[SCIP] 回复: About using SCIP to solve a MINLP problem
Ambros Gleixner
gleixner at zib.de
Fri Aug 21 09:42:24 CEST 2020
Dear Yin Qyy,
I think your follow-up question has not been answered, yet. It is quite
typical that closing the gap can take the largest amount of time.
Except for Stefan's initial suggestions, which seem to have helped
already, it is unfortunately hard to give more specific advice. These
MINLPs can be difficult to solve. In the last log it still seems like
finding the optimal solution of value 3 is the issue. Maybe you have to
think of designing a problem-specific heuristic for your model. For
this you would probably need to switch from AMPL to something more
flexible like PySCIPOpt.
Best,
Ambros
Am 06.08.20 um 08:13 schrieb yin qyy:
> Dear,
>
> Thank you very much for your guidance. After trying many parameter
> settings, I did greatly improve the calculation speed of SCIP. I set the
> emphasis for heuristics to aggressive，the overall speed is much faster.
> Now sometimes the primalboundcan be found quickly, but the primal bound
> is very far from the dualbound. This part is often the most
> time-consuming, such as:
>
> L 8.0s| 1 | 0 | 0 | - |trustreg| 0 | 156 | 10 | 32 | 0 | 2 | 1 | 0 |
> 3.000000e+00 | 3.700000e+01 |1133.33%| unknown
>
> L 9.0s| 1 | 0 | 0 | - |trustreg| 0 | 156 | 10 | 32 | 0 | 2 | 1 | 0 |
> 3.000000e+00 | 3.600000e+01 |1100.00%| unknown
>
> L11.0s| 1 | 0 | 0 | - |trustreg| 0 | 156 | 10 | 32 | 0 | 2 | 1 | 0 |
> 3.000000e+00 | 3.500000e+01 |1066.67%| unknown
>
> 19.0s| 1 | 0 | 0 | - | 3654k | 0 | 156 | 10 | 32 | 0 | 2 | 1 | 0 |
> 3.000000e+00 | 3.500000e+01 |1066.67%| unknown
>
> 19.0s| 1 | 0 | 3 | - | 3654k | 0 | 156 | 10 | 35 | 3 | 4 | 1 | 0 |
> 3.000000e+00 | 3.500000e+01 |1066.67%| unknown
>
> 19.0s| 1 | 0 | 5 | - | 3746k | 0 | 156 | 10 | 36 | 4 | 5 | 1 | 0 |
> 3.000000e+00 | 3.500000e+01 |1066.67%| unknown
>
> 19.0s| 1 | 2 | 5 | - | 3747k | 0 | 156 | 10 | 36 | 4 | 6 | 1 | 0 |
> 3.000000e+00 | 3.500000e+01 |1066.67%| unknown
>
> 33.0s| 100 | 101 | 2999 | 30.2 | 16M | 15 | 156 | 10 | 13 |1077 | 1 | 1
> | 0 | 3.000000e+00 | 3.500000e+01 |1066.67%| unknown
>
> 49.0s| 200 | 201 | 5431 | 27.3 | 31M | 23 | 156 | 10 | 37 |2063 | 3 | 1
> | 0 | 3.000000e+00 | 3.500000e+01 |1066.67%| unknown
>
> L56.0s| 201 | 202 | 5445 | 27.2 |mutation| 23 | 156 | 10 | 43 |2075 | 3
> | 1 | 0 | 3.000000e+00 | 2.700000e+01 | 800.00%| unknown
>
> L58.0s| 206 | 207 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 2.600000e+01 | 766.67%| unknown
>
> L59.0s| 206 | 207 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 2.500000e+01 | 733.33%| unknown
>
> L60.0s| 206 | 207 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 2.400000e+01 | 700.00%| unknown
>
> L61.0s| 206 | 207 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 2.300000e+01 | 666.67%| unknown
>
> time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows
> |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl.
>
> L62.0s| 206 | 207 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 2.200000e+01 | 633.33%| unknown
>
> L63.0s| 206 | 207 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 2.100000e+01 | 600.00%| unknown
>
> L64.0s| 206 | 207 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 2.000000e+01 | 566.67%| unknown
>
> L65.0s| 206 | 207 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 1.900000e+01 | 533.33%| unknown
>
> L66.0s| 206 | 207 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 1.800000e+01 | 500.00%| unknown
>
> L67.0s| 206 | 207 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 1.700000e+01 | 466.67%| unknown
>
> L68.0s| 206 | 207 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 1.600000e+01 | 433.33%| unknown
>
> L69.0s| 206 | 207 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 1.500000e+01 | 400.00%| unknown
>
> L70.0s| 206 | 207 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 1.400000e+01 | 366.67%| unknown
>
> L71.0s| 206 | 207 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 1.300000e+01 | 333.33%| unknown
>
> L72.0s| 206 | 207 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 1.200000e+01 | 300.00%| unknown
>
> L73.0s| 206 | 184 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 1.100000e+01 | 266.67%| unknown
>
> L75.0s| 206 | 169 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 1.000000e+01 | 233.33%| unknown
>
> L76.0s| 206 | 148 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 9.000000e+00 | 200.00%| unknown
>
> L77.0s| 206 | 129 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 8.000000e+00 | 166.67%| 4.72%
>
> time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows
> |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl.
