Scippy

SCIP

Solving Constraint Integer Programs

heur_proximity.c
Go to the documentation of this file.
1 /* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */
2 /* */
3 /* This file is part of the program and library */
4 /* SCIP --- Solving Constraint Integer Programs */
5 /* */
6 /* Copyright (c) 2002-2023 Zuse Institute Berlin (ZIB) */
7 /* */
8 /* Licensed under the Apache License, Version 2.0 (the "License"); */
9 /* you may not use this file except in compliance with the License. */
10 /* You may obtain a copy of the License at */
11 /* */
12 /* http://www.apache.org/licenses/LICENSE-2.0 */
13 /* */
14 /* Unless required by applicable law or agreed to in writing, software */
15 /* distributed under the License is distributed on an "AS IS" BASIS, */
16 /* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. */
17 /* See the License for the specific language governing permissions and */
18 /* limitations under the License. */
19 /* */
20 /* You should have received a copy of the Apache-2.0 license */
21 /* along with SCIP; see the file LICENSE. If not visit scipopt.org. */
22 /* */
23 /* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */
24 
25 /**@file heur_proximity.c
26  * @ingroup DEFPLUGINS_HEUR
27  * @brief improvement heuristic which uses an auxiliary objective instead of the original objective function which
28  * is itself added as a constraint to a sub-SCIP instance. The heuristic was presented by Matteo Fischetti
29  * and Michele Monaci.
30  * @author Gregor Hendel
31  */
32 
33 /*---+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8----+----9----+----0----+----1----+----2*/
34 
35 #include "blockmemshell/memory.h"
36 #include "scip/cons_linear.h"
37 #include "scip/heuristics.h"
38 #include "scip/heur_proximity.h"
39 #include "scip/pub_event.h"
40 #include "scip/pub_heur.h"
41 #include "scip/pub_message.h"
42 #include "scip/pub_misc.h"
43 #include "scip/pub_sol.h"
44 #include "scip/pub_var.h"
45 #include "scip/scip_branch.h"
46 #include "scip/scip_cons.h"
47 #include "scip/scip_copy.h"
48 #include "scip/scip_event.h"
49 #include "scip/scip_general.h"
50 #include "scip/scip_heur.h"
51 #include "scip/scip_lp.h"
52 #include "scip/scip_mem.h"
53 #include "scip/scip_message.h"
54 #include "scip/scip_nlp.h"
55 #include "scip/scip_nodesel.h"
56 #include "scip/scip_numerics.h"
57 #include "scip/scip_param.h"
58 #include "scip/scip_prob.h"
59 #include "scip/scip_sol.h"
60 #include "scip/scip_solve.h"
61 #include "scip/scip_solvingstats.h"
62 #include "scip/scip_timing.h"
63 #include "scip/scip_var.h"
64 #include <string.h>
65 
66 #define HEUR_NAME "proximity"
67 #define HEUR_DESC "heuristic trying to improve the incumbent by an auxiliary proximity objective function"
68 #define HEUR_DISPCHAR SCIP_HEURDISPCHAR_LNS
69 #define HEUR_PRIORITY -2000000
70 #define HEUR_FREQ -1
71 #define HEUR_FREQOFS 0
72 #define HEUR_MAXDEPTH -1
73 #define HEUR_TIMING SCIP_HEURTIMING_AFTERNODE
74 #define HEUR_USESSUBSCIP TRUE /**< does the heuristic use a secondary SCIP instance? */
75 
76 /* event handler properties */
77 #define EVENTHDLR_NAME "Proximity"
78 #define EVENTHDLR_DESC "LP event handler for " HEUR_NAME " heuristic"
79 
80 /* default values for proximity-specific parameters */
81 /* todo refine these values */
82 #define DEFAULT_MAXNODES 10000LL /**< maximum number of nodes to regard in the subproblem */
83 #define DEFAULT_MINIMPROVE 0.02 /**< factor by which proximity should at least improve the incumbent */
84 #define DEFAULT_MINGAP 0.01 /**< minimum primal-dual gap for which the heuristic is executed */
85 #define DEFAULT_MINNODES 1LL /**< minimum number of nodes to regard in the subproblem */
86 #define DEFAULT_MINLPITERS 200LL /**< minimum number of LP iterations to perform in one sub-mip */
87 #define DEFAULT_MAXLPITERS 100000LL /**< maximum number of LP iterations to be performed in the subproblem */
88 #define DEFAULT_NODESOFS 50LL /**< number of nodes added to the contingent of the total nodes */
89 #define DEFAULT_WAITINGNODES 100LL /**< default waiting nodes since last incumbent before heuristic is executed */
90 #define DEFAULT_NODESQUOT 0.1 /**< default quotient of sub-MIP nodes with respect to number of processed nodes*/
91 #define DEFAULT_USELPROWS FALSE /**< should subproblem be constructed based on LP row information? */
92 #define DEFAULT_BINVARQUOT 0.1 /**< default threshold for percentage of binary variables required to start */
93 #define DEFAULT_RESTART TRUE /**< should the heuristic immediately run again on its newly found solution? */
94 #define DEFAULT_USEFINALLP FALSE /**< should the heuristic solve a final LP in case of continuous objective variables? */
95 #define DEFAULT_LPITERSQUOT 0.2 /**< default quotient of sub-MIP LP iterations with respect to LP iterations so far */
96 #define DEFAULT_USEUCT FALSE /**< should uct node selection be used at the beginning of the search? */
97 
98 /*
99  * Data structures
100  */
101 
102 /** primal heuristic data */
103 struct SCIP_HeurData
104 {
105  SCIP_Longint maxnodes; /**< maximum number of nodes to regard in the subproblem */
106  SCIP_Longint minnodes; /**< minimum number of nodes to regard in the subproblem */
107  SCIP_Longint maxlpiters; /**< maximum number of LP iterations to be performed in the subproblem */
108  SCIP_Longint nusedlpiters; /**< number of actually performed LP iterations */
109  SCIP_Longint minlpiters; /**< minimum number of LP iterations to perform in one sub-mip */
110  SCIP_Longint nodesofs; /**< number of nodes added to the contingent of the total nodes */
111  SCIP_Longint usednodes; /**< nodes already used by proximity in earlier calls */
112  SCIP_Longint waitingnodes; /**< waiting nodes since last incumbent before heuristic is executed */
113  SCIP_Real lpitersquot; /**< quotient of sub-MIP LP iterations with respect to LP iterations so far */
