<div dir="ltr"><div dir="ltr"><div dir="ltr">Dear All,<div><br></div><div>With the options below I could set mu_target without the hessian of the lagrangian. Other options maybe won't work.</div><div><br></div><div><div>addOption(prob, "hessian_approximation", "limited-memory")</div><div>addOption(prob, "mu_target", 1.0)</div><div>addOption(prob, "mehrotra_algorithm", "yes")</div></div><div><br></div><div>Best,</div></div></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">Em qui, 17 de jan de 2019 às 11:28, Pedro Borges <<a href="mailto:pedro.borges.melo@gmail.com">pedro.borges.melo@gmail.com</a>> escreveu:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="auto">Just ignore this email. I had forgotten your reply. Thanks. </div><br><div class="gmail_quote"><div dir="ltr">Em Qui, 17 de jan de 2019 10:34, Pedro Borges <<a href="mailto:pedro.borges.melo@gmail.com" target="_blank">pedro.borges.melo@gmail.com</a>> escreveu:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="auto">Dear Professor,<div dir="auto"><br></div><div dir="auto">I am really looking forward to your reply. </div><div dir="auto"><br></div><div dir="auto">Thanks,</div></div><br><div class="gmail_quote"><div dir="ltr">Em Qua, 16 de jan de 2019 06:15, Stefan Vigerske <<a href="mailto:stefan@math.hu-berlin.de" target="_blank">stefan@math.hu-berlin.de</a>> escreveu:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Hi,<br>
<br>
I meant that I don't see where Ipopt would force mu_target to be 0 in <br>
the Ipopt code.<br>
<br>
Using the limited-memory Hessian approximation changes the default for <br>
mu_strategy to adaptive. Maybe that strategy takes mu_target into <br>
account differently, though I don't see that it sets mu_target = 0.<br>
If mu_strategy is adaptive, then option mu_oracle also plays a role.<br>
<br>
And then there is also an option mu_min.<br>
<br>
Stefan<br>
<br>
On 1/15/19 2:01 PM, Pedro Borges wrote:<br>
> Hi!<br>
> <br>
> Check the code below.<br>
> <br>
> Should I consider this is a problem on the julia side?<br>
> <br>
> using JuMP<br>
> using Ipopt<br>
> <br>
> ###############################################################<br>
> <br>
> #=<br>
> ipoptSolver = IpoptSolver(mu_target = 1.0)<br>
> m = Model(solver=ipoptSolver)<br>
> @variable(m, x >= 0)<br>
> @NLobjective(m, :Min, x)<br>
> solve(m)<br>
> getvalue(x)<br>
> =#<br>
> <br>
> ###############################################################<br>
> <br>
> #=<br>
> ipoptSolver = IpoptSolver(mu_target = 1.0)<br>
> m = Model(solver=ipoptSolver)<br>
> @variable(m, x >= 0)<br>
> JuMP.register(m, :my_objective, 1, (x) -> x, (x) -> 1.0, (x) -> 0.0)<br>
> JuMP.setNLobjective(m, :Min, :(my_objective($(x…))))<br>
> solve(m)<br>
> getvalue(x)<br>
> =#<br>
> <br>
> ###############################################################<br>
> <br>
> #=<br>
> function val(x1, x2)<br>
> return x1 + x2<br>
> end<br>
> <br>
> function grad(result, x1, x2)<br>
> result[1] = 1.0<br>
> result[2] = 1.0<br>
> end<br>
> <br>
> ipoptSolver = IpoptSolver(mu_target = 1.0)<br>
> m = Model(solver=ipoptSolver)<br>
> @variable(m, x[1:2] >= 0)<br>
> JuMP.register(m, :my_objective, 2, val, grad, autodiff=false)<br>
> JuMP.setNLobjective(m, :Min, :(my_objective($(x…))))<br>
> solve(m)<br>
> getvalue(x)<br>
> =#<br>
> <br>
> Em Ter, 15 de jan de 2019 12:17, Stefan Vigerske <<a href="mailto:stefan@math.hu-berlin.de" rel="noreferrer" target="_blank">stefan@math.hu-berlin.de</a>><br>
> escreveu:<br>
> <br>
>> Hi,<br>
>><br>
>> I do not see where this is happening in the code.<br>
>><br>
>> Stefan<br>
>><br>
>> On 1/9/19 2:28 PM, Pedro Borges wrote:<br>
>>> Hello guys!<br>
>>><br>
>>> Is it possible to use mu_target != 0 without proving the hessian of the<br>
>>> lagrangian?<br>
>>> If I don't provide the hessian Ipopt sets mu_target = 0 against my will.<br>
>>><br>
>>> Best,<br>
>>><br>
>>><br>
>>> _______________________________________________<br>
>>> Ipopt mailing list<br>
>>> <a href="mailto:Ipopt@list.coin-or.org" rel="noreferrer" target="_blank">Ipopt@list.coin-or.org</a><br>
>>> <a href="https://list.coin-or.org/mailman/listinfo/ipopt" rel="noreferrer noreferrer" target="_blank">https://list.coin-or.org/mailman/listinfo/ipopt</a><br>
>>><br>
>><br>
>><br>
> <br>
<br>
</blockquote></div></blockquote></div>
</blockquote></div><br clear="all"><div><br></div>-- <br><div dir="ltr" class="gmail_signature"><div dir="ltr"><div>Pedro Henrique Borges de Melo</div></div></div>