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

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

[Octave-bug-tracker] [bug #57028] Interior-point solver in glpk generate


From: Robert Taylor
Subject: [Octave-bug-tracker] [bug #57028] Interior-point solver in glpk generates a message that cannot be turned off.
Date: Wed, 9 Oct 2019 20:45:12 -0400 (EDT)
User-agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:69.0) Gecko/20100101 Firefox/69.0

URL:
  <https://savannah.gnu.org/bugs/?57028>

                 Summary: Interior-point solver in glpk generates a message
that cannot be turned off.
                 Project: GNU Octave
            Submitted by: btaylor
            Submitted on: Thu 10 Oct 2019 12:45:10 AM UTC
                Category: Octave Function
                Severity: 3 - Normal
                Priority: 5 - Normal
              Item Group: Other
                  Status: None
             Assigned to: None
         Originator Name: btaylor
        Originator Email: 
             Open/Closed: Open
         Discussion Lock: Any
                 Release: 5.1.0
        Operating System: Microsoft Windows

    _______________________________________________________

Details:

The interior-point solver in glpk generates a "scaling" message that cannot be
turned off (neither the .msglev nor .scale parameters for glpk curtail this
behavior). 

The problem appears in Octave versions 4.2.2 to 5.5.0.

Sample code:


f = randn(1000, 1);
I = eye(5);
A = repmat(I, 1, 200);
b = ones(1, 5);
lb = zeros(1000, 1);
ctype = repmat('S', 1, 5);
vartype = repmat('C', 1, 1000);
params.lpsolver = 2;
[~, fmin] = glpk(f, A, b, lb, [], ctype, vartype, 1, params);
disp(fmin);


Sample output is:

Scaling...
 A: min|aij| = 1.000e+000  max|aij| = 1.000e+000  ratio = 1.000e+000
Problem data seem to be well scaled
EQ: min|aij| = 1.000e+000  max|aij| = 1.000e+000  ratio = 1.000e+000
-13.893

I understand that, during troubleshooting, this message might be useful but it
should be an optional message.

Incidentally, the performance of the interior-point solver is amazing. I am
working on problems with over 1 million variables that are being called
hundreds of times and, aside from a lot of Command Window output, I am getting
excellent results.





    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?57028>

_______________________________________________
  Message sent via Savannah
  https://savannah.gnu.org/




reply via email to

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