<html><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /></head><body style='font-size: 10pt; font-family: Verdana,Geneva,sans-serif'>
<div><span>Dear Steve and Gerald,</span></div>
<div><span> </span></div>
<div><span>Many thanks for the two ideas you give us to our question, and sorry for the false hope.</span></div>
<div><span> </span></div>
<div><span>Unfornately, either the two solutions or a mix of the two appears sometimes to raise a bug into SCIP.</span></div>
<div><span> </span></div>
<div><span>Our aim is still to freely decide to stop the pricing phase in order to add some inequalities before pricing convergence.</span></div>
<div><span> </span></div>
<div><span>The Steve's idea was to use SCIP_DITNOTRUN return value at the end of the pricing Callback.</span></div>
<div><span>The Gerald's idea was to previously perform the separation algo within the pricing Callback in order to be sure to generate a cut in the next separator Callback: unfortunately, when no column is added in the pricing callback, the current relaxation is taken as a lower bound which induces bad decisions in the following.</span></div>
<div><span>We tried to mix the two idea and we then use SCIP_DITNOTRUN with Gerald's idea.</span></div>
<div><span> </span></div>
<div><span>Unfortunately, as Eddie has noticed with his own code, for some instances, the use of SCIP_DITNOTRUN in the pricing Callback appears to stop the whole SCIP BCP process before convergence was reached.</span></div>
<div><span> </span></div>
<div><span>If someone wants to test our code, please find here a link to an instance which always fails. When SCIP stops, the convergence is not reached: this can be seen with our second code where all the columns and all the inequalities that have been produced in the first code, have been added till the beginning... and in that case SCIP doesn't stop! (but stops a few iteration later without reaching the convergence).</span></div>
<div><span> </span></div>
<div><span>We guess that using several SCIP_DITNOTRUN can raise a bug for some instance.</span></div>
<div><span> </span></div>
<div><span>Perhaps the case Eddie indicates correspond to the fact that SCIP_DITNOTRUN is not to be used when the solution to be priced is integral. Unfortunately the bug is reached for our code with a fractional value.</span></div>
<div> </div>
<div><span>If you wanna check what happends, there is our code : https://github.com/alexandredupontbouillard/B-C-P-coloring</span></div>
<div><span> </span></div>
<div><span>You can find in the readme explanations on how to reproduce the bug.</span></div>
<div> </div>
<div>Regards,</div>
<div>Pierre and Alexandre</div>
<div id="signature"> </div>
<p><br /></p>
<p id="reply-intro">Le 2021-11-24 21:23, Alexandre Dupont-Bouillard a écrit :</p>
<blockquote type="cite" style="padding: 0 0.4em; border-left: #1010ff 2px solid; margin: 0"><!-- html ignored --><!-- head ignored --><!-- meta ignored -->
<p>Dear Gerald,</p>
<p><br /></p>
<p>It now works as we wanted, many thanks for your precious help.</p>
<p><br /></p>
<p>Regards,</p>
<p>Pierre and Alexandre</p>
<div id="signature"> </div>
<p><br /></p>
<p id="reply-intro">Le 2021-11-11 15:22, Gerald Gamrath a écrit :</p>
<blockquote style="padding: 0 0.4em; border-left: #1010ff 2px solid; margin: 0;">
<p>Hi Alexandre,</p>
<p>what you could do to avoid changing the SCIP code would be to check within your pricer if your separator will find cuts for the current LP solution. For example, you could have an external function that you can call from your pricer which returns whether a cut can be generated. If that check may be expensive, you could buffer the generated cut so that the separation call can directly add that cut. I could imagine that having this close relation between cutting and pricing implemented in your plugins could be beneficial anyway, e.g., if you want to make the decision whether or not to continue pricing depend on the strength of the cuts that your separator would generate.</p>
<p>Best,<br />Gerald</p>
<div class="moz-cite-prefix">Am 08.11.21 um 20:11 schrieb Alexandre Dupont-Bouillard:</div>
<blockquote style="padding: 0 0.4em; border-left: #1010ff 2px solid; margin: 0;">
<p><span style="white-space: nowrap;">Dear Steve</span><br /><br /><span style="white-space: nowrap;">Many thanks for your answer.</span><br /><span style="white-space: nowrap;">The question of free alternation between cut and price seems to be a new</span><br /><span style="white-space: nowrap;">strategy since it's not classical for SCIP!</span><br /><br /><span style="white-space: nowrap;">Unfortunately, as said by Edward, your proposal seems not to work.</span><br /><span style="white-space: nowrap;">Indeed, setting the result as SCIP_DIDNOTRUN in PRICERREDCOST callback</span><br /><span style="white-space: nowrap;">appears to exactly work as artificially saying that no column is produced.</span><br /><br />This brings the problem you point out: no pricing round is launched after a cutting<br /><span style="white-space: nowrap;">round with no generated cut.</span><br />This a problem for us because the previous pricing round has been artificially<br /><span style="white-space: nowrap;">inhibit in order to launch a cutting phase.