114  SCIP_Real minimprove; /**< factor by which proximity should at least improve the incumbent */
115  SCIP_Real mingap; /**< minimum primal-dual gap for which the heuristic is executed */
116  SCIP_Real nodesquot; /**< quotient of sub-MIP nodes with respect to number of processed nodes */
117  SCIP_Real binvarquot; /**< threshold for percantage of binary variables required to start */
118 
119  SCIP* subscip; /**< the subscip used by the heuristic */
120  SCIP_HASHMAP* varmapfw; /**< map between scip variables and subscip variables */
121  SCIP_VAR** subvars; /**< variables in subscip */
122  SCIP_CONS* objcons; /**< the objective cutoff constraint of the subproblem */
123 
124  int nsubvars; /**< the number of subvars */
125  int lastsolidx; /**< index of last solution on which the heuristic was processed */
126  int subprobidx; /**< counter for the subproblem index to be solved by proximity */
127 
128  SCIP_Bool uselprows; /**< should subproblem be constructed based on LP row information? */
129  SCIP_Bool restart; /**< should the heuristic immediately run again on its newly found solution? */
130  SCIP_Bool usefinallp; /**< should the heuristic solve a final LP in case of continuous objective variables? */
131  SCIP_Bool useuct; /**< should uct node selection be used at the beginning of the search? */
132 };
133 
134 
135 /*
136  * Local methods
137  */
138 
139 /** optimizes the continuous variables in an LP diving by fixing all integer variables to the given solution values */
140 static
142  SCIP* scip, /**< SCIP data structure */
143  SCIP_SOL* sol, /**< candidate solution for which continuous variables should be optimized */
144  SCIP_Bool* success /**< was the dive successful? */
145  )
146 {
147  SCIP_VAR** vars;
148  SCIP_RETCODE retstat;
149 
150  int v;
151  int nvars;
152  int ncontvars;
153  int nintvars;
154 
155  SCIP_Bool lperror;
156  SCIP_Bool requiresnlp;
157 
158  assert(success != NULL);
159 
160  SCIP_CALL( SCIPgetVarsData(scip, &vars, &nvars, NULL, NULL, NULL, &ncontvars) );
161 
162  nintvars = nvars - ncontvars;
163 
164  /**@todo in case of an MINLP, if SCIPisNLPConstructed() is TRUE rather solve the NLP instead of the LP */
165  requiresnlp = SCIPisNLPConstructed(scip);
166  if( requiresnlp || ncontvars == 0 )
167  return SCIP_OKAY;
168 
169  /* start diving to calculate the LP relaxation */
170  SCIP_CALL( SCIPstartDive(scip) );
171 
172  /* set the bounds of the variables: fixed for integers, global bounds for continuous */
173  for( v = 0; v < nvars; ++v )
174  {
175  if( SCIPvarGetStatus(vars[v]) == SCIP_VARSTATUS_COLUMN )
176  {
177  SCIP_CALL( SCIPchgVarLbDive(scip, vars[v], SCIPvarGetLbGlobal(vars[v])) );
178  SCIP_CALL( SCIPchgVarUbDive(scip, vars[v], SCIPvarGetUbGlobal(vars[v])) );
179  }
180  }
181 
182  /* apply this after global bounds to not cause an error with intermediate empty domains */
183  for( v = 0; v < nintvars; ++v )
184  {
185  if( SCIPvarGetStatus(vars[v]) == SCIP_VARSTATUS_COLUMN )
186  {
187  SCIP_Real solval;
188 
189  solval = SCIPgetSolVal(scip, sol, vars[v]);
190  SCIP_CALL( SCIPchgVarLbDive(scip, vars[v], solval) );
191  SCIP_CALL( SCIPchgVarUbDive(scip, vars[v], solval) );
192  }
193  }
194 
195  /* solve LP */
196  SCIPdebugMsg(scip, " -> old LP iterations: %" SCIP_LONGINT_FORMAT "\n", SCIPgetNLPIterations(scip));
197 
198  /* Errors in the LP solver should not kill the overall solving process, if the LP is just needed for a heuristic.
199  * Hence in optimized mode, the return code is caught and a warning is printed, only in debug mode, SCIP will stop.
200  */
201  retstat = SCIPsolveDiveLP(scip, -1, &lperror, NULL);
202  if( retstat != SCIP_OKAY )
203  {
204 #ifdef NDEBUG
205  SCIPwarningMessage(scip, "Error while solving LP in Proximity heuristic; LP solve terminated with code <%d>\n",retstat);
206 #else
207  SCIP_CALL( retstat );
208 #endif
209  }
210 
211  SCIPdebugMsg(scip, " -> new LP iterations: %" SCIP_LONGINT_FORMAT "\n", SCIPgetNLPIterations(scip));
212  SCIPdebugMsg(scip, " -> error=%u, status=%d\n", lperror, SCIPgetLPSolstat(scip));
213  if( !lperror && SCIPgetLPSolstat(scip) == SCIP_LPSOLSTAT_OPTIMAL )
214  {
215  SCIP_CALL( SCIPlinkLPSol(scip, sol) );
216  SCIP_CALL( SCIPtrySol(scip, sol, FALSE, FALSE, TRUE, TRUE, TRUE, success) );
217  }
218 
219  /* terminate diving mode */
220  SCIP_CALL( SCIPendDive(scip) );
221 
222  return SCIP_OKAY;
223 }
224 
225 /** creates a new solution for the original problem by copying the solution of the subproblem */
226 static
228  SCIP* scip, /**< original SCIP data structure */
229  SCIP* subscip, /**< SCIP structure of the subproblem */
230  SCIP_VAR** subvars, /**< the variables of the subproblem */
231  SCIP_HEUR* heur, /**< proximity heuristic structure */
232  SCIP_SOL* subsol, /**< solution of the subproblem */
233  SCIP_Bool usefinallp, /**< should continuous variables be optimized by a final LP */
234  SCIP_Bool* success /**< used to store whether new solution was found or not */
235  )
236 {
237  SCIP_VAR** vars; /* the original problem's variables */
238  int nvars; /* the original problem's number of variables */
239  int ncontvars; /* the original problem's number of continuous variables */
240  SCIP_Real* subsolvals; /* solution values of the subproblem */
241  SCIP_SOL* newsol; /* solution to be created for the original problem */
242  int i;
243 
244  assert(scip != NULL);
245  assert(subscip != NULL);
246  assert(subvars != NULL);
247  assert(subsol != NULL);
248  assert(success != NULL);
249 
250  /* get variables' data */
251  SCIP_CALL( SCIPgetVarsData(scip, &vars, &nvars, NULL, NULL, NULL, &ncontvars) );
252 
253  SCIP_CALL( SCIPallocBufferArray(scip, &subsolvals, nvars) );
254 
255  /* copy the solution */
256  for( i = 0; i < nvars; ++i )
257  {
258  if( subvars[i] == NULL )
259  subsolvals[i] = MIN(MAX(0.0, SCIPvarGetLbLocal(vars[i])), SCIPvarGetUbLocal(vars[i])); /*lint !e666*/
260  else
