octave-patch-tracker
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Octave-patch-tracker] [patch #8974] GSOC16 - Improve iterative methods


From: Cristiano Dorigo
Subject: [Octave-patch-tracker] [patch #8974] GSOC16 - Improve iterative methods for sparse linear systems
Date: Sat, 09 Apr 2016 17:31:12 +0000
User-agent: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:45.0) Gecko/20100101 Firefox/45.0

URL:
  <http://savannah.gnu.org/patch/?8974>

                 Summary: GSOC16 - Improve iterative methods for sparse linear
systems
                 Project: GNU Octave
            Submitted by: cristianodorigo
            Submitted on: Sat 09 Apr 2016 05:31:11 PM GMT
                Category: None
                Priority: 5 - Normal
                  Status: None
                 Privacy: Public
             Assigned to: None
        Originator Email: 
             Open/Closed: Open
         Discussion Lock: Any

    _______________________________________________________

Details:

Dear all,
I submitted a proposal for the GSOC16 project mentioned in the summary.
In the attached patch there is an idea of how to work for the first part of
the project. 

I wrote the private function __default__input__.m that performs as follows:
[varargout] = __default__input.m__(def_val, varargin) and verifies the values
of the varargin variables, if they are filled then varargout{i} = varargin{i},
otherwise if they miss or are empty thevarargout{i} = def_val{i}, where
def_val is a cell array that contains the defaul desired values.

Then I substituted the internal input control of restart,rtol,maxit in the
gmres code using the previous function and the same for tol,maxit,x0 in pcg.

I made the function __default__input__ more general as possible because the
variables that need a control in gmres and pcg are different (in this specific
case gmres has the restart values that pcg hasn't).

The next step could be at first to improve and/or fix this function if there
are some useless things and/or missing features , then to extend this input
check for the other algorithms mentioned in the GSOC project (pcr, bicg,
bicgstab, cgs, and qmr). Then I think that can be useful another private
function that check if are consistent the matrix A, the preconditioners M1, M2
and the right-hand-side vector b.

Please don't hesistate to give me some advices, observations or tell me if I'm
working in an uncorrected road.








    _______________________________________________________

File Attachments:


-------------------------------------------------------
Date: Sat 09 Apr 2016 05:31:11 PM GMT  Name: input_control_sparse.cset  Size:
7kB   By: cristianodorigo

<http://savannah.gnu.org/patch/download.php?file_id=36878>

    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/patch/?8974>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/




reply via email to

[Prev in Thread] Current Thread [Next in Thread]