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

References | Namespaces | Classes | Enumerations
Limited-Memory-Broyden-Fletcher-Goldfarb-Shanno Algorithm

Contains classes for computing the limited memory Broyden-Fletcher-Goldfarb-Shanno algorithm. More...

References

 Batch
 

Namespaces

 daal::algorithms::optimization_solver::lbfgs
 Contains classes for computing the limited memory Broyden-Fletcher-Goldfarb-Shanno algorithm.
 
 daal::algorithms::optimization_solver::lbfgs::interface1
 Contains version 1.0 of Intel(R) Data Analytics Acceleration Library (Intel(R) DAAL) interface.
 

Classes

struct  Parameter
 Parameter class for LBFGS algorithm More...
 
class  Input
 Input class for LBFGS algorithm More...
 
class  Result
 Results obtained with the compute() method of the LBFGS algorithm in the batch processing mode. More...
 

Enumerations

enum  Method { defaultDense = 0 }
 
enum  OptionalDataId { correctionPairs = iterative_solver::lastOptionalData + 1, correctionIndices, averageArgumentLIterations }
 

Enumeration Type Documentation

enum Method

Available methods for computing LBFGS

Enumerator
defaultDense 

Default: performance-oriented method.

enum OptionalDataId

Available identifiers of optional input for the iterative solver

Enumerator
correctionPairs 

Correction pairs table. Numeric table 2*m x n, where rows (0, m-1) represent correction vectors S and rows (m, 2*m-1) represent correction vectors Y

correctionIndices 

Numeric table of size 1 x 2 with 32-bit integer indexes. The first value is the index of correction pair t, the second value is the index of the last iteration k from the previous run

averageArgumentLIterations 

Numeric table of size 2 x n, where row 0 represent average arguments for the previous L iterations and row 1 represent average arguments for the last L iterations. These values are required to compute S correction vectors on the next step

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