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

References | Namespaces | Classes | Enumerations
Singular Value Decomposition

Contains classes to run the singular-value decomposition (SVD) algorithm. More...

References

 Batch
 
 Distributed
 
 Online
 

Namespaces

 daal::algorithms::svd
 Contains classes to run the singular-value decomposition (SVD) algorithm.
 
 daal::algorithms::svd::interface1
 Contains version 1.0 of Intel(R) Data Analytics Acceleration Library (Intel(R) DAAL) interface.
 

Classes

struct  Parameter
 Parameters for the computation method of the SVD algorithm. More...
 
class  Input
 Input objects for the SVD algorithm in the batch processing and online processing modes, and the first step in the distributed processing mode. More...
 
class  DistributedStep2Input
 Input objects for the second step of the SVD algorithm in the distributed processing mode More...
 
class  DistributedStep3Input
 Input objects for the third step of the SVD algorithm in the distributed processing mode More...
 
class  OnlinePartialResult
 Provides methods to access partial results obtained with the compute() method of the SVD algorithm in the online processing mode or the first step in the distributed processing mode. More...
 
class  Result
 Provides methods to access final results obtained with the compute() method of the SVD algorithm in the batch processing mode or with the finalizeCompute() method in the online processing mode or steps 2 and 3 in the distributed processing mode. More...
 
class  DistributedPartialResult
 Provides methods to access partial results obtained with the compute() method of the SVD algorithm in the second step in the distributed processing mode. More...
 
class  DistributedPartialResultStep3
 Provides methods to access partial results obtained with the compute() method of the SVD algorithm in the third step in the distributed processing mode. More...
 

Enumerations

enum  Method { defaultDense = 0 }
 
enum  SVDResultFormat { notRequired, requiredInPackedForm }
 
enum  InputId { data }
 Available types of input objects for the SVD algorithm. More...
 
enum  ResultId { singularValues, leftSingularMatrix, rightSingularMatrix }
 Available types of results of the SVD algorithm. More...
 
enum  PartialResultId { outputOfStep1ForStep3, outputOfStep1ForStep2 }
 Available types of partial results of the SVD algorithm obtained in the online processing mode and in the first step in the distributed processing mode. More...
 
enum  DistributedPartialResultCollectionId { outputOfStep2ForStep3 }
 Available types of partial results obtained in the second step of the SVD algorithm in the distributed processing mode, stored in the DataCollection object. More...
 
enum  DistributedPartialResultId { finalResultFromStep2Master = lastDistributedPartialResultCollectionId + 1 }
 Available types of partial results obtained in the second step of the SVD algorithm in the distributed processing mode, stored in the Result object. More...
 
enum  DistributedPartialResultStep3Id { finalResultFromStep3 }
 Available types of partial results obtained in the third step of the SVD algorithm in the distributed processing mode, stored in the Result object. More...
 
enum  MasterInputId { inputOfStep2FromStep1 }
 Partial results from previous steps in the distributed processing mode, required by the second step. More...
 
enum  FinalizeOnLocalInputId { inputOfStep3FromStep1, inputOfStep3FromStep2 }
 Partial results from previous steps in the distributed processing mode, required by the third step. More...
 

Enumeration Type Documentation

enum DistributedPartialResultCollectionId

Enumerator
outputOfStep2ForStep3 

DataCollection with data to be transferred to the third step in the distributed processing mode

enum DistributedPartialResultId

Enumerator
finalResultFromStep2Master 

Result object with singular values and the right orthogonal matrix

enum DistributedPartialResultStep3Id

Enumerator
finalResultFromStep3 

Result object with singular values and the left orthogonal matrix

enum FinalizeOnLocalInputId

Enumerator
inputOfStep3FromStep1 

DataCollection with data transferred from the first step to the third step in the distributed processing mode

inputOfStep3FromStep2 

DataCollection with data transferred from the second step to the third step in the distributed processing mode

enum InputId

Enumerator
data 

Input data table

enum MasterInputId

Enumerator
inputOfStep2FromStep1 

DataCollection with data transferred from the first step to the second step in the distributed processing mode

enum Method

Available methods to compute results of the SVD algorithm

Enumerator
defaultDense 

Default method

enum PartialResultId

Enumerator
outputOfStep1ForStep3 

DataCollection with data computed in the first step to be transferred to the third step in the distributed processing mode

outputOfStep1ForStep2 

DataCollection with data computed in the first step to be transferred to the second step in the distributed processing mode

enum ResultId

Enumerator
singularValues 

Singular values

leftSingularMatrix 

Left orthogonal matrix

rightSingularMatrix 

Right orthogonal matrix

enum SVDResultFormat

Available options to return result matrices

Enumerator
notRequired 

Matrix is not required

requiredInPackedForm 

Matrix in the packed format is required

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