<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<style type="text/css" style="display:none;"><!-- P {margin-top:0;margin-bottom:0;} --></style>
</head>
<body dir="ltr">
<div id="divtagdefaultwrapper" style="font-size:12pt;color:#000000;font-family:Calibri,Helvetica,sans-serif;" dir="ltr">
<p>Thanks Ambros, you <span>hit the nail on the head. Were using SCIPtransformVar() instead of SCIPgetTranformVar().</span></p>
<div><br>
</div>
Cheers
<div>Christian<br>
<br>
<div style="color: rgb(0, 0, 0);">
<div>
<hr tabindex="-1" style="display:inline-block; width:98%">
<div id="x_divRplyFwdMsg" dir="ltr"><font face="Calibri, sans-serif" color="#000000" style="font-size:11pt"><b>Von:</b> Scip <scip-bounces@zib.de> im Auftrag von Ambros Gleixner <gleixner@zib.de><br>
<b>Gesendet:</b> Montag, 7. Dezember 2020 22:04<br>
<b>An:</b> scip@zib.de<br>
<b>Betreff:</b> Re: [SCIP] Warning about unreleased variables?</font>
<div> </div>
</div>
</div>
<font size="2"><span style="font-size:10pt;">
<div class="PlainText">Dear Christian,<br>
<br>
One suspicion: Are you using SCIPtransformVar()? This adds a capture <br>
even if the transformed variable already exists. In this case, you <br>
could use SCIPgetTransformVar() instead.<br>
<br>
Otherwise, I don't think any of the branching related methods capture <br>
variables automatically.<br>
<br>
Does that help?<br>
<br>
Best,<br>
Ambros<br>
<br>
<br>
Am 02.12.20 um 11:05 schrieb Franzen, Christian:<br>
> Dear SCIP Team.<br>
> <br>
> <br>
> Currently I am implementing a branching rule within my <br>
> branch-and-price-and-cut solver. That seems to work correctly so far. <br>
> However, whenever I am done with solving and call SCIPfree() to cleanup <br>
> all resources I see the following warning in the logs<br>
> <br>
> <br>
> WARNING: Transformed variable <PathFlow_agg4191_path25> not released <br>
> when freeing SCIP.<br>
> <br>
> Better to say, I see thousands of them. These seems to be related to my <br>
> branching rule, since I do not see the warnings when solving only the <br>
> root node. So it seems to be not related to my Pricer or Separators. I <br>
> couldn't find anything in the documentation, but is there a chance that <br>
> one of the methods for finding branching candidates is capturing the <br>
> corrsponding variables (e.g., SCIPgetLPBranchCands)? I do not capture <br>
> them on my one and I am quite sure I am releasing all <br>
> (original) variables after creation. Any ideas what I missed here?<br>
> <br>
> <br>
> Regards<br>
> <br>
> Christian<br>
> <br>
> <br>
> _______________________________________________<br>
> Scip mailing list<br>
> Scip@zib.de<br>
> <a href="https://listserv.zib.de/mailman/listinfo/scip">https://listserv.zib.de/mailman/listinfo/scip</a><br>
> <br>
_______________________________________________<br>
Scip mailing list<br>
Scip@zib.de<br>
<a href="https://listserv.zib.de/mailman/listinfo/scip">https://listserv.zib.de/mailman/listinfo/scip</a><br>
</div>
</span></font></div>
</div>
</div>
</body>
</html>