261  subsolvals[i] = SCIPgetSolVal(subscip, subsol, subvars[i]);
262  }
263 
264  /* create new solution for the original problem */
265  SCIP_CALL( SCIPcreateSol(scip, &newsol, heur) );
266  SCIP_CALL( SCIPsetSolVals(scip, newsol, nvars, vars, subsolvals) );
267 
268  *success = FALSE;
269 
270  /* solve an LP with all integer variables fixed to improve solution quality */
271  if( ncontvars > 0 && usefinallp && SCIPisLPConstructed(scip) )
272  {
273  int v;
274  int ncontobjvars = 0; /* does the problem instance have continuous variables with nonzero objective coefficients? */
275  SCIP_Real sumofobjsquares = 0.0;
276 
277  /* check if continuous variables with nonzero objective coefficient are present */
278  for( v = nvars - 1; v >= nvars - ncontvars; --v )
279  {
280  SCIP_VAR* var;
281 
282  var = vars[v];
283  assert(vars[v] != NULL);
284  assert(!SCIPvarIsIntegral(var));
285 
286  if( SCIPvarGetStatus(var) == SCIP_VARSTATUS_COLUMN && !SCIPisZero(scip, SCIPvarGetObj(var)) )
287  {
288  ++ncontobjvars;
289  sumofobjsquares += SCIPvarGetObj(var) * SCIPvarGetObj(var);
290  }
291  }
292 
293  SCIPstatisticMessage(" Continuous Objective variables: %d, Euclidean OBJ: %g total, %g continuous\n", ncontobjvars, SCIPgetObjNorm(scip), sumofobjsquares);
294 
295  /* solve a final LP to optimize solution values of continuous problem variables */
296  SCIPstatisticMessage("Solution Value before LP resolve: %g\n", SCIPgetSolOrigObj(scip, newsol));
297  SCIP_CALL( solveLp(scip, newsol, success) );
298 
299  /* if the LP solve was not successful, reset the solution */
300  if( !*success )
301  {
302  for( v = nvars - 1; v >= nvars - ncontvars; --v )
303  {
304  SCIP_CALL( SCIPsetSolVal(scip, newsol, vars[v], subsolvals[v]) );
305  }
306  }
307  }
308 
309  /* try to add new solution to SCIP and free it immediately */
310  if( !*success )
311  {
312  SCIP_CALL( SCIPtrySol(scip, newsol, FALSE, FALSE, TRUE, TRUE, TRUE, success) );
313  }
314  SCIP_CALL( SCIPfreeSol(scip, &newsol) );
315 
316  SCIPfreeBufferArray(scip, &subsolvals);
317 
318  return SCIP_OKAY;
319 }
320 
321 /** sets solving parameters for the subproblem created by the heuristic */
322 static
324  SCIP_HEURDATA* heurdata, /**< heuristic data structure */
325  SCIP* subscip /**< copied SCIP data structure */
326  )
327 {
328  assert(subscip != NULL);
329 
330  /* do not abort subproblem on CTRL-C */
331  SCIP_CALL( SCIPsetBoolParam(subscip, "misc/catchctrlc", FALSE) );
332 
333 #ifdef SCIP_DEBUG
334  /* for debugging, enable full output */
335  SCIP_CALL( SCIPsetIntParam(subscip, "display/verblevel", 5) );
336  SCIP_CALL( SCIPsetIntParam(subscip, "display/freq", 100000000) );
337 #else
338  /* disable statistic timing inside sub SCIP and output to console */
339  SCIP_CALL( SCIPsetIntParam(subscip, "display/verblevel", 0) );
340  SCIP_CALL( SCIPsetBoolParam(subscip, "timing/statistictiming", FALSE) );
341 #endif
342 
343  /* forbid recursive call of heuristics and separators solving sub-SCIPs */
344  SCIP_CALL( SCIPsetSubscipsOff(subscip, TRUE) );
345 
346  /* use restart dfs node selection */
347  if( SCIPfindNodesel(subscip, "restartdfs") != NULL && !SCIPisParamFixed(subscip, "nodeselection/restartdfs/stdpriority") )
348  {
349  SCIP_CALL( SCIPsetIntParam(subscip, "nodeselection/restartdfs/stdpriority", INT_MAX/4) );
350  }
351 
352  /* activate uct node selection at the top of the tree */
353  if( heurdata->useuct && SCIPfindNodesel(subscip, "uct") != NULL && !SCIPisParamFixed(subscip, "nodeselection/uct/stdpriority") )
354  {
355  SCIP_CALL( SCIPsetIntParam(subscip, "nodeselection/uct/stdpriority", INT_MAX/2) );
356  }
357 
358  /* disable expensive presolving
359  * todo maybe presolving can be entirely turned off here - parameter???
360  */
362 
363  /* SCIP_CALL( SCIPsetPresolving(scip, SCIP_PARAMSETTING_OFF, TRUE) ); */
364  if( !SCIPisParamFixed(subscip, "presolving/maxrounds") )
365  {
366  SCIP_CALL( SCIPsetIntParam(subscip, "presolving/maxrounds", 50) );
367  }
368 
369  /* disable cutting plane separation */
371 
372  /* todo: check branching rule in sub-SCIP */
373  if( SCIPfindBranchrule(subscip, "inference") != NULL && !SCIPisParamFixed(subscip, "branching/inference/priority") )
374  {
375  SCIP_CALL( SCIPsetIntParam(subscip, "branching/inference/priority", INT_MAX/4) );
376  }
377 
378  /* disable feasibility pump and fractional diving */
379  if( !SCIPisParamFixed(subscip, "heuristics/feaspump/freq") )
380  {
381  SCIP_CALL( SCIPsetIntParam(subscip, "heuristics/feaspump/freq", -1) );
382  }
383  if( !SCIPisParamFixed(subscip, "heuristics/fracdiving/freq") )
384  {
385  SCIP_CALL( SCIPsetIntParam(subscip, "heuristics/fracdiving/freq", -1) );
386  }
387 
388  /* todo check if
389  * SCIP_CALL( SCIPsetEmphasis(subscip, SCIP_PARAMEMPHASIS_FEASIBILITY, TRUE) );
390  * improves performance */
391 
392  return SCIP_OKAY;
393 }
394 
395 /** frees the subproblem */
396 static
398  SCIP* scip, /**< SCIP data structure */
399  SCIP_HEURDATA* heurdata /**< heuristic data */
400  )
401 {
402  /* free remaining memory from heuristic execution */
403  if( heurdata->subscip != NULL )
404  {
405  assert(heurdata->varmapfw != NULL);
406  assert(heurdata->subvars != NULL);
407  assert(heurdata->objcons != NULL);
408 
409  SCIPdebugMsg(scip, "Freeing subproblem of proximity heuristic\n");
410  SCIPfreeBlockMemoryArray(scip, &heurdata->subvars, heurdata->nsubvars);
411  SCIPhashmapFree(&heurdata->varmapfw);
412  SCIP_CALL( SCIPreleaseCons(heurdata->subscip, &heurdata->objcons) );
413  SCIP_CALL( SCIPfree(&heurdata->subscip) );
414 
415  heurdata->subscip = NULL;
416  heurdata->varmapfw = NULL;
417  heurdata->subvars = NULL;
418  heurdata->objcons = NULL;
419  }
420  return SCIP_OKAY;
421 }
422 
423 /* ---------------- Callback methods of event handler ---------------- */
424 
425 /** exec the event handler
426  *
427  * We interrupt the solution process.
