I'm very grateful for your last reply. I had gone over my own function evaluations. There's no logarithm function or a divide by zero. <div>My test suite have 311 decision variable and 253 equality constraints, let alone 311 variable bound constraints. What scale does ipopt could deal with. But I thought my test suite wasn't big enough.</div>
<div>Is there any other reason that could lead to the same error message "<span style="font-family:arial,sans-serif;font-size:14.399999618530273px">EXIT: Invalid number in NLP function or derivative detected.</span> " </div>
<div>Expecting your reply.</div><div>Thank you.<br clear="all"><div>------------------------------------------------------------<br>Regards<br>Neo Ma 马平川 | Graduate<br><br>T:+86-10-6278 2545 | <a href="mailto:E%3Achambertinofn@gmail.com" target="_blank">E:chambertinofn@gmail.com</a><br>
------------------------------------------------------------<br>Dept. of Electrical Engineering, Tsinghua Univ.<br>BLDG. 28# RM.312 Tsinghua Univ. Beijing 100084 P.R.CHINA<br>------------------------------------------------------------<br>
Remember what should be remembered, and forget what should be forgotten. Alter what is changeable, and accept what is immutable.</div><br>
<br><br><div class="gmail_quote">On Tue, Nov 6, 2012 at 9:29 PM, Miao Li <span dir="ltr"><<a href="mailto:miao.li@epfl.ch" target="_blank">miao.li@epfl.ch</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<span style="color:rgb(41,41,41);font-family:Arial,Verdana,sans-serif;font-size:12.800000190734863px;line-height:16px;background-color:rgb(250,250,250)">"</span>
<span style="font-family:arial,sans-serif;font-size:14.399999618530273px">EXIT: Invalid number in NLP function or derivative detected.</span> <span style="color:rgb(41,41,41);font-family:Arial,Verdana,sans-serif;font-size:12.800000190734863px;line-height:16px;background-color:rgb(250,250,250)">":The error message means that Ipopt detected an error in function evaluations, typically a divide by zero or a logarithm of a negative number. You'd better check you model about this issue.</span><br>
<div>For the second, you can change the number of maximum iteration if I understand your problem correctly.</div><div><br></div><div>Best,</div><div>Miao</div><div class="gmail_extra"><br><br><div class="gmail_quote">2012/11/6 马平川 <span dir="ltr"><<a href="mailto:chambertinofn@gmail.com" target="_blank">chambertinofn@gmail.com</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><div class="h5">First, this is the only way to post my problems that I can find on the Ipopt home page. Maybe this message would be send to all the members on the mailing list. I'm sorry for bothering you with this mail. But if you could solve my problems, I would appriciate you for your reply.<div>
<br></div><div>I'm coding in Matlab for my master degree issue. But there was some unknown bugs in my program. I worried about that mistakes in my Primal-dual Inner Point Method codes had led these bugs. So I compiled the Ipopt matlab interface on Windows (ipopt.mexw32) on my PC. And then all the test which was from the Ipopt code (\build\Ipopt\contrib\MatlabInterface\examples) using my own compiled Ipopt matlab interface(ipopt.mexw32) has passed.</div>
<div><br></div><div>But the first running result after inputting 'clear all' in the matlab prompt is below.</div><div><div>******************************************************************************</div><div>
This program contains Ipopt, a library for large-scale nonlinear optimization.</div>
<div> Ipopt is released as open source code under the Eclipse Public License (EPL).</div><div> For more information visit <a href="http://projects.coin-or.org/Ipopt" target="_blank">http://projects.coin-or.org/Ipopt</a></div>
<div>
******************************************************************************</div><div><br></div><div>This is Ipopt version 3.10.0, running with linear solver ma27.</div><div><br></div><div>Number of nonzeros in equality constraint Jacobian...: 1336</div>
<div>Number of nonzeros in inequality constraint Jacobian.: 0</div><div>Number of nonzeros in Lagrangian Hessian.............: 3236</div><div><br></div><div><br></div><div>Number of Iterations....: 0</div><div>
<br></div><div>Number of objective function evaluations = 0</div><div>Number of objective gradient evaluations = 1</div><div>Number of equality constraint evaluations = 0</div><div>Number of inequality constraint evaluations = 0</div>
<div>Number of equality constraint Jacobian evaluations = 1</div><div>Number of inequality constraint Jacobian evaluations = 0</div><div>Number of Lagrangian Hessian evaluations = 0</div><div>Total CPU secs in IPOPT (w/o function evaluations) = 0.082</div>
