[root] Log of /trunk/math/matrix/inc/TDecompBase.h

Revision 20882 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Mon Nov 19 11:31:26 2007 UTC (7 years, 2 months ago) by rdm
Original Path: trunk/matrix/inc/TDecompBase.h
File length: 4760 byte(s)
Diff to previous 19826

Set property svn:eol-style LF on all source and Makefiles. This should avoid
problems with Win32 line endings ending up in the repository. All MS tools
support LF eols fine.

Revision 13937 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Thu Jan 26 15:16:11 2006 UTC (8 years, 11 months ago) by brun
Original Path: trunk/matrix/inc/TDecompBase.h
File length: 4826 byte(s)
Diff to previous 13929

From Eddy:
1) As pointed out by Philippe, LinkDef.h contained two functions with
the wrong
    signature , resulting in CINT problems as seen in running the
tutorial solveLinear.C

2) A few enum declarations for internal usage haved been moved from
public to protected

Revision 13929 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Wed Jan 25 18:49:03 2006 UTC (9 years ago) by brun
Original Path: trunk/matrix/inc/TDecompBase.h
File length: 4826 byte(s)
Diff to previous 13751

From Eddy:
1) Added to LinkDef.h the NormalEqn function from TDecompChol . They
got
    lost in this file during the transition to the templated version of
the matrix library .

2) Replaced in the templated code everywhere references to DB_EPSILON
and DBL_MIN
    by the templated version through <numeric_limits> .

Revision 13751 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Thu Dec 22 08:41:32 2005 UTC (9 years, 1 month ago) by brun
Original Path: trunk/matrix/inc/TDecompBase.h
File length: 4828 byte(s)
Diff to previous 10329

From Eddy Offermann;

Introduce a templated version of the matrix package.

WARNING: Read carefully the Release Notes of version 5.08
concerning the forward declarations of classes TMatriXF,TMatrixD,TVectorF,TVectorD. Use #include "TMatrixDfwh.h, etc" instead.

If uou update from CVS, do
  cd $ROOTSYS/matrix/src
  rm *.d *.o

Revision 10329 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Sat Oct 16 18:09:17 2004 UTC (10 years, 3 months ago) by brun
Original Path: trunk/matrix/inc/TDecompBase.h
File length: 4823 byte(s)
Diff to previous 9893

From Eddy Offermann:
- added new decomposition class TDecompBK: Bunch-Kaufman algorithm
  this class is designed to decompose real symmetric matrices . Now
  we can guarantee that inverse of a symmetric matrix is again
  symmetric .
- As a consequence of the introduction of TDecompBK, we added
  Invert(..) functionality to TMatrixD/FSym
  Also added InvertFast to TMatrixD/FSym which uses Cramer
  inversion for matrices <= (6 x 6) , resulting in large speedup
  but less accuracy . The necessary routines are encapsulated
  in the new TMatrixDSymCramerInv class .

  Note of caution :
  Unlike the other decompositions, Bunch-Kaufmann does not result
  in a triangular matrix of which the determinant is trivially calculated
  by multiplying the diagonal elements . As a consequence,
  we have to invoke yet another decomposition when the user
  requests the determinant .
  So Invert() will be roughly twice faster than Double_t *det; Invert(det)

- Changed the algoritm for the Cholesky decomposition so that we
  do not need to have a copy of the original matrix around !
  After removal of class member fA, increased the version to 2
- Added sorting of eigen -values/-vectors to TMatrixDEigen .
  They are now sorted according to Re^2+Im^2 of the (possible)
  complex eigenvalues in descending order . TMatrixDSymEigen
  already sorted . Since most users use TMatrixD instead of
  TMatrixDSym, this will ensure backward compatiblity since
  old root contained the symmetric algorithms in TMatrixD .
- Added rank 1 update algorihms : A += alpha * x y^T
  (x and y are vectors)
  to the non-sparse matrix classes (adding this to sparse
  would make this matrix immediately dense :)
- Moved Invert fom TDecompBase to the different decomposition
  classes . Reason is that some classes can decompose A(m,n)
  where m > n . The result of Invert() is then a "pseudo-invert"
  : A_pseudo = (A^T A)^-1 A^T .
- Removed an unfortunate bug in the TMatrixD/FSub class :
  both operators
      void operator+=(Double_t val);
      void operator*=(Double_t val);
  performed actually
      void operator=(Double_t val);
~

Revision 9893 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Fri Sep 3 13:41:35 2004 UTC (10 years, 4 months ago) by brun
Original Path: trunk/matrix/inc/TDecompBase.h
File length: 4904 byte(s)
Diff to previous 9573

From Eddy Offermann
The matrix includes contained circular dependencies (which were of
course
stopped by the #ifdef's) . Situation has been cleaned up by inserting
several forward declarations in the header files and moving some
includes
to the source files . A change in a header file now causes less
compilation
in other directories like hist.
This cleanup made adding an include file in quadp/src/TQpVar necessary
and some additional include files in the tutorials and test files

Revision 9503 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Mon Jul 12 20:00:41 2004 UTC (10 years, 6 months ago) by brun
Original Path: trunk/matrix/inc/TDecompBase.h
File length: 4798 byte(s)
Diff to previous 9181

From Eddy Offermann:
Some code re-organization:

- made TDecompLU::DecomposeLUCrout and TDecompLU::DecomposeLUGauss
protected.
  As a consequence changed TDecompLU::InvertLU so that it now calls
  the decomposition step .
  This change prohibits the user of calling InvertLU without applying
first
  the decomposition step

- Routines in TMatrixD/FCramerInv return now a Bool_t instead of Int_t
.
  More important, these routines check that the matrix is indeed of the
  appropriate shape, since the user could call them directly instead
  of going through TMatrixD/F::InvertFast

- Removed the redundant TDecompBase::MultiSolve(TMatrixDSym &B), it did
  exactly what TDecompBase::MultiSolve(TMatrixD &B) did .