428  */
429 static
430 SCIP_DECL_EVENTEXEC(eventExecProximity)
431 {
432  SCIP_HEURDATA* heurdata;
433 
434  assert(eventhdlr != NULL);
435  assert(eventdata != NULL);
436  assert(strcmp(SCIPeventhdlrGetName(eventhdlr), EVENTHDLR_NAME) == 0);
437  assert(event != NULL);
439 
440  heurdata = (SCIP_HEURDATA*)eventdata;
441  assert(heurdata != NULL);
442 
443  /* interrupt solution process of sub-SCIP
444  * todo adjust interruption limit */
445  if( SCIPgetLPSolstat(scip) == SCIP_LPSOLSTAT_ITERLIMIT || SCIPgetNLPIterations(scip) >= heurdata->maxlpiters )
446  {
448  }
449 
450  return SCIP_OKAY;
451 }
452 
453 
454 /* ---------------- Callback methods of primal heuristic ---------------- */
455 
456 /** copy method for primal heuristic plugins (called when SCIP copies plugins) */
457 static
458 SCIP_DECL_HEURCOPY(heurCopyProximity)
459 { /*lint --e{715}*/
460  assert(scip != NULL);
461  assert(heur != NULL);
462  assert(strcmp(SCIPheurGetName(heur), HEUR_NAME) == 0);
463 
464  /* call inclusion method of primal heuristic */
466 
467  return SCIP_OKAY;
468 }
469 
470 /** destructor of primal heuristic to free user data (called when SCIP is exiting) */
471 static
472 SCIP_DECL_HEURFREE(heurFreeProximity)
473 { /*lint --e{715}*/
474  SCIP_HEURDATA* heurdata;
475 
476  assert( heur != NULL );
477  assert( scip != NULL );
478 
479  /* get heuristic data */
480  heurdata = SCIPheurGetData(heur);
481  assert( heurdata != NULL );
482 
483  /* free heuristic data */
484  SCIPfreeBlockMemory(scip, &heurdata);
485  SCIPheurSetData(heur, NULL);
486 
487  return SCIP_OKAY;
488 }
489 
490 
491 /** initialization method of primal heuristic (called after problem was transformed) */
492 static
493 SCIP_DECL_HEURINIT(heurInitProximity)
494 { /*lint --e{715}*/
495  SCIP_HEURDATA* heurdata;
496 
497  assert( heur != NULL );
498  assert( scip != NULL );
499 
500  /* get heuristic data */
501  heurdata = SCIPheurGetData(heur);
502  assert( heurdata != NULL );
503 
504  /* initialize data */
505  heurdata->usednodes = 0LL;
506  heurdata->lastsolidx = -1;
507  heurdata->nusedlpiters = 0LL;
508  heurdata->subprobidx = 0;
509 
510  heurdata->subscip = NULL;
511  heurdata->varmapfw = NULL;
512  heurdata->subvars = NULL;
513  heurdata->objcons = NULL;
514 
515  heurdata->nsubvars = 0;
516 
517  return SCIP_OKAY;
518 }
519 
520 /** solution process exiting method of proximity heuristic */
521 static
522 SCIP_DECL_HEUREXITSOL(heurExitsolProximity)
523 {
524  SCIP_HEURDATA* heurdata;
525 
526  assert( heur != NULL );
527  assert( scip != NULL );
528 
529  /* get heuristic data */
530  heurdata = SCIPheurGetData(heur);
531  assert( heurdata != NULL );
532 
533  SCIP_CALL( deleteSubproblem(scip, heurdata) );
534 
535  assert(heurdata->subscip == NULL && heurdata->varmapfw == NULL && heurdata->subvars == NULL && heurdata->objcons == NULL);
536 
537  return SCIP_OKAY;
538 }
539 
540 /** execution method of primal heuristic */
541 static
542 SCIP_DECL_HEUREXEC(heurExecProximity)
543 { /*lint --e{715}*/
544  SCIP_HEURDATA* heurdata; /* heuristic's data */
545  SCIP_Longint nnodes; /* number of stalling nodes for the subproblem */
546  SCIP_Longint nlpiters; /* lp iteration limit for the subproblem */
547  SCIP_Bool foundsol = FALSE;
548 
549  assert(heur != NULL);
550  assert(scip != NULL);
551  assert(result != NULL);
552 
553  *result = SCIP_DIDNOTRUN;
554 
555  /* get heuristic data */
556  heurdata = SCIPheurGetData(heur);
557  assert(heurdata != NULL);
558 
559  /* do not run heuristic when there are only few binary varables */
560  if( SCIPgetNBinVars(scip) < heurdata->binvarquot * SCIPgetNVars(scip) )
561  return SCIP_OKAY;
562 
563  /* calculate branching node limit for sub problem */
564  /* todo maybe treat root node differently */
565  nnodes = (SCIP_Longint) (heurdata->nodesquot * SCIPgetNNodes(scip));
566  nnodes += heurdata->nodesofs;
567 
568  /* determine the node and LP iteration limit for the solve of the sub-SCIP */
569  nnodes -= heurdata->usednodes;
570  nnodes = MIN(nnodes, heurdata->maxnodes);
571 
572  nlpiters = (SCIP_Longint) (heurdata->lpitersquot * SCIPgetNRootFirstLPIterations(scip));
573  nlpiters = MIN(nlpiters, heurdata->maxlpiters);
574 
575  /* check whether we have enough nodes left to call subproblem solving */
576  if( nnodes < heurdata->minnodes )
577  {
578  SCIPdebugMsg(scip, "skipping proximity: nnodes=%" SCIP_LONGINT_FORMAT ", minnodes=%" SCIP_LONGINT_FORMAT "\n", nnodes, heurdata->minnodes);
579  return SCIP_OKAY;
580  }
581 
582  /* do not run proximity, if the problem does not have an objective function anyway */
583  if( SCIPgetNObjVars(scip) == 0 )
584  {
585  SCIPdebugMsg(scip, "skipping proximity: pure feasibility problem anyway\n");
586  return SCIP_OKAY;
587  }
588 
589  do
590  {
591  /* main loop of proximity: in every iteration, a new subproblem is set up and solved until no improved solution
592  * is found or one of the heuristic limits on nodes or LP iterations is hit
593  * heuristic performs only one iteration if restart parameter is set to FALSE
594  */
595  SCIP_Longint nusednodes = 0LL;
596  SCIP_Longint nusedlpiters = 0LL;
597 
598  nlpiters = MAX(nlpiters, heurdata->minlpiters);
599 
600  /* define and solve the proximity subproblem */
601  SCIP_CALL( SCIPapplyProximity(scip, heur, result, heurdata->minimprove, nnodes, nlpiters, &nusednodes, &nusedlpiters, FALSE) );
602 
603  /* adjust node limit and LP iteration limit for future iterations */
604  assert(nusednodes <= nnodes);
605  heurdata->usednodes += nusednodes;
606  nnodes -= nusednodes;
607 
608  nlpiters -= nusedlpiters;
609  heurdata->nusedlpiters += nusedlpiters;
610 
611  /* memorize if a new solution has been found in at least one iteration */
612  if( *result == SCIP_FOUNDSOL )
613  foundsol = TRUE;
614  }
615  while( *result == SCIP_FOUNDSOL && heurdata->restart && !SCIPisStopped(scip) && nnodes > 0 );
616 
617  /* reset result pointer if solution has been found in previous iteration */
618  if( foundsol )
619  *result = SCIP_FOUNDSOL;
620 
621  /* free the occupied memory */
622  if( heurdata->subscip != NULL )
623  {
624  /* just for testing the library method, in debug mode, we call the wrapper method for the actual delete method */
625 #ifndef NDEBUG
627 #else
628  SCIP_CALL( deleteSubproblem(scip, heurdata) );
629 #endif
630  }
631  return SCIP_OKAY;
632 }
633 
634 
635 /*
636  * primal heuristic specific interface methods
637  */
638 
639 /** frees the sub-MIP created by proximity */
641  SCIP* scip /** SCIP data structure */
642  )
643 {
644  SCIP_HEUR* heur;
645  SCIP_HEURDATA* heurdata;
646 
647  assert(scip != NULL);
648 
649  heur = SCIPfindHeur(scip, HEUR_NAME);
650  assert(heur != NULL);
651 
652  heurdata = SCIPheurGetData(heur);
653  if( heurdata != NULL )
654  {
655  SCIP_CALL( deleteSubproblem(scip, heurdata) );
656  }
657 
658  return SCIP_OKAY;
659 }
660 
661 /** main procedure of the proximity heuristic, creates and solves a sub-SCIP
662  *
663  * @note The method can be applied in an iterative way, keeping the same subscip in between. If the @p freesubscip
664  * parameter is set to FALSE, the heuristic will keep the subscip data structures. Always set this parameter
665  * to TRUE, or call SCIPdeleteSubproblemProximity() afterwards.
