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

References | Namespaces | Classes | Enumerations
Correlation and Variance-Covariance Matrices

Contains classes for computing the correlation or variance-covariance matrix. More...

References

 Batch
 
 Distributed
 
 Online
 

Namespaces

 daal::algorithms::covariance
 Contains classes for computing the correlation or variance-covariance matrix.
 
 daal::algorithms::covariance::interface1
 Contains version 1.0 of Intel(R) Data Analytics Acceleration Library (Intel(R) DAAL) interface.
 

Classes

class  InputIface
 Abstract class that specifies interface for classes that declare input of the correlation or variance-covariance matrix algorithm. More...
 
class  Input
 Input objects of the correlation or variance-covariance matrix algorithm More...
 
class  PartialResult
 Provides methods to access partial results obtained with the compute() method of the correlation or variance-covariance matrix algorithm in the online or distributed processing mode. More...
 
struct  Parameter
 Parameters of the correlation or variance-covariance matrix algorithm. More...
 
struct  OnlineParameter
 Parameters of the correlation or variance-covariance matrix algorithm in the online processing mode. More...
 
class  Result
 Provides methods to access final results obtained with the compute() method of the correlation or variance-covariance matrix algorithm in the batch processing mode. More...
 
class  DistributedInput< step >
 Input parameters of the distributed Covariance algorithm. More...
 
class  DistributedInput< step1Local >
 Input parameters of the distributed Covariance algorithm. Represents inputs of the algorithm on local node. More...
 
class  DistributedInput< step2Master >
 Input parameters of the distributed Covariance algorithm. Represents inputs of the algorithm on master node. More...
 

Enumerations

enum  Method {
  defaultDense = 0, singlePassDense = 1, sumDense = 2, fastCSR = 3,
  singlePassCSR = 4, sumCSR = 5
}
 
enum  InputId { data }
 
enum  PartialResultId { nObservations, crossProduct, sum }
 
enum  ResultId { covariance, correlation = covariance, mean }
 Available identifiers of results of the correlation or variance-covariance matrix algorithm. More...
 
enum  OutputMatrixType { covarianceMatrix, correlationMatrix }
 
enum  MasterInputId { partialResults }
 Available identifiers of master node input arguments of the Covariance algorithm. More...
 

Enumeration Type Documentation

enum InputId

Available identifiers of input objects for the correlation or variance-covariance matrix algorithm

Enumerator
data 

Input data table

enum MasterInputId

Enumerator
partialResults 

Collection of partial results trained on local nodes

enum Method

Available computation methods for variance-covariance or correlation matrix

Enumerator
defaultDense 

Default: performance-oriented method. Works with all types of numeric tables

singlePassDense 

Single-pass: implementation of the single-pass algorithm proposed by D.H.D. West. Works with all types of numeric tables

sumDense 

Precomputed sum: implementation of moments computation algorithm in the case of a precomputed sum. Works with all types of numeric tables

fastCSR 

Fast: performance-oriented method. Works with Compressed Sparse Rows (CSR) numeric tables

singlePassCSR 

Single-pass: implementation of the single-pass algorithm proposed by D.H.D. West. Works with CSR numeric tables

sumCSR 

Precomputed sum: implementation of the algorithm in the case of a precomputed sum. Works with CSR numeric tables

enum OutputMatrixType

Available types of the computed matrix for Covariance

Enumerator
covarianceMatrix 

Variance-Covariance matrix

correlationMatrix 

Correlation matrix

enum PartialResultId

Available identifiers of partial results of the correlation or variance-covariance matrix algorithm

Enumerator
nObservations 

Number of observations processed so far

crossProduct 

Cross-product matrix computed so far

sum 

Vector of sums computed so far

enum ResultId

Enumerator
covariance 

Variance-covariance matrix

correlation 

Correlation matrix

mean 

Vector of means

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