help-glpk
[Top][All Lists]
Advanced

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

[Help-glpk] lpx_integer confusion


From: dtucker
Subject: [Help-glpk] lpx_integer confusion
Date: Mon, 12 Jul 2004 16:36:05 -0500


Hello, I am attempting to solve a binary integer programming problem.  I call the lpx_integer function, but first I call the lpx_simplex for an initial optimal solution.  For some reason, occasionally, the lpx_simplex function seems to get some information in it that makes all further calls to it useless, even when the coefficients for A and f have changed.  The messages I continually get from lpx_simplex after this point are:

spx_invert: trying to factorize the basis using threshold tolerance 0.3
spx_invert: trying to factorize the basis using threshold tolerance 0.7
spx_invert: the basis matrix is singular
spx_simplex: initial basis is singular

So, if lpx_simplex does not return LPX_E_OK, I call lpx_interior instead.  lpx_interior seems to do fine, and I get the following response:

lpx_interior: original LP problem has 66 rows and 24 columns
lpx_interior: transformed LP problem has 90 rows and 114 columns
Factorizing S = A*A' symbolically...
nz(A) = 743
nz(S) = 2220 (upper triangular part)
nz(U) = 2307
Guessing initial point...
Optimization begins...
  0: F =  -9.406178874e+00; rpi =  1.5e+00; rdi =  1.7e+00; gap =  1.0e-00
  1: F =  -4.534167157e+00; rpi =  2.3e-01; rdi =  3.7e-01; gap =  4.6e+00
  2: F =  -6.075538098e+00; rpi =  4.7e-02; rdi =  1.2e-01; gap =  1.7e+00
  3: F =  -6.809413830e+00; rpi =  1.4e-02; rdi =  1.2e-02; gap =  3.7e-01
  4: F =  -6.933764409e+00; rpi =  1.5e-03; rdi =  1.4e-03; gap =  4.1e-02
  5: F =  -6.950002036e+00; rpi =  1.5e-04; rdi =  1.4e-04; gap =  4.1e-03
  6: F =  -6.951630025e+00; rpi =  1.5e-05; rdi =  1.4e-05; gap =  4.1e-04
  7: F =  -6.951792828e+00; rpi =  1.5e-06; rdi =  1.4e-06; gap =  4.1e-05
  8: F =  -6.951809109e+00; rpi =  1.5e-07; rdi =  1.4e-07; gap =  4.1e-06
  9: F =  -6.951810737e+00; rpi =  1.5e-08; rdi =  1.4e-08; gap =  4.1e-07
 10: F =  -6.951810900e+00; rpi =  1.5e-09; rdi =  1.4e-09; gap =  4.1e-08
 11: F =  -6.951810916e+00; rpi =  1.5e-10; rdi =  1.4e-10; gap =  4.1e-09
OPTIMAL SOLUTION FOUND

Now, however, if I attempt to call lpx_integer, I get the following messages:

lpx_integer: optimal solution of LP relaxation required
Unable to start lpx_integer search.

Should'nt there already be an optimal solution stored in the problem object from the successful execution of lpx_interior?  Why is lpx_integer unable to start, if so?  If not, why does lpx_simplex get locked into this mode, even when new input is passed to it?

Don

[Please respond to me as well as the list.  I am not subscribed to the list.]
reply via email to

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