666  */
668  SCIP* scip, /**< original SCIP data structure */
669  SCIP_HEUR* heur, /**< heuristic data structure */
670  SCIP_RESULT* result, /**< result data structure */
671  SCIP_Real minimprove, /**< factor by which proximity should at least improve the incumbent */
672  SCIP_Longint nnodes, /**< node limit for the subproblem */
673  SCIP_Longint nlpiters, /**< LP iteration limit for the subproblem */
674  SCIP_Longint* nusednodes, /**< pointer to store number of used nodes in subscip */
675  SCIP_Longint* nusedlpiters, /**< pointer to store number of used LP iterations in subscip */
676  SCIP_Bool freesubscip /**< should the created sub-MIP be freed at the end of the method? */
677  )
678 {
679  SCIP* subscip; /* the subproblem created by proximity */
680  SCIP_HASHMAP* varmapfw; /* mapping of SCIP variables to sub-SCIP variables */
681  SCIP_VAR** vars; /* original problem's variables */
682  SCIP_VAR** subvars; /* subproblem's variables */
683  SCIP_HEURDATA* heurdata; /* heuristic's private data structure */
684  SCIP_EVENTHDLR* eventhdlr; /* event handler for LP events */
685 
686  SCIP_SOL* incumbent;
687  SCIP_CONS* objcons;
688  SCIP_Longint iterlim;
689 
690  SCIP_Real large;
691  SCIP_Real inf;
692 
693  SCIP_Real bestobj;
694  SCIP_Real objcutoff;
695  SCIP_Real lowerbound;
696 
697  int nvars; /* number of original problem's variables */
698  int nfixedvars;
699  int nsubsols;
700  int solidx;
701  int i;
702 
703  SCIP_Bool valid;
704  SCIP_Bool success;
705 
706  assert(scip != NULL);
707  assert(heur != NULL);
708  assert(result != NULL);
709 
710  assert(nnodes >= 0);
711  assert(0.0 <= minimprove && minimprove <= 1.0);
712 
713  *result = SCIP_DIDNOTRUN;
714 
715  /* get heuristic data */
716  heurdata = SCIPheurGetData(heur);
717  assert(heurdata != NULL);
718 
719  /* only call the heuristic if we have an incumbent */
720  if( SCIPgetNSolsFound(scip) == 0 )
721  return SCIP_OKAY;
722 
723  /* do not use heuristic on problems without binary variables */
724  if( SCIPgetNBinVars(scip) == 0 )
725  return SCIP_OKAY;
726 
727  incumbent = SCIPgetBestSol(scip);
728  assert(incumbent != NULL);
729 
730  /* make sure that the incumbent is valid for the transformed space, otherwise terminate */
731  if( SCIPsolIsOriginal(incumbent) )
732  return SCIP_OKAY;
733 
734  solidx = SCIPsolGetIndex(incumbent);
735 
736  if( heurdata->lastsolidx == solidx )
737  return SCIP_OKAY;
738 
739  /* only call heuristic, if the best solution does not come from trivial heuristic */
740  if( SCIPsolGetHeur(incumbent) != NULL && strcmp(SCIPheurGetName(SCIPsolGetHeur(incumbent)), "trivial") == 0 )
741  return SCIP_OKAY;
742 
743  /* waitingnodes parameter defines the minimum number of nodes to wait before a new incumbent is processed */
744  if( SCIPgetNNodes(scip) > 1 && SCIPgetNNodes(scip) - SCIPsolGetNodenum(incumbent) < heurdata->waitingnodes )
745  return SCIP_OKAY;
746 
747  bestobj = SCIPgetSolTransObj(scip, incumbent);
748  lowerbound = SCIPgetLowerbound(scip);
749 
750  /* use knowledge about integrality of objective to round up lower bound */
751  if( SCIPisObjIntegral(scip) )
752  {
753  SCIPdebugMsg(scip, " Rounding up lower bound: %f --> %f \n", lowerbound, SCIPfeasCeil(scip, lowerbound));
754  lowerbound = SCIPfeasCeil(scip, lowerbound);
755  }
756 
757  /* do not trigger heuristic if primal and dual bound are already close together */
758  if( SCIPisFeasLE(scip, bestobj, lowerbound) || SCIPgetGap(scip) <= heurdata->mingap )
759  return SCIP_OKAY;
760 
761  /* calculate the minimum improvement for a heuristic solution in terms of the distance between incumbent objective
762  * and the lower bound */
763  if( SCIPisInfinity(scip, REALABS(lowerbound)) )
764  {
765  if( SCIPisZero(scip, bestobj) )
766  objcutoff = bestobj - 1;
767  else
768  objcutoff = (1 - minimprove) * bestobj;
769  }
770  else
771  objcutoff = minimprove * lowerbound + (1 - minimprove) * (bestobj);
772 
773  /* use integrality of the objective function to round down (and thus strengthen) the objective cutoff */
774  if( SCIPisObjIntegral(scip) )
775  objcutoff = SCIPfeasFloor(scip, objcutoff);
776 
777  if( SCIPisFeasLT(scip, objcutoff, lowerbound) )
778  objcutoff = lowerbound;
779 
780  /* exit execution if the right hand side of the objective constraint does not change (suggests that the heuristic
781  * was not successful in a previous iteration) */
782  if( heurdata->objcons != NULL && SCIPisFeasEQ(scip, SCIPgetRhsLinear(heurdata->subscip, heurdata->objcons), objcutoff) )
783  return SCIP_OKAY;
784 
785  /* check whether there is enough time and memory left */
786  SCIP_CALL( SCIPcheckCopyLimits(scip, &valid) );
787 
788  if( ! valid )
789  return SCIP_OKAY;
790 
791  *result = SCIP_DIDNOTFIND;
792 
793  heurdata->lastsolidx = solidx;
794 
795  /* get variable data */
796  SCIP_CALL( SCIPgetVarsData(scip, &vars, &nvars, NULL, NULL, NULL, NULL) );
797 
798  /* create a subscip and copy the original scip instance into it */
799  if( heurdata->subscip == NULL )
800  {
801  assert(heurdata->varmapfw == NULL);
802  assert(heurdata->objcons == NULL);
803 
804  /* initialize the subproblem */
805  SCIP_CALL( SCIPcreate(&subscip) );
806 
807  /* create the variable mapping hash map */
808  SCIP_CALL( SCIPhashmapCreate(&varmapfw, SCIPblkmem(subscip), nvars) );
809  SCIP_CALL( SCIPallocBlockMemoryArray(scip, &subvars, nvars) );
810 
811  /* copy complete SCIP instance */
812  valid = FALSE;
813 
814  /* create a problem copy as sub SCIP */