</span><br />Indeed, when there is no generated cut in a cutting round, a supplementary exact<br />pricing round is needed to assert convergence (and potentially to go on with the<br /><span style="white-space: nowrap;">price&cut process).</span><br /><br /><span style="white-space: nowrap;">A solution appears to turn off the test stopping the price&cut process</span><br /><span style="white-space: nowrap;">when no cut have been generated during a cutting round</span><br /><span style="white-space: nowrap;">With such a deleted test, the price&cut iterations will have no end since</span><br /><span style="white-space: nowrap;">the cutting and pricing rounds will call each other in a never ending loop.</span><br />A way to stop this loop is to add a test having the memory of the previous round<br />in order to stop when two successive non-inhibited price and cut rounds have produced<br /><span style="white-space: nowrap;">nothing.</span></p>
<p><span style="white-space: nowrap;">Regards,</span><br /><span style="white-space: nowrap;">Pierre and Alexandre</span><br /><br /><br />Ps: Another disturbing problem for us was that a cutting round is launched only<br />if the current solution is not integer: this is a problem for us because an integer<br /><span style="white-space: nowrap;">solution can appear (before convergence) right after a pricing round.</span><br />In this case, it is mandatory for us to do a second pricing round, even if our strategy<br /><span style="white-space: nowrap;">is to alternate with a cutting round (our cutting procedure cannot produce cuts for integral solutions).</span><br />We add a test at the beginning of the PRICERREDCOST callback for no-inhibiting the pricing<br /><span style="white-space: nowrap;">when the current solution is integer.</span></p>
<div id="signature"> </div>
<p><br /></p>
<p id="reply-intro">Le 2021-11-08 11:25, Edward Lam a écrit :</p>
<blockquote style="padding: 0 0.4em; border-left: #1010ff 2px solid; margin: 0;">I have tried this. It sounds like this would work according to the documentation but this does not work. Sometimes it skips pricing and separation and goes straight to branching. I guess branching is technically okay if the master problem is fractional but sometimes its integral and gets messy quickly.<br />
<div> </div>
<div>
<div dir="auto" style="caret-color: #000000; color: #000000; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;">
<div dir="auto" style="caret-color: #000000; color: #000000; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;"><span style="font-style: normal;">Cheers<br />Eddie</span></div>
</div>
</div>
<div><br />
<blockquote style="padding: 0 0.4em; border-left: #1010ff 2px solid; margin: 0;">
<div>On 8 Nov 2021, at 7:10 pm, Maher, Stephen <<a href="mailto:S.J.Maher@exeter.ac.uk" rel="noreferrer">S.J.Maher@exeter.ac.uk</a>> wrote:</div>
<div>
<div style="font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt;">Hi Alexandre,</div>
<div style="font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt;"> </div>
<div style="font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt;">Sorry about the delay in response to this question. This happens to be a question that we have not encountered before, so we needed to work out how to actually achieve this.</div>
<div style="font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt;"> </div>
<div style="font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt;">I am not sure whether this would work, since I don't have an example to test with. I believe that by setting the result in your PRICERREDCOST callback to SCIP_DIDNOTRUN, you will terminate pricing and then continue with the rest of the node processing. The separation round will begin after you terminate the pricing round.</div>
<div style="font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt;"> </div>
<div style="font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt;">There are two things to note here. The first is that if you add a cut, then the node processing will enter back into pricing. So you may want to keep returning the result of SCIP_DIDNOTRUN from your pricer while you want to keep generating cuts. The second is that if you don't add a cut during the separation round, then you will not enter back into the pricing round. So you must ensure that you have completely finished pricing before the final separation round.</div>
<div style="font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt;"> </div>
<div style="font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt;">Since this is untested, please let us know if it works for your setting. If not, then we can think of a different approach.</div>
<div style="font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt;"> </div>
<div style="font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt;">Cheers,</div>
<div style="font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt;"> </div>
<div style="font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt;">Steve</div>
<div style="caret-color: #000000; font-family: Helvetica; font-size: 14px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;">
<div id="appendonsend"> </div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt;"> </div>
