[Dip] examples no longer working

Matthew Galati matthew.galati at gmail.com
Mon Jan 10 18:39:41 EST 2011


I've been using __DECOMP_IP_CPX__ lately and everything has been working
fine (when using --with-ip-solver=cpx).

I found the issue. __DECOMP_IP_CBC__ is no longer being defined by default.
It is odd, because __DECOMP_LP_CLP__ is defined as normal.

So, basically, DIP is building with no IP solver which is causing failures
on all the tests.

Something must have changed recently in the build tools or DIP build files.
Ted is in charge of that stuff, so I will need his help on this one. Ted -
can you take a look?

Thanks,
Matt



On Mon, Jan 10, 2011 at 2:51 PM, Matthew Galati <matthew.galati at gmail.com>wrote:

> Make test fails for me too (at trunk). I'll take a look ASAP.
>
>   unitTest: ../../../Dip/src/DecompAlgo.cpp:748: virtual void
> DecompAlgo::createMa
> sterProblem(DecompVarList&): Assertion `initVars.size() > 0' failed.
>
> Thanks,
> Matt
>
>
>
>
> On Sun, Jan 9, 2011 at 5:33 AM, kim <kim at hamilton-vejlin.dk> wrote:
>
>> For the newer versions of DIP ("0.81.1", "0.81.2" and "trunk") I can no
>> longer get the examples to work.
>>
>> After installation, when I run examples they return simply:
>> "Alps0240I Processed 0 nodes, has 0 nodes, best relaxed 1e+75, best
>> feasible 1e+75
>>
>> Alps0202I Problem is infeasible. 1 nodes processed, 0 nodes left
>> Alps0264I No solution found
>> "
>>
>> This is the output from "make test" after installation as well as the
>> "MILP" and "MILPBlock" examples. I cannot get "GAP" and "MMKP" to
>> compile at all.
>>
>> My install sequence, in case I'm doing something wrong:
>> 1) svn checkout https://projects.coin-or.org/svn/Dip/releases/0.81.2
>> Dip-0.81.2
>> 2) cd Dip-0.81.2
>> 3) mkdir build
>> 4) cd build
>> 5) ../configure
>> 6) make
>> 7) make install
>> 8) make test (this is the first time an example unexpectedly returns "no
>> feasible solution")
>> 9) cd Dip/examples/MILPBlock
>> 10) sh data_extract.sh
>> 11) make
>> 12) ./decomp_milpblock --param milpblock.parm (this too gives a "no
>> solution" result)
>>
>> I have since tried with different instances and with different examples
>> with the same result.
>>
>> - Kim, University of Copenhagen
>>
>>
>> _______________________________________________
>> Dip mailing list
>> Dip at list.coin-or.org
>> http://list.coin-or.org/mailman/listinfo/dip
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://list.coin-or.org/pipermail/dip/attachments/20110110/c43ebad9/attachment.html 


More information about the Dip mailing list