815  SCIP_CALL( SCIPcopyLargeNeighborhoodSearch(scip, subscip, varmapfw, "proximity", NULL, NULL, 0, heurdata->uselprows, TRUE,
816  &success, &valid) );
817 
818  SCIPdebugMsg(scip, "Copying the SCIP instance was %s complete.\n", valid ? "" : "not ");
819 
820  /* create event handler for LP events */
821  eventhdlr = NULL;
822  SCIP_CALL( SCIPincludeEventhdlrBasic(subscip, &eventhdlr, EVENTHDLR_NAME, EVENTHDLR_DESC, eventExecProximity, NULL) );
823  if( eventhdlr == NULL )
824  {
825  SCIPerrorMessage("event handler for " HEUR_NAME " heuristic not found.\n");
826  return SCIP_PLUGINNOTFOUND;
827  }
828 
829  /* set up parameters for the copied instance */
830  SCIP_CALL( setupSubproblem(heurdata, subscip) );
831 
832  /* create the objective constraint in the sub scip, first without variables and values which will be added later */
833  SCIP_CALL( SCIPcreateConsBasicLinear(subscip, &objcons, "objbound_of_origscip", 0, NULL, NULL, -SCIPinfinity(subscip), SCIPinfinity(subscip)) );
834 
835  /* determine large value to set variable bounds to, safe-guard to avoid fixings to infinite values */
836  large = SCIPinfinity(scip);
837  if( !SCIPisInfinity(scip, 0.1 / SCIPfeastol(scip)) )
838  large = 0.1 / SCIPfeastol(scip);
839  inf = SCIPinfinity(subscip);
840 
841  /* get variable image and change objective to proximity function (Manhattan distance) in sub-SCIP */
842  for( i = 0; i < nvars; i++ )
843  {
844  SCIP_Real adjustedbound;
845  SCIP_Real lb;
846  SCIP_Real ub;
847 
848  subvars[i] = (SCIP_VAR*) SCIPhashmapGetImage(varmapfw, vars[i]);
849 
850  if( subvars[i] == NULL )
851  continue;
852 
853  SCIP_CALL( SCIPchgVarObj(subscip, subvars[i], 0.0) );
854 
855  lb = SCIPvarGetLbGlobal(subvars[i]);
856  ub = SCIPvarGetUbGlobal(subvars[i]);
857 
858  /* adjust infinite bounds in order to avoid that variables with non-zero objective
859  * get fixed to infinite value in proximity subproblem
860  */
861  if( SCIPisInfinity(subscip, ub) )
862  {
863  adjustedbound = MAX(large, lb + large);
864  adjustedbound = MIN(adjustedbound, inf);
865  SCIP_CALL( SCIPchgVarUbGlobal(subscip, subvars[i], adjustedbound) );
866  }
867  if( SCIPisInfinity(subscip, -lb) )
868  {
869  adjustedbound = MIN(-large, ub - large);
870  adjustedbound = MAX(adjustedbound, -inf);
871  SCIP_CALL( SCIPchgVarLbGlobal(subscip, subvars[i], adjustedbound) );
872  }
873 
874  /* add all nonzero objective coefficients to the objective constraint */
875  if( !SCIPisFeasZero(subscip, SCIPvarGetObj(vars[i])) )
876  {
877  SCIP_CALL( SCIPaddCoefLinear(subscip, objcons, subvars[i], SCIPvarGetObj(vars[i])) );
878  }
879  }
880 
881  /* add objective constraint to the subscip */
882  SCIP_CALL( SCIPaddCons(subscip, objcons) );
883  }
884  else
885  {
886  /* the instance, event handler, hash map and variable array were already copied in a previous iteration
887  * and stored in heuristic data
888  */
889  assert(heurdata->varmapfw != NULL);
890  assert(heurdata->subvars != NULL);
891  assert(heurdata->objcons != NULL);
892 
893  subscip = heurdata->subscip;
894  varmapfw = heurdata->varmapfw;
895  subvars = heurdata->subvars;
896  objcons = heurdata->objcons;
897 
898  eventhdlr = SCIPfindEventhdlr(subscip, EVENTHDLR_NAME);
899  assert(eventhdlr != NULL);
900  }
901 
902  SCIP_CALL( SCIPchgRhsLinear(subscip, objcons, objcutoff) );
903 
904  for( i = 0; i < SCIPgetNBinVars(scip); ++i )
905  {
906  SCIP_Real solval;
907 
908  if( subvars[i] == NULL )
909  continue;
910 
911  /* objective coefficients are only set for binary variables of the problem */
912  assert(SCIPvarIsBinary(subvars[i]));
913 
914  solval = SCIPgetSolVal(scip, incumbent, vars[i]);
915  assert(SCIPisFeasGE(scip, solval, 0.0));
916  assert(SCIPisFeasLE(scip, solval, 1.0));
917  assert(SCIPisFeasIntegral(scip, solval));
918 
919  if( solval < 0.5 )
920  {
921  SCIP_CALL( SCIPchgVarObj(subscip, subvars[i], 1.0) );
922  }
923  else
924  {
925  SCIP_CALL( SCIPchgVarObj(subscip, subvars[i], -1.0) );
926  }
927  }
928 
929  /* set limits for the subproblem */
930  SCIP_CALL( SCIPcopyLimits(scip, subscip) );
931  SCIP_CALL( SCIPsetLongintParam(subscip, "limits/nodes", nnodes) );
932  SCIP_CALL( SCIPsetIntParam(subscip, "limits/solutions", 1) );
933 
934  /* restrict LP iterations */
935  /* todo set iterations limit depending on the number of iterations of the original problem root */
936  iterlim = nlpiters;
937  SCIP_CALL( SCIPsetLongintParam(subscip, "lp/iterlim", MAX(1, iterlim / MIN(10, nnodes))) );
938  SCIP_CALL( SCIPsetLongintParam(subscip, "lp/rootiterlim", iterlim) );
939 
940  /* catch LP events of sub-SCIP */
941  SCIP_CALL( SCIPtransformProb(subscip) );
942  SCIP_CALL( SCIPcatchEvent(subscip, SCIP_EVENTTYPE_NODESOLVED, eventhdlr, (SCIP_EVENTDATA*) heurdata, NULL) );
943 
944  SCIPstatisticMessage("solving subproblem at Node: %" SCIP_LONGINT_FORMAT " "
945  "nnodes: %" SCIP_LONGINT_FORMAT " "
946  "iterlim: %" SCIP_LONGINT_FORMAT "\n", SCIPgetNNodes(scip), nnodes, iterlim);
947 
948  /* solve the subproblem with all previously adjusted parameters */
949  nfixedvars = SCIPgetNFixedVars(subscip);
950 
951  SCIP_CALL( SCIPpresolve(subscip) );
952 
953  nfixedvars = SCIPgetNFixedVars(subscip) - nfixedvars;
954  assert(nfixedvars >= 0);
955  SCIPstatisticMessage("presolve fixings %d: %d\n", ++(heurdata->subprobidx), nfixedvars);
956 
957  /* errors in solving the subproblem should not kill the overall solving process;
958  * hence, the return code is caught and a warning is printed, only in debug mode, SCIP will stop.
