GIRAFFE Pipeline Reference Manual

Matrix Utilities

Functions

cxdouble giraffe_matrix_sigma_mean (const cpl_matrix *matrix, cxdouble mean)
 Compute sigma of matrix elements, with a given mean value.
 
cxdouble giraffe_matrix_sigma_fit (const cpl_matrix *matrix, const cpl_matrix *matrix_fit)
 Compute sigma of matrix fit.
 
cpl_image * giraffe_matrix_create_image (const cpl_matrix *matrix)
 Converts a matrix into an image.
 
cxint giraffe_matrix_sort (cpl_matrix *mA)
 Sort in place the matrix elements in ascending order.
 
cpl_matrix * giraffe_matrix_leastsq (const cpl_matrix *mA, const cpl_matrix *mB)
 Computes the solution of an equation using a pseudo-inverse.
 
cpl_matrix * giraffe_matrix_solve_cholesky (const cpl_matrix *A, const cpl_matrix *b, const cpl_matrix *Cb, cpl_matrix *Cx)
 Solve a linear system using the Cholesky decomposition.
 
cxint giraffe_matrix_clear (cpl_matrix *matrix)
 Set all elements of a matrix to zero.
 
void giraffe_matrix_dump (const cpl_matrix *matrix, cxint max_rows)
 Output a maximum number of rows of the input matrix.
 

Detailed Description

TBD

Function Documentation

cxint giraffe_matrix_clear ( cpl_matrix *  matrix)

Set all elements of a matrix to zero.

Parameters
matrixThe matrix to update.
Returns
The function returns 0 on success, or a non-zero value otherwise.

The function replaces the value of each matrix element of matrix with 0.

Definition at line 804 of file gimatrix.c.

cpl_image* giraffe_matrix_create_image ( const cpl_matrix *  matrix)

Converts a matrix into an image.

Parameters
matrixThe matrix to convert.
Returns
The newly image created from the matrix, or NULL if the image cannot be created.

The function converts a CPL matrix into a CPL image. The matrix elements are copied into the image pixel buffer.

Definition at line 345 of file gimatrix.c.

void giraffe_matrix_dump ( const cpl_matrix *  matrix,
cxint  max_rows 
)

Output a maximum number of rows of the input matrix.

Parameters
matrixMatrix to print.
max_rowsMaximum number of rows to print.
Returns
Nothing.

Function prints part of a matrix. Printed are all columns, but only up to max_rows, number of rows. Each column has the column number printed as a header.

Note
This function should be moved to CPL proper...

Definition at line 844 of file gimatrix.c.

Referenced by giraffe_slitgeometry_print(), and giraffe_wcalsolution_dump().

cpl_matrix* giraffe_matrix_leastsq ( const cpl_matrix *  mA,
const cpl_matrix *  mB 
)

Computes the solution of an equation using a pseudo-inverse.

Parameters
mAleft hand side [nr, nc] coefficients matrix.
mBright hand side [nr, nc] matrix of the equation.
Returns
The function returns the solution vector as a [nr, 1] column matrix.

The function solves the linear system

mX * mA = mB

for the unknown column matrix mX. The solution is given by the expression

mX = mB * transpose(mA) * inverse(mA * transpose(mA))

The solution matrix is solving the equation according to a least-squares criterion. To destroy this matrix the function cpl_matrix_delete() should be used.

Note
This function works properly only in the case all the elements of the input matrix do not contain garbage (such as NaN or infinity).

Definition at line 511 of file gimatrix.c.

cxdouble giraffe_matrix_sigma_fit ( const cpl_matrix *  matrix,
const cpl_matrix *  matrix_fit 
)

Compute sigma of matrix fit.

Parameters
matrixOriginal matrix
matrix_fitMatrix to compare
Returns
Sigma of fit

Function computes sigma of fit between two matrixes

Note
This function may be moved to CPL

Definition at line 283 of file gimatrix.c.

cxdouble giraffe_matrix_sigma_mean ( const cpl_matrix *  matrix,
cxdouble  mean 
)

Compute sigma of matrix elements, with a given mean value.

Parameters
matrixInput matrix
meanValue to be used as mean value during calculation
Returns
Sigma of matrix

Function computes sigma of matrix elements, with a given mean value

Note
This function might be moved to CPL.

Definition at line 237 of file gimatrix.c.

cpl_matrix* giraffe_matrix_solve_cholesky ( const cpl_matrix *  A,
const cpl_matrix *  b,
const cpl_matrix *  Cb,
cpl_matrix *  Cx 
)

Solve a linear system using the Cholesky decomposition.

Parameters
AThe design matrix of the system
bThe right hand side of the system
CbThe covariance matrix of the right hand side.
CxThe covariance matrix of the solution vector.
Returns
The function returns the solution vector of the linear system, or NULL in case of an error.

The function solves the over-determined linear system Ax = b, where A is an m times n matrix, b is an m times 1 matrix (a column vector), and x is the n times 1 solution matrix.

A covariance matrix Cb of the right hand side vector b may be given. If it is not NULL, it is when the system is solved. It has to be a m times m matrix.

Similarly if the covariance matrix of the solution vector x, Cx, is not NULL, the computed covariance matrix of the solution is stored there. It must be a n times n matrix. Any contents of this matrix is overwritten, and its contents is undetermined if the function returns with an error.

Definition at line 587 of file gimatrix.c.

cxint giraffe_matrix_sort ( cpl_matrix *  mA)

Sort in place the matrix elements in ascending order.

Parameters
mAThe matrix to sort.
Returns
0 on success, -1 on failure

The input matrix mA is sorted in place. On the successfull return the matrix mA contains the original values sorted in ascending order.

Definition at line 388 of file gimatrix.c.


This file is part of the GIRAFFE Pipeline Reference Manual 2.12.
Documentation copyright © 2002-2006 European Southern Observatory.
Generated on Mon Mar 24 2014 11:43:53 by doxygen 1.8.2 written by Dimitri van Heesch, © 1997-2004