org.apache.spark.ml.linalg

Class BLAS

Object

org.apache.spark.ml.linalg.BLAS

public class BLAS
extends Object

BLAS routines for MLlib's vectors and matrices.

Constructor Summary

Constructors

Constructor and Description
BLAS()

Method Summary

Modifier and Type	Method and Description
static void	axpy(double a, Vector x, Vector y) y += a * x
static void	copy(Vector x, Vector y) y = x
static double	dot(Vector x, Vector y) dot(x, y)
static void	dspmv(int n, double alpha, DenseVector A, DenseVector x, double beta, DenseVector y) y := alpha*A*x + beta*y
static void	gemm(double alpha, Matrix A, DenseMatrix B, double beta, DenseMatrix C) C := alpha * A * B + beta * C
static void	gemm(double alpha, Matrix A, DenseMatrix B, double beta, double[] CValues) CValues[0: A.numRows * B.numCols] := alpha * A * B + beta * CValues[0: A.numRows * B.numCols]
static void	gemv(double alpha, Matrix A, double[] x, double beta, double[] y) y[0: A.numRows] := alpha * A * x[0: A.numCols] + beta * y[0: A.numRows]
static void	gemv(double alpha, Matrix A, Vector x, double beta, DenseVector y) y := alpha * A * x + beta * y
static void	scal(double a, Vector x) x = a * x
static void	spr(double alpha, Vector v, DenseVector U) Adds alpha * x * x.t to a matrix in-place.
static void	spr(double alpha, Vector v, double[] U) Adds alpha * v * v.t to a matrix in-place.
static void	syr(double alpha, Vector x, DenseMatrix A) A := alpha * x * x^T^ + A

Methods inherited from class Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

BLAS

public BLAS()

Method Detail

axpy

public static void axpy(double a,
                        Vector x,
                        Vector y)

y += a * x

Parameters:

a - (undocumented)

x - (undocumented)

y - (undocumented)

dot

public static double dot(Vector x,
                         Vector y)

dot(x, y)

Parameters:

x - (undocumented)

y - (undocumented)

Returns:

(undocumented)

copy

public static void copy(Vector x,
                        Vector y)

y = x

Parameters:

x - (undocumented)

y - (undocumented)

scal

public static void scal(double a,
                        Vector x)

x = a * x

Parameters:

a - (undocumented)

x - (undocumented)

spr

public static void spr(double alpha,
                       Vector v,
                       DenseVector U)

Adds alpha * x * x.t to a matrix in-place. This is the same as BLAS's ?SPR.

Parameters:

U - the upper triangular part of the matrix in a DenseVector(column major)

alpha - (undocumented)

v - (undocumented)

dspmv

public static void dspmv(int n,
                         double alpha,
                         DenseVector A,
                         DenseVector x,
                         double beta,
                         DenseVector y)

y := alpha*A*x + beta*y

Parameters:

n - The order of the n by n matrix A.

A - The upper triangular part of A in a DenseVector (column major).

x - The DenseVector transformed by A.

y - The DenseVector to be modified in place.

alpha - (undocumented)

beta - (undocumented)

spr

public static void spr(double alpha,
                       Vector v,
                       double[] U)

Adds alpha * v * v.t to a matrix in-place. This is the same as BLAS's ?SPR.

Parameters:

U - the upper triangular part of the matrix packed in an array (column major)

alpha - (undocumented)

v - (undocumented)

syr

public static void syr(double alpha,
                       Vector x,
                       DenseMatrix A)

A := alpha * x * x^T^ + A

Parameters:

alpha - a real scalar that will be multiplied to x * x^T^.

x - the vector x that contains the n elements.

A - the symmetric matrix A. Size of n x n.

gemm

public static void gemm(double alpha,
                        Matrix A,
                        DenseMatrix B,
                        double beta,
                        DenseMatrix C)

C := alpha * A * B + beta * C

Parameters:

alpha - a scalar to scale the multiplication A * B.

A - the matrix A that will be left multiplied to B. Size of m x k.

B - the matrix B that will be left multiplied by A. Size of k x n.

beta - a scalar that can be used to scale matrix C.

C - the resulting matrix C. Size of m x n. C.isTransposed must be false.

gemm

public static void gemm(double alpha,
                        Matrix A,
                        DenseMatrix B,
                        double beta,
                        double[] CValues)

CValues[0: A.numRows * B.numCols] := alpha * A * B + beta * CValues[0: A.numRows * B.numCols]

Parameters:

alpha - a scalar to scale the multiplication A * B.

A - the matrix A that will be left multiplied to B. Size of m x k.

B - the matrix B that will be left multiplied by A. Size of k x n.

beta - a scalar that can be used to scale matrix C.

CValues - the values of matrix C. C.isTransposed is supposed to be false.

gemv

public static void gemv(double alpha,
                        Matrix A,
                        double[] x,
                        double beta,
                        double[] y)

y[0: A.numRows] := alpha * A * x[0: A.numCols] + beta * y[0: A.numRows]

Parameters:

alpha - (undocumented)

A - (undocumented)

x - (undocumented)

beta - (undocumented)

y - (undocumented)

gemv

public static void gemv(double alpha,
                        Matrix A,
                        Vector x,
                        double beta,
                        DenseVector y)

y := alpha * A * x + beta * y

Parameters:

alpha - a scalar to scale the multiplication A * x.

A - the matrix A that will be left multiplied to x. Size of m x n.

x - the vector x that will be left multiplied by A. Size of n x 1.

beta - a scalar that can be used to scale vector y.

y - the resulting vector y. Size of m x 1.