>
> L78.0s| 206 | 105 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1
> | 1 | 0 | 3.000000e+00 | 7.000000e+00 | 133.33%| 8.31%
>
> L79.0s| 206 | 79 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1 |
> 1 | 0 | 3.000000e+00 | 6.000000e+00 | 100.00%| 15.99%
>
> L80.0s| 206 | 54 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1 |
> 1 | 0 | 3.000000e+00 | 5.000000e+00 | 66.67%| 29.13%
>
> L81.0s| 206 | 28 | 5489 | 26.8 |trustreg| 23 | 156 | 10 | 55 |2106 | 1 |
> 1 | 0 | 3.000000e+00 | 4.000000e+00 | 33.33%| 57.88%
>
> 90.0s| 300 | 108 | 6328 | 21.1 | 32M | 31 | 156 | 10 | 22 |2292 | 1 | 1
> | 0 | 3.000000e+00 | 4.000000e+00 | 33.33%| 57.88%
>
> 96.0s| 400 | 190 | 6902 | 17.3 | 32M | 41 | 156 | 13 | 26 |2473 | 3 | 4
> | 0 | 3.000000e+00 | 4.000000e+00 | 33.33%| 57.88%
>
> SCIP Status : solving was interrupted [time limit reached]
>
> Solving Time (sec) : 100.00
>
> Solving Nodes : 421
>
> Primal Bound : +4.00000000000000e+00 (28 solutions)
>
> Dual Bound : +3.00000000000000e+00
>
> Gap : 33.33 %
>
> How can I improve this situation?
>
> Thank you very much again for your guidance!
>
> Best,
>
> Jax
>
> 发送自Windows 10 版邮件 <https://go.microsoft.com/fwlink/?LinkId=550986>应用
>
> *发件人: *Stefan Vigerske <mailto:svigerske at gams.com>
> *发送时间: *2020年8月4日17:02
> *收件人: *yin qyy <mailto:yinqyy at outlook.com>; scip at zib.de
> <mailto:scip at zib.de>
> *主题: *Re: [SCIP] About using SCIP to solve a MINLP problem
>
> Hi,
>
> you could try to crank up how much primal heuristics are run. At least
> for that small example, the dual bound seems to be best possible from
> the beginning one and it only took a while to find a feasible solution
> with same objective value.
> The primal heuristic that found the solution with objective function
> value 4 was nlpdiving, which makes some sense. So changing
> heuristics/nlpdiving/freq and other parameters of this heuristic may
> help to make it run more often or more extensively.
> You can also make all primal heuristics run more extensively by setting
> the emphasis for heuristics to aggressive (doing this through AMPL may
> be a bit difficult; maybe just open a SCIP shell from a SCIP binary,
> type set heuristics emphasis aggr
> (https://www.scipopt.org/doc-7.0.1/html/SHELL.php#TUTORIAL_PARAMETERS)
> and copy the parameter changes).
>
> Since the only nonlinearity are products of binary and bounded
> continuous variable, if I get that right, then you could also try
> whether letting SCIP reformulate the problem into a MIP in presolve
> (constraints/quadratic/binreformbinaryonly = FALSE, maybe try also
> constraints/quadratic/replacebinaryprod = 1) helps.
>
> Best,
> Stefan
>
>
>
> On 8/3/20 12:08 PM, yin qyy wrote:
>> Thank you for your patience. I am a student. I am using SCIP to solve a
>> MINLP problem. This is my model:
>>
>> bl is a binary variable, z is an element of the Zr matrix, Zr can be
>> calculated, but the composition of Zr is very complicated; a is an
>> element of matrix A, this constraint can be understood as A is the
>> inverse of Zr, A* Zr=1; So in fact Z is linear, and A is also linear,
>> which is equivalent to bilinear; both theta and epsilon are set
>> parameters, and specific values can be calculated;
>>
>> Here is a simple example, we set nq=2，nz=1，ntp=4；
>>
>> Zr=[3.64545486 3.85360847 2.21292579 3.55129215
>>
>> 2.44837376 3.25989972 2.88290441 2.19697746]
>>
>> We use AMPL to model, the model and data are as follows M1.mod and M1.us;
>>
>> In this small example, scip can quickly find the optimal solution, but
>> when the scale of our problem expands, scip sometimes spends a lot of
>> time finding the optimal solution, such as:
>>
>> time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows
>> |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl.