959  */
960  SCIP_CALL_ABORT( SCIPsolve(subscip) );
961 
962  /* print solving statistics of subproblem if we are in SCIP's debug mode */
964  SCIPstatisticMessage("solve of subscip %d:"
965  "usednodes: %" SCIP_LONGINT_FORMAT " "
966  "lp iters: %" SCIP_LONGINT_FORMAT " "
967  "root iters: %" SCIP_LONGINT_FORMAT " "
968  "Presolving Time: %.2f\n", heurdata->subprobidx,
970 
971  SCIPstatisticMessage("Solving Time %d: %.2f\n", heurdata->subprobidx, SCIPgetSolvingTime(subscip) );
972 
973  /* drop LP events of sub-SCIP */
974  SCIP_CALL( SCIPdropEvent(subscip, SCIP_EVENTTYPE_NODESOLVED, eventhdlr, (SCIP_EVENTDATA*) heurdata, -1) );
975 
976  /* keep track of relevant information for future runs of heuristic */
977  if( nusednodes != NULL )
978  *nusednodes = SCIPgetNNodes(subscip);
979  if( nusedlpiters != NULL )
980  *nusedlpiters = SCIPgetNLPIterations(subscip);
981 
982  /* check whether a solution was found */
983  nsubsols = SCIPgetNSols(subscip);
984  incumbent = SCIPgetBestSol(subscip);
985  assert(nsubsols == 0 || incumbent != NULL);
986 
987  SCIPstatisticMessage("primal bound before subproblem %d: %g\n", heurdata->subprobidx, SCIPgetPrimalbound(scip));
988  if( nsubsols > 0 )
989  {
990  /* try to translate the sub problem solution to the original scip instance */
991  success = FALSE;
992  SCIP_CALL( createNewSol(scip, subscip, subvars, heur, incumbent, heurdata->usefinallp, &success) );
993 
994  if( success )
995  *result = SCIP_FOUNDSOL;
996  }
997  SCIPstatisticMessage("primal bound after subproblem %d: %g\n", heurdata->subprobidx, SCIPgetPrimalbound(scip));
998 
999  /* free the transformed subproblem data */
1000  SCIP_CALL( SCIPfreeTransform(subscip) );
1001 
1002  /* save subproblem in heuristic data for subsequent runs if it has been successful, otherwise free subproblem */
1003  heurdata->subscip = subscip;
1004  heurdata->varmapfw = varmapfw;
1005  heurdata->subvars = subvars;
1006  heurdata->objcons = objcons;
1007  heurdata->nsubvars = nvars;
1008 
1009  /* delete the sub problem */
1010  if( freesubscip )
1011  {
1012  SCIP_CALL( deleteSubproblem(scip, heurdata) );
1013  }
1014 
1015  return SCIP_OKAY;
1016 }
1017 
1018 
1019 /** creates the proximity primal heuristic and includes it in SCIP */
1021  SCIP* scip /**< SCIP data structure */
1022  )
1023 {
1024  SCIP_HEURDATA* heurdata;
1025  SCIP_HEUR* heur = NULL;
1026 
1027  /* create heuristic data */
1028  SCIP_CALL( SCIPallocBlockMemory(scip, &heurdata) );
1029 
1030  /* include primal heuristic */
1031  SCIP_CALL( SCIPincludeHeurBasic(scip, &heur,
1033  HEUR_MAXDEPTH, HEUR_TIMING, HEUR_USESSUBSCIP, heurExecProximity, heurdata) );
1034  assert(heur != NULL);
1035 
1036  /* set non-NULL pointers to callback methods */
1037  SCIP_CALL( SCIPsetHeurCopy(scip, heur, heurCopyProximity) );
1038  SCIP_CALL( SCIPsetHeurFree(scip, heur, heurFreeProximity) );
1039  SCIP_CALL( SCIPsetHeurInit(scip, heur, heurInitProximity) );
1040  SCIP_CALL( SCIPsetHeurExitsol(scip, heur, heurExitsolProximity) );
1041 
1042  /* add proximity primal heuristic parameters */
1043  SCIP_CALL( SCIPaddBoolParam(scip, "heuristics/" HEUR_NAME "/uselprows",
1044  "should subproblem be constructed based on LP row information?",
1045  &heurdata->uselprows, TRUE, DEFAULT_USELPROWS, NULL, NULL) );
1046 
1047  SCIP_CALL( SCIPaddBoolParam(scip, "heuristics/" HEUR_NAME "/restart",
1048  "should the heuristic immediately run again on its newly found solution?",
1049  &heurdata->restart, TRUE, DEFAULT_RESTART, NULL, NULL) );
1050 
1051  SCIP_CALL( SCIPaddBoolParam(scip, "heuristics/" HEUR_NAME "/usefinallp",
1052  "should the heuristic solve a final LP in case of continuous objective variables?",
1053  &heurdata->usefinallp, TRUE, DEFAULT_USEFINALLP, NULL, NULL) );
1054 
1055  SCIP_CALL( SCIPaddLongintParam(scip, "heuristics/" HEUR_NAME "/maxnodes",
1056  "maximum number of nodes to regard in the subproblem",
1057  &heurdata->maxnodes, TRUE,DEFAULT_MAXNODES, 0LL, SCIP_LONGINT_MAX, NULL, NULL) );
1058 
1059  SCIP_CALL( SCIPaddLongintParam(scip, "heuristics/" HEUR_NAME "/nodesofs",
1060  "number of nodes added to the contingent of the total nodes",
1061  &heurdata->nodesofs, TRUE, DEFAULT_NODESOFS, 0LL, SCIP_LONGINT_MAX, NULL, NULL) );
1062 
1063  SCIP_CALL( SCIPaddLongintParam(scip, "heuristics/" HEUR_NAME "/minnodes",
1064  "minimum number of nodes required to start the subproblem",
1065  &heurdata->minnodes, TRUE, DEFAULT_MINNODES, 0LL, SCIP_LONGINT_MAX, NULL, NULL) );
1066 
1067  SCIP_CALL( SCIPaddLongintParam(scip, "heuristics/" HEUR_NAME "/maxlpiters",
1068  "maximum number of LP iterations to be performed in the subproblem",
1069  &heurdata->maxlpiters, TRUE, DEFAULT_MAXLPITERS, -1LL, SCIP_LONGINT_MAX, NULL, NULL) );
