Error:unable to factorize basis matrix

[Alexander Schnell]

Hello,

i`m using pymathprog 0.3 and when trying to solve a big Binary Integer Program, i get the following error message and the evaluation stops:

: obj =  2.719712443e+007  infeas = 9.756e+001 (1409)
  80200: obj =  2.719712443e+007  infeas = 9.756e+001 (1409)
  80400: obj =  2.719712443e+007  infeas = 9.756e+001 (1409)
  80600: obj =  2.719712443e+007  infeas = 9.756e+001 (1409)
  80800: obj =  2.719712443e+007  infeas = 9.756e+001 (1400)
  81000: obj =  2.723652533e+007  infeas = 8.235e+001 (1376)
  81200: obj =  2.723652533e+007  infeas = 8.235e+001 (1369)
Warning: numerical instability (primal simplex, phase I)
  81259: obj =  2.723652533e+007  infeas = 8.235e+001 (1368)
Warning: numerical instability (primal simplex, phase I)
  81299: obj =  2.723652533e+007  infeas = 8.235e+001 (1368)
  81400: obj =  2.723810723e+007  infeas = 7.967e+001 (1361)
Error: unable to factorize the basis matrix (1)
Sorry, basis recovery procedure not implemented yet
ios_driver: unable to solve current LP relaxation; glp_simplex returned 5
+ 81411: mip =     not found yet >=  2.618187908e+007        (341; 1)
glp_intopt: cannot solve current LP relaxation

What does this mean? Has it something to do with the numerical instability warning?

Best Regards,
Alex
_______________________________________________________________
Neu: WEB.DE DSL bis 50.000 kBit/s und 200,- Euro Startguthaben!
http://produkte.web.de/go/02/






_______________________________________________
Help-glpk mailing list
Help-glpk@...
http://lists.gnu.org/mailman/listinfo/help-glpk

[Yingjie Lan]

Hi Alex,

All I can say given the information you provided is that this seems to be a problem with the GLPK solver, not the pymathprog, because the model is already committed to the GLPK solver. Exactly why this message pops out from GLPK I don't know for sure.

Yingjie

--- On Wed, 10/28/09, Alexander Schnell <lexei@...> wrote:

> From: Alexander Schnell <lexei@...>
> Subject: [Help-glpk] Error:unable to factorize basis matrix
> To: help-glpk@...
> Date: Wednesday, October 28, 2009, 3:22 PM
> Hello,
>
> i`m using pymathprog 0.3 and when trying to solve a big
> Binary Integer Program, i get the following error message
> and the evaluation stops:
>
> : obj =  2.719712443e+007  infeas = 9.756e+001
> (1409)
>   80200: obj =  2.719712443e+007  infeas =
> 9.756e+001 (1409)
>   80400: obj =  2.719712443e+007  infeas =
> 9.756e+001 (1409)
>   80600: obj =  2.719712443e+007  infeas =
> 9.756e+001 (1409)
>   80800: obj =  2.719712443e+007  infeas =
> 9.756e+001 (1400)
>   81000: obj =  2.723652533e+007  infeas =
> 8.235e+001 (1376)
>   81200: obj =  2.723652533e+007  infeas =
> 8.235e+001 (1369)
> Warning: numerical instability (primal simplex, phase I)
>   81259: obj =  2.723652533e+007  infeas =
> 8.235e+001 (1368)
> Warning: numerical instability (primal simplex, phase I)
>   81299: obj =  2.723652533e+007  infeas =
> 8.235e+001 (1368)
>   81400: obj =  2.723810723e+007  infeas =
> 7.967e+001 (1361)
> Error: unable to factorize the basis matrix (1)
> Sorry, basis recovery procedure not implemented yet
> ios_driver: unable to solve current LP relaxation;
> glp_simplex returned 5
> + 81411: mip =     not found yet
> >=  2.618187908e+007       
> (341; 1)
> glp_intopt: cannot solve current LP relaxation
>
> What does this mean? Has it something to do with the
> numerical instability warning?
>
> Best Regards,
> Alex
> _______________________________________________________________
> Neu: WEB.DE DSL bis 50.000 kBit/s und 200,- Euro
> Startguthaben!
> http://produkte.web.de/go/02/
>
>
>
>
>
>
> _______________________________________________
> Help-glpk mailing list
> Help-glpk@...
> http://lists.gnu.org/mailman/listinfo/help-glpk
>

_______________________________________________
Help-glpk mailing list
Help-glpk@...
http://lists.gnu.org/mailman/listinfo/help-glpk