>>
>> 4.0s| 1 | 0 | 1 | - | 3700k | 0 | 156 | 10 | 29 | 0 | 0 | 0 | 0 |
>> 3.000000e+00 | -- | Inf | unknown
>>
>> 4.0s| 1 | 0 | 3 | - | 3729k | 0 | 156 | 11 | 31 | 2 | 1 | 0 | 0 |
>> 3.000000e+00 | -- | Inf | unknown
>>
>> 4.0s| 1 | 0 | 5 | - | 3892k | 0 | 156 | 11 | 32 | 3 | 2 | 0 | 0 |
>> 3.000000e+00 | -- | Inf | unknown
>>
>> 4.0s| 1 | 0 | 7 | - | 3892k | 0 | 156 | 11 | 33 | 4 | 3 | 0 | 0 |
>> 3.000000e+00 | -- | Inf | unknown
>>
>> 4.0s| 1 | 0 | 8 | - | 3896k | 0 | 156 | 11 | 34 | 5 | 4 | 0 | 0 |
>> 3.000000e+00 | -- | Inf | unknown
>>
>> 4.0s| 1 | 0 | 9 | - | 3899k | 0 | 156 | 11 | 35 | 6 | 5 | 0 | 0 |
>> 3.000000e+00 | -- | Inf | unknown
>>
>> 4.0s| 1 | 0 | 10 | - | 3994k | 0 | 156 | 11 | 36 | 7 | 6 | 0 | 0 |
>> 3.000000e+00 | -- | Inf | unknown
>>
>> 4.0s| 1 | 0 | 13 | - | 3998k | 0 | 156 | 11 | 37 | 8 | 7 | 0 | 0 |
>> 3.000000e+00 | -- | Inf | unknown
>>
>> 4.0s| 1 | 0 | 14 | - | 4001k | 0 | 156 | 11 | 38 | 9 | 8 | 0 | 0 |
>> 3.000000e+00 | -- | Inf | unknown
>>
>> 4.0s| 1 | 0 | 15 | - | 4004k | 0 | 156 | 11 | 39 | 10 | 9 | 0 | 0 |
>> 3.000000e+00 | -- | Inf | unknown
>>
>> 4.0s| 1 | 0 | 16 | - | 4008k | 0 | 156 | 11 | 40 | 11 | 10 | 0 | 0 |
>> 3.000000e+00 | -- | Inf | unknown
>>
>> 4.0s| 1 | 0 | 17 | - | 4012k | 0 | 156 | 11 | 11 | 12 | 11 | 0 | 0 |
>> 3.000000e+00 | -- | Inf | unknown
>>
>> 4.0s| 1 | 0 | 20 | - | 4013k | 0 | 156 | 11 | 12 | 13 | 12 | 0 | 0 |
>> 3.000000e+00 | -- | Inf | unknown
>>
>> ******************************************************************************
>>
>> This program contains Ipopt, a library for large-scale nonlinear
>> optimization.
>>
>> Ipopt is released as open source code under the Eclipse Public License
>> (EPL).
>>
>> For more information visit http://projects.coin-or.org/Ipopt
>>
>> ******************************************************************************
>>
>> 5.0s| 1 | 0 | 27 | - | 4228k | 0 | 156 | 13 | 21 | 22 | 12 | 0 | 0 |
>> 3.000000e+00 | -- | Inf | unknown
>>
>> 5.0s| 1 | 0 | 30 | - | 4228k | 0 | 156 | 13 | 19 | 23 | 13 | 0 | 0 |
>> 3.000000e+00 | -- | Inf | unknown
>>
>> time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows
>> |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl.
