<div dir="ltr">Hi Stefan,<div><br></div><div>Thanks!</div><div><br></div><div>Best regards,</div><div><br></div><div>Ramon.</div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On 25 May 2014 21:25, Stefan Vigerske <span dir="ltr"><<a href="mailto:stefan@math.hu-berlin.de" target="_blank">stefan@math.hu-berlin.de</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi,<br>
<br>
This is normal. Constraints have to be satisfied only within some tolerance, which is 1e-6 by default. See the numerics/feastol parameter.<br>
<br>
Stefan<div><div class="h5"><br>
<br>
On 05/25/2014 08:02 PM, Ramón Casero Cañas wrote:<br>
</div></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><div class="h5">
Dear all,<br>
<br>
I have found a couple of cases where I provide a problem with constraints<br>
to SCIP, I set the solution limit to 1, optimize, SCIP finds a solution,<br>
but at least one of the constraints is not fulfilled.<br>
<br>
I'm not too sure whether this is a finite precision problem (maybe I'm<br>
doing something wrong). I have attached a model and solution to illustrate<br>
the issue.<br>
<br>
******************************<u></u>******************************<u></u>**********************<br>
SCIP version 3.1.0, Linux 64 bit<br>
SCIP> read /tmp/model-tpd4cbba1c_26fc_<u></u>4b5c_81bf_0dcdffc1ec38.pip<br>
<br>
read problem </tmp/model-tpd4cbba1c_26fc_<u></u>4b5c_81bf_0dcdffc1ec38.pip><br>
============<br>
<br>
original problem has 61 variables (0 bin, 0 int, 0 impl, 61 cont) and 161<br>
constraints<br>
SCIP> set limits solutions 1<br>
limits/solutions = 1<br>
SCIP> optimize<br>
...<br>
2.7s| 1 | 0 | 32399 | - |2495k| 0 | 0 | 139 | 205 | 139<br>
|1731 |1697 | 0 | 0 |-2.187679e+04 |-2.003787e+04 | 9.18%<br>
2.7s| 1 | 1 | 32399 | - |2495k| 0 | 0 | 139 | 205 | 139<br>
|1731 |1697 | 0 | 0 |-2.187679e+04 |-2.003787e+04 | 9.18%<br>
<br>
SCIP Status : solving was interrupted [solution limit reached]<br>
Solving Time (sec) : 2.67<br>
Solving Nodes : 1<br>
Primal Bound : -2.00378746938832e+04 (1 solutions)<br>
Dual Bound : -2.18767884307964e+04<br>
Gap : 9.18 %<br>
<br>
SCIP> checksol<br>
<br>
check best solution<br>
solution is feasible in original problem<br>
******************************<u></u>******************************<u></u>**********************<br>
<br>
However, if we look at constraint 17:<br>
<br>
c17: -0.166666666666667 x197 y204 z202 + 0.166666666666667 x204 y197 z202<br>
+ 0.166666666666667 x197 y202 z204 + -0.166666666666667 x202 y197 z204 +<br>
-0.166666666666667 x204 y202 z197 + 0.166666666666667 x202 y204 z197 >=<br>
4.51681931542477e-08<br>
<br>
The values provided by the solution are:<br>
<br>
x197 = 0.371685887161333;<br>
y197 = -1.53809757847603;<br>
z197 = -1.32501091239681;<br>
<br>
x202 = 0.0972528846406156;<br>
y202 = -1.24399107862034;<br>
z202 = -1.64395150486303;<br>
<br>
x204 = -0.371685854531146;<br>
y204 = 1.53809760270505;<br>
z204 = 1.32501089342457;<br>
<br>
<br>
If we put these values into the constraint above, we get<br>
<br>
7.723300589346227e-09 >= 4.51681931542477e-08<br>
<br>
which is false. I have found another case with a similar problem where the<br>
left hand side of the inequality comes as a small negative number (~ -1e-8).<br>
<br>
Is this kind of "constraint almost fulfilled" behaviour to be expected? Am<br>
I missing something?<br>
<br>
Best regards,<br>
<br>
Ramon.<br>
<br>
<br>
<br></div></div>
______________________________<u></u>_________________<br>
Scip mailing list<br>
<a href="mailto:Scip@zib.de" target="_blank">Scip@zib.de</a><br>
<a href="http://listserv.zib.de/mailman/listinfo/scip" target="_blank">http://listserv.zib.de/<u></u>mailman/listinfo/scip</a><br>
<br>
</blockquote>
<br>
</blockquote></div><br><br clear="all"><div><br></div>-- <br><div dir="ltr">Dr. Ramón Casero Cañas<br><br>Institute of Biomedical Engineering<br>Department of Engineering Science, University of Oxford<br>Old Road Campus Research Building, Headington <br>
Oxford <br>OX3 7DQ<br>UK<br><br>tlf +44 (0) 1865 617716<br>twitter @Ramon_Casero<br>web <a href="http://www.cs.ox.ac.uk/people/Ramon.CaseroCanas/" target="_blank">http://www.cs.ox.ac.uk/people/Ramon.CaseroCanas/</a><br>
</div>
</div>