<hr style="display: inline-block; width: 1281.828125px;" />
<div id="divRplyFwdMsg" dir="ltr"><span style="font-size: 11pt; font-family: Calibri, sans-serif;"><strong>From:</strong><span class="Apple-converted-space"> </span>Scip <<a href="mailto:scip-bounces@zib.de" rel="noreferrer">scip-bounces@zib.de</a>> on behalf of Alexandre Dupont-Bouillard <<a href="mailto:dupont-bouillard@lipn.univ-paris13.fr" rel="noreferrer">dupont-bouillard@lipn.univ-paris13.fr</a>><br /><strong>Sent:</strong><span class="Apple-converted-space"> </span>25 October 2021 09:54<br /><strong>To:</strong><span class="Apple-converted-space"> </span><a href="mailto:scip@zib.de" rel="noreferrer">scip@zib.de</a><span class="Apple-converted-space"> </span><<a href="mailto:scip@zib.de" rel="noreferrer">scip@zib.de</a>><br /><strong>Subject:</strong><span class="Apple-converted-space"> </span>[SCIP] How to freely tune between pricing and cutting ?</span>
<div> </div>
</div>
<div class="BodyFragment">
<div class="PlainText">CAUTION: This email originated from outside of the organisation. Do not click links or open attachments unless you recognise the sender and know the content is safe.<br /><br /><br />Hi<br /><br />We try to build a BCP method where cuts are added through a separator.<br />We would like to have a free hand on the alternation between pricing and<br />cutting phases.<br /><br />We haven't found how to add parameters to choose for instance to begin<br />by a cutting phase<br />or to stop pricing before convergence to add some inequalities and come<br />back to pricing<br />and so on...<br /><br />The best would be to be able to switch from cutting to pricing (and from<br />pricing to cutting) using a test done by an algorithm.<br /><br />Does someone know how to perform such alternation?<br /><br />Thanks<br />Pierre and Alexandre<br />_______________________________________________<br />Scip mailing list<br /><a href="mailto:Scip@zib.de" rel="noreferrer">Scip@zib.de</a><br /><a href="https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Flistserv.zib.de%2Fmailman%2Flistinfo%2Fscip&data=04%7C01%7CS.J.Maher%40exeter.ac.uk%7Ce1f9ee8f46304091c1a308d99796bbb3%7C912a5d77fb984eeeaf321334d8f04a53%7C0%7C0%7C637707496323265507%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=BjaINZ3HZglnha70BFGyyDMOEtBY9ygHd9Nz%2F6VsGGM%3D&reserved=0" target="_blank" rel="noopener noreferrer">https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Flistserv.zib.de%2Fmailman%2Flistinfo%2Fscip&data=04%7C01%7CS.J.Maher%40exeter.ac.uk%7Ce1f9ee8f46304091c1a308d99796bbb3%7C912a5d77fb984eeeaf321334d8f04a53%7C0%7C0%7C637707496323265507%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=BjaINZ3HZglnha70BFGyyDMOEtBY9ygHd9Nz%2F6VsGGM%3D&reserved=0</a></div>
</div>
</div>
<span style="caret-color: #000000; font-family: Helvetica; font-size: 14px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;">_______________________________________________</span><br style="caret-color: #000000; font-family: Helvetica; font-size: 14px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" /><span style="caret-color: #000000; font-family: Helvetica; font-size: 14px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;">Scip mailing list</span><br style="caret-color: #000000; font-family: Helvetica; font-size: 14px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" /><a style="font-family: Helvetica; font-size: 14px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px;" href="mailto:Scip@zib.de" rel="noreferrer">Scip@zib.de</a><br style="caret-color: #000000; font-family: Helvetica; font-size: 14px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" /><a style="font-family: Helvetica; font-size: 14px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px;" href="https://listserv.zib.de/mailman/listinfo/scip" target="_blank" rel="noopener noreferrer">https://listserv.zib.de/mailman/listinfo/scip</a></div>
</blockquote>
</div>
<br /><br />
<div class="pre" style="margin: 0; padding: 0; font-family: monospace;">_______________________________________________<br />Scip mailing list<br /><a href="mailto:Scip@zib.de" rel="noreferrer">Scip@zib.de</a><br /><a href="https://listserv.zib.de/mailman/listinfo/scip" target="_blank" rel="noopener noreferrer">https://listserv.zib.de/mailman/listinfo/scip</a></div>
</blockquote>
<br /><fieldset class="mimeAttachmentHeader"></fieldset>
<pre class="moz-quote-pre">_______________________________________________
Scip mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Scip@zib.de" rel="noreferrer">Scip@zib.de</a>
<a class="moz-txt-link-freetext" href="https://listserv.zib.de/mailman/listinfo/scip" target="_blank" rel="noopener noreferrer">https://listserv.zib.de/mailman/listinfo/scip</a>
</pre>
</blockquote>
</blockquote>
<br />
<div class="pre" style="margin: 0; padding: 0; font-family: monospace">_______________________________________________<br />Scip mailing list<br /><a href="mailto:Scip@zib.de" rel="noreferrer">Scip@zib.de</a><br /><a href="https://listserv.zib.de/mailman/listinfo/scip" target="_blank" rel="noopener noreferrer">https://listserv.zib.de/mailman/listinfo/scip</a></div>
</blockquote>
</body></html>