>>
>> 5.0s| 1 | 2 | 30 | - | 4228k | 0 | 156 | 13 | 19 | 23 | 14 | 0 | 2 |
>> 3.000000e+00 | -- | Inf | unknown
>>
>> r 5.0s| 7 | 7 | 301 | 45.8 |rounding| 6 | 156 | 18 | 51 | 112 | 4 | 0 |
>> 4 | 3.000000e+00 | 7.000000e+00 | 133.33%| 0.29%
>>
>> r 7.0s| 29 | 26 | 649 | 22.3 |rounding| 21 | 156 | 18 | 58 | 205 | 4 | 0
>> | 14 | 3.000000e+00 | 5.000000e+00 | 66.67%| 1.15%
>>
>> d10.0s| 71 | 13 | 1070 | 14.9 |nlpdivin| 23 | 156 | 18 | 25 | 291 | 1 |
>> 0 | 14 | 3.000000e+00 | 4.000000e+00 | 33.33%| 23.92%
>>
>> 10.0s| 100 | 24 | 1351 | 13.4 | 5979k | 23 | 156 | 18 | 49 | 437 | 3 | 0
>> | 15 | 3.000000e+00 | 4.000000e+00 | 33.33%| 24.49%
>>
>> 12.0s| 200 | 24 | 2393 | 11.9 | 6636k | 65 | 156 | 18 | 122 |1046 | 3 |
>> 0 | 23 | 3.000000e+00 | 4.000000e+00 | 33.33%| 24.49%
>>
>> 13.0s| 300 | 24 | 3251 | 10.8 | 7738k | 115 | 156 | 18 | 222 |1696 | 3 |
>> 0 | 23 | 3.000000e+00 | 4.000000e+00 | 33.33%| 24.49%
>>
>> 14.0s| 400 | 54 | 4154 | 10.3 | 8345k | 143 | 156 | 18 | 124 |2340 | 2 |
>> 0 | 23 | 3.000000e+00 | 4.000000e+00 | 33.33%| 24.49%
>>
>> 15.0s| 500 | 54 | 5352 | 10.7 | 8964k | 143 | 156 | 18 | 234 |3043 | 2 |
>> 0 | 23 | 3.000000e+00 | 4.000000e+00 | 33.33%| 24.49%
>>
>> 16.0s| 600 | 90 | 6299 | 10.5 | 10M | 143 | 156 | 18 | 110 |3561 | 2 | 0
>> | 25 | 3.000000e+00 | 4.000000e+00 | 33.33%| 26.78%
>>
>> 17.0s| 700 | 162 | 7015 | 10.0 | 12M | 143 | 156 | 18 | 72 |3963 | 2 | 0
>> | 34 | 3.000000e+00 | 4.000000e+00 | 33.33%| 27.93%
>>
>> 18.0s| 800 | 236 | 7630 | 9.5 | 12M | 143 | 156 | 18 | 94 |4311 | 2 | 0
>> | 46 | 3.000000e+00 | 4.000000e+00 | 33.33%| 28.50%
>>
>> 20.0s| 900 | 310 | 8387 | 9.3 | 14M | 143 | 156 | 18 | 22 |4659 | 2 | 0
>> | 61 | 3.000000e+00 | 4.000000e+00 | 33.33%| 28.79%
>>
>> 21.0s| 1000 | 340 | 9725 | 9.7 | 16M | 143 | 156 | 18 | 182 |5253 | 2 |
>> 0 | 71 | 3.000000e+00 | 4.000000e+00 | 33.33%| 28.93%
>>
>> 22.0s| 1100 | 414 | 10609 | 9.6 | 18M | 143 | 156 | 18 | 147 |5696 | 2 |
>> 0 | 92 | 3.000000e+00 | 4.000000e+00 | 33.33%| 29.00%
>>
>> time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows
>> |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl.
>>
>> 24.0s| 1200 | 450 | 11828 | 9.8 | 20M | 143 | 156 | 18 | 144 |6247 | 2 |
>> 0 | 107 | 3.000000e+00 | 4.000000e+00 | 33.33%| 29.04%
>>
>> 25.0s| 1300 | 490 | 13189 | 10.1 | 21M | 143 | 156 | 18 | 96 |6775 | 2 |
>> 0 | 121 | 3.000000e+00 | 4.000000e+00 | 33.33%| 29.05%
>>
>> 26.0s| 1400 | 490 | 14129 | 10.1 | 21M | 143 | 156 | 18 | 147 |7326 | 2
>> | 0 | 121 | 3.000000e+00 | 4.000000e+00 | 33.33%| 29.05%
>>
>> 26.0s| 1500 | 508 | 14955 | 10.0 | 23M | 144 | 156 | 18 | 234 |7747 | 2
>> | 0 | 121 | 3.000000e+00 | 4.000000e+00 | 33.33%| 29.05%
>>
>> 27.0s| 1600 | 550 | 16127 | 10.1 | 23M | 144 | 156 | 18 | 254 |8241 | 2
>> | 0 | 121 | 3.000000e+00 | 4.000000e+00 | 33.33%| 29.05%
>>
>> 28.0s| 1700 | 594 | 17362 | 10.2 | 25M | 144 | 156 | 18 | 181 |8685 | 2
>> | 0 | 121 | 3.000000e+00 | 4.000000e+00 | 33.33%| 29.05%
>>
>> 29.0s| 1800 | 612 | 20188 | 11.2 | 25M | 144 | 156 | 18 | 341 |9373 | 2
>> | 0 | 126 | 3.000000e+00 | 4.000000e+00 | 33.33%| 29.05%
>>
>> 30.0s| 1900 | 646 | 21011 | 11.1 | 27M | 144 | 156 | 18 | 128 |9955 | 3
>> | 0 | 133 | 3.000000e+00 | 4.000000e+00 | 33.33%| 29.05%
>>
>> 30.0s| 2000 | 662 | 22015 | 11.0 | 27M | 144 | 156 | 18 | 228 | 10k| 2 |