<div>Total CPU secs in NLP function evaluations = 0.006</div><div><br></div><div>EXIT: Invalid number in NLP function or derivative detected.</div></div><div><br></div><div>And if I rerun this program without doing anything at once. Usually I can get this result.</div>
<div><div>This is Ipopt version 3.10.0, running with linear solver ma27.</div><div><br></div><div>Number of nonzeros in equality constraint Jacobian...: 1336</div><div>Number of nonzeros in inequality constraint Jacobian.: 0</div>
<div>Number of nonzeros in Lagrangian Hessian.............: 3236</div><div><br></div><div>Total number of variables............................: 310</div><div> variables with only lower bounds: 39</div>
<div> variables with lower and upper bounds: 38</div><div> variables with only upper bounds: 0</div><div>Total number of equality constraints.................: 253</div>
<div>Total number of inequality constraints...............: 0</div><div> inequality constraints with only lower bounds: 0</div><div> inequality constraints with lower and upper bounds: 0</div>
<div> inequality constraints with only upper bounds: 0</div><div><br></div><div>iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls</div><div> 0 0.0000000e+000 8.31e+000 4.49e+299 0.0 0.00e+000 - 0.00e+000 0.00e+000 0</div>
<div> 1 8.4037298e+005 8.30e+000 1.59e+000 -1.5 1.84e+301 - 6.87e-300 6.50e-301h 5</div><div> 2 2.0336766e+008 6.34e+002 1.64e+002 0.3 9.20e+001 - 8.39e-001 1.00e+000f 1</div></div><div>...</div><div>...</div>
<div>...</div><div><div>Number of Iterations....: 100</div><div><br></div><div> (scaled) (unscaled)</div><div>Objective...............: 8.5002463604944314e+003 8.5002463604944312e+011</div>
<div>Dual infeasibility......: 1.8701445737198796e+003 1.8701445737198798e+011</div><div>Constraint violation....: 3.9817380411393437e+001 7.1727917826186584e+001</div><div>Complementarity.........: 1.8222872952405129e+002 1.8222872952405128e+010</div>
<div>Overall NLP error.......: 1.8701445737198796e+003 1.8701445737198798e+011</div><div><br></div><div><br></div><div>Number of objective function evaluations = 1016</div><div>Number of objective gradient evaluations = 101</div>
<div>Number of equality constraint evaluations = 1216</div><div>Number of inequality constraint evaluations = 0</div><div>Number of equality constraint Jacobian evaluations = 101</div><div>Number of inequality constraint Jacobian evaluations = 0</div>
<div>Number of Lagrangian Hessian evaluations = 100</div><div>Total CPU secs in IPOPT (w/o function evaluations) = 0.391</div><div>Total CPU secs in NLP function evaluations = 6.904</div>
<div><br></div><div>EXIT: Maximum Number of Iterations Exceeded.</div></div><div><br></div><div>Please neglect the unconverged problem because my own faults may cause this problems. My problem is that the running result of exceeding the maximum number iterations or of jumping out without any iteration happened without any regular patterns. I wonder what had led this problem and how to avoid this.</div>
<div>My PC configuration:</div><div>OS: windows vista home premium X86</div><div>Matlab 7.10.0(R2010a)</div><div>Ipopt 3.10.0</div><div><br></div><div>Thank you for your reading. Hoping your reply.</div><div><div>------------------------------------------------------------<br>
Regards<br>Neo Ma 马平川 | Graduate<br><br>T:<a href="tel:%2B86-10-6278%202545" value="+861062782545" target="_blank">+86-10-6278 2545</a> | <a href="mailto:E%3Achambertinofn@gmail.com" target="_blank">E:chambertinofn@gmail.com</a><br>
------------------------------------------------------------<br>Dept. of Electrical Engineering, Tsinghua Univ.<br>
BLDG. 28# RM.312 Tsinghua Univ. Beijing 100084 P.R.CHINA<br>------------------------------------------------------------<br>Remember what should be remembered, and forget what should be forgotten. Alter what is changeable, and accept what is immutable.</div>
<br>
</div>
<br></div></div>_______________________________________________<br>
Ipopt mailing list<br>
<a href="mailto:Ipopt@list.coin-or.org" target="_blank">Ipopt@list.coin-or.org</a><br>
<a href="http://list.coin-or.org/mailman/listinfo/ipopt" target="_blank">http://list.coin-or.org/mailman/listinfo/ipopt</a><br>
<br></blockquote></div><span class="HOEnZb"><font color="#888888"><br><br clear="all"><div><br></div>-- <br><b><br><font size="1">Miao Li</font></b><font size="1"><br><b> Learning Algorithms and Systems
Laboratory</b><b> (LASA)</b><br></font><b><font size="1">École polytechnique fédérale de Lausanne (</font><acronym title="École Polytechnique Fédérale de Lausanne"><font size="1">EPFL)</font><br><br></acronym></b><br>
</font></span></div>
</blockquote></div><br></div>