[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
re: [Help-glpk] Re: computational efforts, message level and KKT
From: |
Robbie Morrison |
Subject: |
re: [Help-glpk] Re: computational efforts, message level and KKT |
Date: |
Thu, 2 Dec 2010 23:31:19 +1300 (NZDT) |
User-agent: |
SquirrelMail/1.4.17 |
Hi all
>> >> int niter = lpx_get_int_parm(d_prob, LPX_K_ITCNT); // lpx is correct
>> >>
>> >> Are you sure the call is broken? But I agree, it would
>> >> be better if the API was updated.
>> >
>> > Last time I checked it always returned zero. Maybe I messed up
something...
>>
>> I just double checked with 4.44. You are correct. Zero iterations!
>> That is relatively recent I think.
>
> Just checked the code and found a bug. To fix it please remove the
> following line in the local routine reset_parms, file src/glplpx01.c:
>
> lp->it_cnt = 0;
Just tested the given patch. Before and after:
simplex iterations : 0
simplex iterations : 10
Just one small word of caution. Make sure you test
this patch on a problem that is sufficiently complex to
generate some iterations and cannot simply be solved by
the presolver. :-)
Great! Thanks Andrew.
---
Robbie Morrison
PhD student -- policy-oriented energy system simulation
Technical University of Berlin (TU-Berlin), Germany
University email (redirected) : address@hidden
Webmail (preferred) : address@hidden
[from Webmail client]