>> 0 | 139 | 3.000000e+00 | 4.000000e+00 | 33.33%| 29.05%
>>
>> 31.0s| 2100 | 666 | 23205 | 11.0 | 28M | 144 | 156 | 18 | 184 | 11k| 2 |
>> 0 | 144 | 3.000000e+00 | 4.000000e+00 | 33.33%| 29.05%
>>
>> 32.0s| 2200 | 662 | 25329 | 11.5 | 28M | 149 | 156 | 18 | 346 | 12k| 2 |
>> 0 | 144 | 3.000000e+00 | 4.000000e+00 | 33.33%| 29.05%
>>
>> 33.0s| 2300 | 656 | 27038 | 11.7 | 28M | 149 | 156 | 18 | 52 | 12k| 2 |
>> 0 | 145 | 3.000000e+00 | 4.000000e+00 | 33.33%| 33.64%
>>
>> 34.0s| 2400 | 656 | 28141 | 11.7 | 28M | 149 | 156 | 18 | 125 | 13k| 3 |
>> 0 | 145 | 3.000000e+00 | 4.000000e+00 | 33.33%| 33.64%
>>
>> 34.0s| 2500 | 656 | 29170 | 11.7 | 28M | 149 | 156 | 18 | 176 | 13k| 3 |
>> 0 | 145 | 3.000000e+00 | 4.000000e+00 | 33.33%| 33.64%
>>
>> 35.0s| 2600 | 658 | 30550 | 11.7 | 28M | 149 | 156 | 18 | 28 | 14k| 1 |
>> 0 | 145 | 3.000000e+00 | 4.000000e+00 | 33.33%| 35.93%
>>
>> time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows
>> |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl.
>>
>> 36.0s| 2700 | 658 | 31593 | 11.7 | 28M | 149 | 156 | 18 | 28 | 15k| 1 |
>> 0 | 145 | 3.000000e+00 | 4.000000e+00 | 33.33%| 35.93%
>>
>> 36.0s| 2800 | 658 | 32449 | 11.6 | 28M | 149 | 156 | 18 | 28 | 15k| 1 |
>> 0 | 145 | 3.000000e+00 | 4.000000e+00 | 33.33%| 35.93%
>>
>> 37.0s| 2900 | 660 | 33411 | 11.5 | 28M | 149 | 156 | 18 | 39 | 16k| 3 |
>> 0 | 146 | 3.000000e+00 | 4.000000e+00 | 33.33%| 37.08%
>>
>> 38.0s| 3000 | 660 | 34321 | 11.4 | 28M | 149 | 156 | 18 | 39 | 16k| 3 |
>> 0 | 146 | 3.000000e+00 | 4.000000e+00 | 33.33%| 37.08%
>>
>> 38.0s| 3100 | 660 | 35265 | 11.4 | 28M | 149 | 156 | 18 | 39 | 17k| 3 |
>> 0 | 146 | 3.000000e+00 | 4.000000e+00 | 33.33%| 37.08%
>>
>> 39.0s| 3200 | 662 | 36354 | 11.4 | 28M | 149 | 156 | 18 | 43 | 17k| 3 |
>> 0 | 147 | 3.000000e+00 | 4.000000e+00 | 33.33%| 37.65%
>>
>> 40.0s| 3300 | 662 | 37378 | 11.3 | 28M | 149 | 156 | 18 | 43 | 18k| 3 |
>> 0 | 147 | 3.000000e+00 | 4.000000e+00 | 33.33%| 37.65%
>>
>> 40.0s| 3400 | 666 | 38304 | 11.3 | 28M | 149 | 156 | 18 | 31 | 18k| 2 |
>> 0 | 147 | 3.000000e+00 | 4.000000e+00 | 33.33%| 37.79%
>>
>> 41.0s| 3500 | 666 | 39331 | 11.2 | 28M | 149 | 156 | 18 | 31 | 19k| 1 |
>> 0 | 147 | 3.000000e+00 | 4.000000e+00 | 33.33%| 37.94%
>>
>> 42.0s| 3600 | 666 | 40398 | 11.2 | 28M | 149 | 156 | 18 | 31 | 20k| 1 |
>> 0 | 147 | 3.000000e+00 | 4.000000e+00 | 33.33%| 37.94%
>>
>> 43.0s| 3700 | 668 | 41433 | 11.2 | 28M | 149 | 156 | 18 | 44 | 20k| 3 |
>> 0 | 147 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.08%
>>
>> 43.0s| 3800 | 668 | 42458 | 11.2 | 28M | 149 | 156 | 18 | 44 | 21k| 3 |
>> 0 | 147 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.08%
>>
>> 44.0s| 3900 | 668 | 43553 | 11.2 | 28M | 149 | 156 | 18 | 48 | 21k| 3 |
>> 0 | 147 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.08%
>>
>> 45.0s| 4000 | 670 | 44720 | 11.2 | 28M | 149 | 156 | 18 | 51 | 22k| 3 |
>> 0 | 148 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.15%
>>
>> 45.0s| 4100 | 670 | 45986 | 11.2 | 28M | 149 | 156 | 18 | 54 | 23k| 3 |
>> 0 | 148 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.15%
>>
>> time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows
>> |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl.
