[SCIP] pricing question

James Cussens james.cussens at york.ac.uk
Thu Jul 30 11:12:52 CEST 2020


Dear SCIPers,

I have a branch-price-and-cut application, and am using SCIP 7.0.1. When I
set
pricing/maxvarsroot = 3
I get the following behaviour:

  time | node  | left  |LP iter|mem/heur|depth|mdpt |cons |cols |rows |cuts
|   dualbound   | primalbound
p 3.1s|     1 |     0 |     3 |  clique| 671 |   0 |2038 |1556 |2035 |   0
|       --      |-1.672124e+03
  5.7s|     1 |     0 |   114 |   150M |   0 |   0 |2038 |1559 |2035 |   0
|       --      |-1.672124e+03
 11.0s|     1 |     0 |   421 |   150M |   0 |   0 |2038 |1859 |2035 |   0
|       --      |-1.672124e+03
 17.3s|     1 |     0 |   747 |   150M |   0 |   0 |2038 |2159 |2035 |   0
|       --      |-1.672124e+03
 25.8s|     1 |     0 |   920 |   691M |   0 |   0 |2038 |2340 |2035 |   0
| -9.382209e+02 |-1.672124e+03
 50.5s|     1 |     0 |  1056 |   730M |   0 |   0 |2038 |2343 |2105 |  70
| -9.382209e+02 |-1.672124e+03
 52.1s|     1 |     0 |  1346 |   730M |   0 |   0 |2038 |2417 |2057 |  70
| -9.655814e+02 |-1.672124e+03

So at time 5.7s we have 3 new columns as expected. However for the next two
lines we see that 3*100 new columns are introduced.
I have had a look at what is going on in SCIPpricestoreApplyVars and each
time it is called we only get 3 new variables priced in. So where do the
100 iterations come from?

James

James
-- 
James Cussens
Room CSE/239
Dept of Computer Science
University of York
York YO10 5GE, UK
Tel    +44 (0)1904 325371
http://www.cs.york.ac.uk/~jc
http://www.york.ac.uk/docs/disclaimer/email.htm
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.zib.de/pipermail/scip/attachments/20200730/c37f4853/attachment.html>


More information about the Scip mailing list