_t batchSize = 10, size_t correctionPairBatchSize_ = 100,
size_t m = 10, size_t L = 10, size_t seed = 777);
DAAL_DEPRECATED_VIRTUAL virtual ~Parameter() {}
size_t m;
size_t L;
size_t seed;
engines::EnginePtr engine;
data_management::NumericTablePtr batchIndices;
data_management::NumericTablePtr correctionPairBatchIndices;
data_management::NumericTablePtr stepLengthSequence;
DAAL_DEPRECATED_VIRTUAL virtual services::Status check() const DAAL_C11_OVERRIDE;
};

Constructor & Destructor Documentation

DAAL_DEPRECATED Parameter ( sum_of_functions::interface1::BatchPtr  function = sum_of_functions::interface1::BatchPtr(),
size_t  nIterations = 100,
double  accuracyThreshold = 1.0e-5,
size_t  batchSize = 10,
size_t  correctionPairBatchSize_ = 100,
size_t  m = 10,
size_t  L = 10,
size_t  seed = 777 
)

Constructs the parameters of LBFGS algorithm

Parameters
[in]functionObjective function that can be represented as sum
[in]nIterationsMaximal number of iterations of the algorithm
[in]accuracyThresholdAccuracy of the LBFGS algorithm
[in]batchSizeNumber of observations to compute the stochastic gradient
[in]correctionPairBatchSize_The number of observations to compute the sub-sampled Hessian for correction pairs computation
[in]mMemory parameter of LBFGS
[in]LThe number of iterations between the curvature estimates calculations
[in]seedSeed for random choosing terms from objective function
Deprecated:
This item will be removed in a future release. Use engine instead.

Member Function Documentation

virtual DAAL_DEPRECATED_VIRTUAL services::Status check ( ) const
virtual

Checks the correctness of the parameter

Returns
Status of computations

Reimplemented from Parameter.

Member Data Documentation

data_management::NumericTablePtr correctionPairBatchIndices

Numeric table of size (nIterations / L) x correctionPairBatchSize that represent indices that will be used instead of random values for the sub-sampled Hessian matrix computations. If not set then random indices will be chosen.

size_t correctionPairBatchSize

Number of observations to compute the sub-sampled Hessian for correction pairs computation

engines::EnginePtr engine

Engine for random choosing terms from objective function.

size_t L

The number of iterations between the curvature estimates calculations

size_t m

Memory parameter of LBFGS. The maximum number of correction pairs that define the approximation of inverse Hessian matrix.

size_t seed

Seed for random choosing terms from objective function.

Deprecated:
This item will be removed in a future release. Use engine instead.
data_management::NumericTablePtr stepLengthSequence

Numeric table of size:

  • 1 x nIterations that contains values of the step-length sequence a(k), for k = 1, ..., nIterations, or
  • 1 x 1 that contains value of step length at each iteration a(1) = ... = a(nIterations)

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

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