>>
>> 46.0s| 4200 | 670 | 47087 | 11.2 | 28M | 149 | 156 | 18 | 49 | 23k| 3 |
>> 0 | 148 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.15%
>>
>> 47.0s| 4300 | 672 | 48328 | 11.2 | 28M | 149 | 156 | 18 | 37 | 24k| 1 |
>> 0 | 149 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.19%
>>
>> 47.0s| 4400 | 672 | 49605 | 11.3 | 28M | 149 | 156 | 18 | 39 | 25k| 1 |
>> 0 | 149 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.19%
>>
>> 48.0s| 4500 | 672 | 50695 | 11.3 | 28M | 149 | 156 | 18 | 48 | 25k| 3 |
>> 0 | 149 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.20%
>>
>> 49.0s| 4600 | 672 | 51925 | 11.3 | 28M | 149 | 156 | 18 | 48 | 26k| 3 |
>> 0 | 149 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.20%
>>
>> 49.0s| 4700 | 672 | 53120 | 11.3 | 28M | 149 | 156 | 18 | 49 | 26k| 3 |
>> 0 | 149 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.20%
>>
>> 50.0s| 4800 | 674 | 54467 | 11.3 | 28M | 149 | 156 | 18 | 56 | 27k| 3 |
>> 0 | 149 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.21%
>>
>> 51.0s| 4900 | 674 | 55967 | 11.4 | 29M | 149 | 156 | 18 | 56 | 28k| 3 |
>> 0 | 149 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.21%
>>
>> 52.0s| 5000 | 674 | 56922 | 11.4 | 29M | 149 | 156 | 18 | 52 | 28k| 3 |
>> 0 | 149 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.21%
>>
>> 52.0s| 5100 | 678 | 57987 | 11.4 | 29M | 149 | 156 | 18 | 45 | 29k| 1 |
>> 0 | 150 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 53.0s| 5200 | 678 | 59000 | 11.3 | 29M | 149 | 156 | 18 | 45 | 29k| 1 |
>> 0 | 150 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 54.0s| 5300 | 678 | 60023 | 11.3 | 29M | 149 | 156 | 18 | 60 | 30k| 3 |
>> 0 | 151 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 54.0s| 5400 | 678 | 61132 | 11.3 | 29M | 149 | 156 | 18 | 60 | 30k| 3 |
>> 0 | 151 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 55.0s| 5500 | 678 | 62193 | 11.3 | 29M | 149 | 156 | 18 | 60 | 31k| 3 |
>> 0 | 151 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 56.0s| 5600 | 680 | 63284 | 11.3 | 29M | 149 | 156 | 18 | 50 | 31k| 1 |
>> 0 | 151 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows
>> |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl.
>>
>> 56.0s| 5700 | 680 | 64457 | 11.3 | 29M | 149 | 156 | 18 | 50 | 32k| 1 |
>> 0 | 151 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 57.0s| 5800 | 680 | 65597 | 11.3 | 29M | 149 | 156 | 18 | 50 | 33k| 1 |
>> 0 | 151 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 58.0s| 5900 | 680 | 66775 | 11.3 | 29M | 149 | 156 | 18 | 62 | 33k| 3 |
>> 0 | 151 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 58.0s| 6000 | 680 | 67907 | 11.3 | 29M | 149 | 156 | 18 | 62 | 34k| 3 |
>> 0 | 151 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 59.0s| 6100 | 684 | 69117 | 11.3 | 31M | 149 | 156 | 18 | 59 | 34k| 1 |
>> 0 | 152 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 60.0s| 6200 | 684 | 70320 | 11.3 | 31M | 149 | 156 | 18 | 59 | 35k| 1 |
>> 0 | 152 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 60.0s| 6300 | 684 | 71343 | 11.3 | 31M | 149 | 156 | 18 | 59 | 35k| 1 |
>> 0 | 152 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 61.0s| 6400 | 684 | 72784 | 11.4 | 31M | 149 | 156 | 18 | 67 | 36k| 3 |
>> 0 | 152 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 62.0s| 6500 | 684 | 74011 | 11.4 | 31M | 149 | 156 | 18 | 67 | 36k| 3 |
>> 0 | 152 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 63.0s| 6600 | 686 | 75062 | 11.4 | 31M | 149 | 156 | 18 | 70 | 37k| 3 |
>> 0 | 153 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 63.0s| 6700 | 686 | 76130 | 11.4 | 31M | 149 | 156 | 18 | 70 | 38k| 3 |
>> 0 | 153 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 64.0s| 6800 | 686 | 77023 | 11.3 | 31M | 149 | 156 | 18 | 68 | 38k| 2 |
>> 0 | 153 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 65.0s| 6900 | 688 | 78183 | 11.3 | 31M | 149 | 156 | 18 | 61 | 39k| 1 |
>> 0 | 153 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 65.0s| 7000 | 688 | 79338 | 11.3 | 31M | 149 | 156 | 18 | 61 | 39k| 1 |
>> 0 | 153 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 66.0s| 7100 | 690 | 80535 | 11.3 | 31M | 149 | 156 | 18 | 76 | 40k| 3 |
>> 0 | 154 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows
>> |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl.