- void (Trans)Solve(TVectorD &b,Int_t ok) and
  void (Trans)Solve(TMatrixDColumn &cb, Int_t ok) routines in TDecomp*

  now make vector b or matrix b in case of an error (when ok was set to
false)

Revision 9181 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Sun Jun 13 14:53:15 2004 UTC (10 years, 7 months ago) by brun
Original Path: trunk/matrix/inc/TDecompBase.h
File length: 4840 byte(s)
Diff to previous 9043

From Eddy Offermann:
1) Initialized fIndex data member of TDecompLU
   in the constructors
2) Added to all TDecompXXX classes, a Print()
   procedure .

Revision 9043 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Thu May 27 20:20:48 2004 UTC (10 years, 7 months ago) by brun
Original Path: trunk/matrix/inc/TDecompBase.h
File length: 4801 byte(s)
Diff to previous 9032

From Eddy Offermann:
The change of the status handling in the decomp classes
introduced a bug . We have to reset the entire
status flag after a call to SetMatrix .

Revision 9032 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Thu May 27 06:39:53 2004 UTC (10 years, 8 months ago) by brun
Original Path: trunk/matrix/inc/TDecompBase.h
File length: 4719 byte(s)
Diff to previous 8892

From Eddy Offermann
Implemented the matrix status through TObject::fBits .
Also removed the fStatus data member of TDecompbase
and use now TObject::fBits .
User can now reset the matrix status through
MakeValid()

Revision 8892 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Wed May 12 10:39:29 2004 UTC (10 years, 8 months ago) by brun
Original Path: trunk/matrix/inc/TDecompBase.h
File length: 4748 byte(s)
Diff to previous 8456

From Eddy Offermann:
1) patch for the matrix directory (cvs diff -a matrix)
2) tar file containing new modules:
  matrix/inc/TDecompSparse.h
  matrix/inc/TMatrixDSparse.h
  matrix/src/TDecompSparse.cxx
  matrix/src/TMatrixDSparse.cxx
3) new stressLinear.cxx

Description of changes:
1) many updates of comments in code
2) possibility to construct random matrices, general
   symmetric and positive definite
3) Introduction of new matrix class: TMatrixDSparse .
   It defines a general sparse matrix in
Harwell-Boeing
   format (sparse structure definition optimized for
   matrix-vector multiplications).
   Like all the other matrix classes, it derives from
   TMatrixDBase
4) Introduction of a new decomposition class:
   TDecompSparse . It allows decomposition (and
   therefore equation solving) of sparse symmetric
   matrices

Revision 8456 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Mon Mar 22 08:34:36 2004 UTC (10 years, 10 months ago) by brun
Original Path: trunk/matrix/inc/TDecompBase.h
File length: 4633 byte(s)
Diff to previous 8189

From Eddy:
Added to the decomposition class constructors in
which just the matrix size is given as argument .
Also a new state (kMatrixSet) was added .
Instead of

TMatrixD a;
TDecompLU lu(a);

one can do

TMatrixD a;
TDecompLU lu(n);
lu.SetMatrix(a);

The SetMatrix command resizes, if necessary, the
data members in the decomposition class .

This set up willl be usefull when one wants to
decompose
matrices in a loop

Revision 8118 - (view) (download) (as text) (annotate) - [select for diffs]
Modified Wed Feb 4 17:12:44 2004 UTC (10 years, 11 months ago) by brun
Original Path: trunk/matrix/inc/TDecompBase.h
File length: 4345 byte(s)
Diff to previous 8112

From Eddy Offermann:


1. LinkDef.h : Added NormalEqn friend functions from
               TDecompChol class
2. TMatrixDBase: GetDecompMatrix() returns a TMatrixD
                instead of TMatrixDBase
                Added Solve and TransSolve functions
                that return the solution (others
replace
                the input argument which is
inconvinient
                when output vector size is smaller
than
                input (in ncase of fit).

                Also added Invert that returns matrix
                instead of replacing the input
argument.
                This is convinient for a generalized
                (pseudo) inverse: in (m x n) out (n x
m)

3. TDecompChol: Solve/TrnasSolve had incorrect
forward/
                backward substitution function ! Tests
                have been added to stressLinear to
check

                Added NormlEqn solver which can handle
                weights in data

4. TDecompSVD:  Bug in Solve function ! index was
stop-
                ping one too early . Was not noticed
by
                stressLinear because the linear
equation
                test wasnumerically to difficult
                causing accuracy tolerances to be not
                strict enough (was 0.005 , is now
                DBL_EPSILON on a diagonal dominant
                matrix)

Revision 8013 - (view) (download) (as text) (annotate) - [select for diffs]
Added Sun Jan 25 20:33:32 2004 UTC (11 years ago) by brun
Original Path: trunk/matrix/inc/TDecompBase.h
File length: 4113 byte(s)

New Linear Algebra package from Eddy Offermann.
This new package reimplements the previous classes TMatrix and TMatrixD.
The new classes should be back compatible with the previous version except
the function GetElements.
New classes have been introduced for symmetric matrices,
lazy matrices.
New algorithms (LU, SVD) have been introduced.
A new test suite test/stressLinear.cxx is introduced.
A complete description of this package will be posted in the coming days.
The classes are well documented in the implementation headers.

This form allows you to request diffs between any two revisions of this file. For each of the two "sides" of the diff, enter a numeric revision.

Links to HEAD:	(view) (download) (as text) (annotate)
Sticky Revision: