[Ipopt] Negative Lagrange multipliers?

[Ipopt User]

If you don't want to modify any source code, I think the easiest way is to
increase the log level.

On Wed, Sep 7, 2016 at 11:49 AM, Chunhua Men <chhmen at gmail.com> wrote:

> BTW, is it possible to know Lagrange multipliers for each iteration
> (instead of at the end of optimization)? Since I use "limited-memory" as
> the Hessian approximation, the "eval_h" is not used, then I could not
> get  Lagrange-multipliers updates during optimization process.
>
> Thanks again, Chunhua
>
> On Wed, Sep 7, 2016 at 8:42 AM, Chunhua Men <chhmen at gmail.com> wrote:
>
>> Got it - thanks!
>>
>> On Wed, Sep 7, 2016 at 8:14 AM, Ipopt User <ipoptuser at gmail.com> wrote:
>>
>>> IPOPT treats all constraints as equalities with slacks, see (2b) in
>>> http://www.optimization-online.org/DB_FILE/2004/03/836.pdf
>>>
>>> On Tue, Sep 6, 2016 at 8:21 PM, Chunhua Men <chhmen at gmail.com> wrote:
>>>
>>>> Hi, I used IPOPT 3.12.4 to solve a constrained model, and chose
>>>> "limited-memory" as the Hessian approximation. My coding environment is
>>>> visual studio C++.  IPOPT converged to optimal solution.  I wanted to know
>>>> the Lagrange multiplier for each constraint, so I printed out "lambda" in
>>>> "finalize_solution". However, many of them were negative. Did I miss
>>>> anything?
>>>>
>>>> Thanks, Chunhua
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> Ipopt mailing list
>>>> Ipopt at list.coin-or.org
>>>> https://urldefense.proofpoint.com/v2/url?u=http-3A__list.coi
>>>> n-2Dor.org_mailman_listinfo_ipopt&d=CwICAg&c=Ngd-ta5yRYsqeUs
>>>> EDgxhcqsYYY1Xs5ogLxWPA_2Wlc4&r=Krbuvr0BfvPSQKvQuiGrWIUfVTj8k
>>>> ednZerudAN3FWE&m=a_2DH1-JLrp3dNBIbx05pyyLwb4LVhtnTgXAs8ZvlR0
>>>> &s=Pl6RJRxVCqHGy-wEIKdnxo1N2UcAlQGXpBY_LA2CTHY&e=
>>>>
>>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.coin-or.org/pipermail/ipopt/attachments/20160907/530841fb/attachment.html>