>>
>> 67.0s| 7200 | 690 | 81803 | 11.4 | 31M | 149 | 156 | 18 | 116 | 40k| 3 |
>> 0 | 154 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 67.0s| 7300 | 690 | 82707 | 11.3 | 31M | 149 | 156 | 18 | 117 | 41k| 3 |
>> 0 | 154 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 68.0s| 7400 | 694 | 83671 | 11.3 | 31M | 149 | 156 | 18 | 62 | 41k| 1 |
>> 0 | 155 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 69.0s| 7500 | 694 | 84489 | 11.3 | 31M | 149 | 156 | 18 | 63 | 42k| 1 |
>> 0 | 155 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 69.0s| 7600 | 696 | 85576 | 11.3 | 31M | 149 | 156 | 18 | 76 | 43k| 3 |
>> 0 | 155 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 70.0s| 7700 | 696 | 86825 | 11.3 | 31M | 149 | 156 | 18 | 80 | 43k| 3 |
>> 0 | 155 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 71.0s| 7800 | 696 | 87893 | 11.3 | 31M | 149 | 156 | 18 | 80 | 44k| 3 |
>> 0 | 155 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 72.0s| 7900 | 698 | 88866 | 11.2 | 33M | 149 | 156 | 18 | 65 | 44k| 1 |
>> 0 | 155 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 72.0s| 8000 | 698 | 89659 | 11.2 | 33M | 149 | 156 | 18 | 66 | 45k| 1 |
>> 0 | 155 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 73.0s| 8100 | 700 | 90468 | 11.2 | 33M | 149 | 156 | 18 | 78 | 45k| 3 |
>> 0 | 156 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 74.0s| 8200 | 700 | 91324 | 11.1 | 33M | 149 | 156 | 18 | 83 | 46k| 3 |
>> 0 | 156 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 74.0s| 8300 | 702 | 92294 | 11.1 | 33M | 149 | 156 | 18 | 83 | 46k| 3 |
>> 0 | 156 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 75.0s| 8400 | 702 | 93181 | 11.1 | 33M | 149 | 156 | 18 | 82 | 47k| 3 |
>> 0 | 190 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 76.0s| 8500 | 702 | 94139 | 11.1 | 33M | 149 | 156 | 18 | 82 | 48k| 3 |
>> 0 | 198 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 77.0s| 8600 | 704 | 95255 | 11.1 | 34M | 149 | 156 | 18 | 86 | 48k| 3 |
>> 0 | 200 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows
>> |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl.
>>
>> 78.0s| 8700 | 704 | 96224 | 11.1 | 34M | 149 | 156 | 18 | 86 | 49k| 3 |
>> 0 | 200 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 78.0s| 8800 | 706 | 97230 | 11.0 | 34M | 149 | 156 | 18 | 76 | 49k| 1 |
>> 0 | 200 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 79.0s| 8900 | 706 | 98053 | 11.0 | 34M | 149 | 156 | 18 | 76 | 50k| 1 |
>> 0 | 224 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 80.0s| 9000 | 706 | 99030 | 11.0 | 34M | 149 | 156 | 18 | 86 | 50k| 3 |
>> 0 | 224 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 81.0s| 9100 | 706 | 99830 | 11.0 | 35M | 149 | 156 | 18 | 86 | 51k| 3 |
>> 0 | 224 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 81.0s| 9200 | 706 |100740 | 10.9 | 35M | 149 | 156 | 18 | 87 | 52k| 3 |
>> 0 | 224 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 82.0s| 9300 | 710 |101768 | 10.9 | 35M | 149 | 156 | 18 | 76 | 52k| 1 |
>> 0 | 224 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 83.0s| 9400 | 710 |102664 | 10.9 | 35M | 149 | 156 | 18 | 77 | 53k| 1 |
>> 0 | 224 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 83.0s| 9500 | 712 |103630 | 10.9 | 35M | 149 | 156 | 18 | 91 | 53k| 3 |
>> 0 | 224 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 84.0s| 9600 | 712 |104370 | 10.9 | 35M | 149 | 156 | 18 | 89 | 54k| 2 |
>> 0 | 224 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 85.0s| 9700 | 714 |105400 | 10.9 | 35M | 149 | 156 | 18 | 79 | 54k| 1 |
>> 0 | 224 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 86.0s| 9800 | 714 |106416 | 10.9 | 35M | 149 | 156 | 18 | 79 | 55k| 1 |
>> 0 | 224 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 86.0s| 9900 | 714 |107245 | 10.8 | 35M | 149 | 156 | 18 | 80 | 56k| 1 |
>> 0 | 224 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 87.0s| 10000 | 716 |108308 | 10.8 | 36M | 149 | 156 | 18 | 91 | 56k| 3 |
>> 0 | 224 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 88.0s| 10100 | 716 |109126 | 10.8 | 36M | 149 | 156 | 18 | 90 | 57k| 2 |
>> 0 | 224 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows
>> |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl.
>>
>> 88.0s| 10200 | 720 |110163 | 10.8 | 36M | 149 | 156 | 18 | 81 | 57k| 1 |
>> 0 | 224 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 89.0s| 10300 | 720 |111050 | 10.8 | 36M | 149 | 156 | 18 | 82 | 58k| 1 |
>> 0 | 224 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 90.0s| 10400 | 722 |112029 | 10.8 | 36M | 149 | 156 | 18 | 93 | 58k| 3 |
>> 0 | 224 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 90.0s| 10500 | 722 |112977 | 10.8 | 36M | 149 | 156 | 18 | 94 | 59k| 3 |
>> 0 | 224 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 91.0s| 10600 | 726 |113957 | 10.7 | 36M | 149 | 156 | 18 | 86 | 59k| 1 |
>> 0 | 225 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 92.0s| 10700 | 726 |115069 | 10.8 | 36M | 149 | 156 | 18 | 89 | 60k| 1 |
>> 0 | 225 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 92.0s| 10800 | 728 |116011 | 10.7 | 37M | 149 | 156 | 18 | 99 | 60k| 3 |
>> 0 | 225 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 93.0s| 10900 | 728 |117107 | 10.7 | 37M | 149 | 156 | 18 | 103 | 61k| 3
>> | 0 | 225 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 94.0s| 11000 | 730 |118218 | 10.7 | 37M | 149 | 156 | 18 | 103 | 62k| 3
>> | 0 | 226 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 94.0s| 11100 | 730 |119418 | 10.8 | 37M | 149 | 156 | 18 | 106 | 62k| 3
>> | 0 | 226 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 95.0s| 11200 | 730 |120532 | 10.8 | 37M | 149 | 156 | 18 | 109 | 63k| 3
>> | 0 | 226 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 96.0s| 11300 | 734 |121727 | 10.8 | 37M | 149 | 156 | 18 | 95 | 63k| 1 |
>> 0 | 226 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 97.0s| 11400 | 734 |122717 | 10.8 | 37M | 149 | 156 | 18 | 95 | 64k| 1 |
>> 0 | 226 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 97.0s| 11500 | 734 |123765 | 10.8 | 37M | 149 | 156 | 18 | 103 | 64k| 3
>> | 0 | 226 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> 98.0s| 11600 | 734 |124795 | 10.8 | 37M | 149 | 156 | 18 | 108 | 65k| 3
>> | 0 | 226 | 3.000000e+00 | 4.000000e+00 | 33.33%| 38.22%
>>
>> time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows
>> |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl.
>>
>> *98.0s| 11622 | 0 |125035 | 10.8 | LP | 149 | 156 | 18 | 106 | 65k| 3 |
>> 0 | 226 | 3.000000e+00 | 3.000000e+00 | 0.00%| 100.00%
>>
>> I want to make SCIP calculation faster, can it be achieved by changing
>> the parameters?
>>
>> And can I speed up the calculation of SCIP through a good initialization
>> strategy?
>>
>> Best,
>>
>> Jax
>>
>> 发送自Windows 10 版邮件<https://go.microsoft.com/fwlink/?LinkId=550986>应用
>>
>>
>> _______________________________________________
>> Scip mailing list
>> Scip at zib.de
>> https://listserv.zib.de/mailman/listinfo/scip
>>
>
>
> _______________________________________________
> Scip mailing list
> Scip at zib.de
> https://listserv.zib.de/mailman/listinfo/scip
>
--
Ambros Gleixner, Research Group Mathematical Optimization Methods at
Zuse Institute Berlin, http://www.zib.de/gleixner
More information about the Scip
mailing list