1070 
1071  SCIP_CALL( SCIPaddLongintParam(scip, "heuristics/" HEUR_NAME "/minlpiters",
1072  "minimum number of LP iterations performed in subproblem",
1073  &heurdata->minlpiters, TRUE, DEFAULT_MINLPITERS, 0LL, SCIP_LONGINT_MAX, NULL, NULL) );
1074 
1075  SCIP_CALL( SCIPaddLongintParam(scip, "heuristics/" HEUR_NAME "/waitingnodes",
1076  "waiting nodes since last incumbent before heuristic is executed",
1077  &heurdata->waitingnodes, TRUE, DEFAULT_WAITINGNODES, 0LL, SCIP_LONGINT_MAX, NULL, NULL) );
1078 
1079  SCIP_CALL( SCIPaddRealParam(scip, "heuristics/" HEUR_NAME "/minimprove",
1080  "factor by which proximity should at least improve the incumbent",
1081  &heurdata->minimprove, TRUE, DEFAULT_MINIMPROVE, 0.0, 1.0, NULL, NULL) );
1082 
1083  SCIP_CALL( SCIPaddRealParam(scip, "heuristics/" HEUR_NAME "/nodesquot",
1084  "sub-MIP node limit w.r.t number of original nodes",
1085  &heurdata->nodesquot, TRUE, DEFAULT_NODESQUOT, 0.0, SCIPinfinity(scip), NULL, NULL) );
1086 
1087  SCIP_CALL( SCIPaddRealParam(scip, "heuristics/" HEUR_NAME "/binvarquot",
1088  "threshold for percentage of binary variables required to start",
1089  &heurdata->binvarquot, TRUE, DEFAULT_BINVARQUOT, 0.0, 1.0, NULL, NULL) );
1090 
1091  SCIP_CALL( SCIPaddRealParam(scip, "heuristics/" HEUR_NAME "/lpitersquot",
1092  "quotient of sub-MIP LP iterations with respect to LP iterations so far",
1093  &heurdata->lpitersquot, TRUE, DEFAULT_LPITERSQUOT, 0.0, 1.0, NULL, NULL) );
1094 
1095  SCIP_CALL( SCIPaddRealParam(scip, "heuristics/" HEUR_NAME "/mingap",
1096  "minimum primal-dual gap for which the heuristic is executed",
1097  &heurdata->mingap, TRUE, DEFAULT_MINGAP, 0.0, SCIPinfinity(scip), NULL, NULL) );
1098 
1099  SCIP_CALL( SCIPaddBoolParam(scip, "heuristics/" HEUR_NAME "/useuct",
1100  "should uct node selection be used at the beginning of the search?",
1101  &heurdata->useuct, TRUE, DEFAULT_USEUCT, NULL, NULL) );
1102 
1103  return SCIP_OKAY;
1104 }
enum SCIP_Result SCIP_RESULT
Definition: type_result.h:61
SCIP_Bool SCIPsolIsOriginal(SCIP_SOL *sol)
Definition: sol.c:2530
SCIP_RETCODE SCIPsetHeurExitsol(SCIP *scip, SCIP_HEUR *heur, SCIP_DECL_HEUREXITSOL((*heurexitsol)))
Definition: scip_heur.c:242
#define SCIPfreeBlockMemoryArray(scip, ptr, num)
Definition: scip_mem.h:110
SCIP_Bool SCIPisFeasZero(SCIP *scip, SCIP_Real val)
#define DEFAULT_USELPROWS
SCIP_RETCODE SCIPchgVarLbGlobal(SCIP *scip, SCIP_VAR *var, SCIP_Real newbound)
Definition: scip_var.c:4943
SCIP_RETCODE SCIPlinkLPSol(SCIP *scip, SCIP_SOL *sol)
Definition: scip_sol.c:1026
SCIP_Real SCIPgetSolvingTime(SCIP *scip)
Definition: scip_timing.c:378
SCIP_Longint SCIPgetNRootLPIterations(SCIP *scip)
SCIP_RETCODE SCIPsetSeparating(SCIP *scip, SCIP_PARAMSETTING paramsetting, SCIP_Bool quiet)
Definition: scip_param.c:958
SCIP_Real SCIPfeastol(SCIP *scip)
#define SCIPallocBlockMemoryArray(scip, ptr, num)
Definition: scip_mem.h:93
SCIP_Bool SCIPisNLPConstructed(SCIP *scip)
Definition: scip_nlp.c:110
#define DEFAULT_MINIMPROVE
SCIP_Bool SCIPisFeasEQ(SCIP *scip, SCIP_Real val1, SCIP_Real val2)
public methods for SCIP parameter handling
SCIP_Longint SCIPgetNLPIterations(SCIP *scip)
SCIP_RETCODE SCIPcreateConsBasicLinear(SCIP *scip, SCIP_CONS **cons, const char *name, int nvars, SCIP_VAR **vars, SCIP_Real *vals, SCIP_Real lhs, SCIP_Real rhs)
SCIP_Bool SCIPisFeasLT(SCIP *scip, SCIP_Real val1, SCIP_Real val2)
public methods for node selector plugins
public methods for memory management
SCIP_RETCODE SCIPincludeHeurProximity(SCIP *scip)
SCIP_Real SCIPgetPrimalbound(SCIP *scip)
SCIP_Real SCIPvarGetLbGlobal(SCIP_VAR *var)
Definition: var.c:17923
static SCIP_DECL_EVENTEXEC(eventExecProximity)
#define EVENTHDLR_DESC
static SCIP_DECL_HEURCOPY(heurCopyProximity)
SCIP_Longint SCIPgetNSolsFound(SCIP *scip)
static SCIP_RETCODE solveLp(SCIP *scip, SCIP_SOL *sol, SCIP_Bool *success)
SCIP_Real SCIPvarGetLbLocal(SCIP_VAR *var)
Definition: var.c:17979
public solving methods
SCIP_RETCODE SCIPincludeEventhdlrBasic(SCIP *scip, SCIP_EVENTHDLR **eventhdlrptr, const char *name, const char *desc, SCIP_DECL_EVENTEXEC((*eventexec)), SCIP_EVENTHDLRDATA *eventhdlrdata)
Definition: scip_event.c:104
public methods for timing
SCIP_Bool SCIPvarIsBinary(SCIP_VAR *var)
Definition: var.c:17444
static SCIP_DECL_HEURINIT(heurInitProximity)
#define DEFAULT_BINVARQUOT
SCIP_Bool SCIPisFeasGE(SCIP *scip, SCIP_Real val1, SCIP_Real val2)
SCIP_RETCODE SCIPgetVarsData(SCIP *scip, SCIP_VAR ***vars, int *nvars, int *nbinvars, int *nintvars, int *nimplvars, int *ncontvars)
Definition: scip_prob.c:1874
static SCIP_RETCODE deleteSubproblem(SCIP *scip, SCIP_HEURDATA *heurdata)
#define FALSE
Definition: def.h:96
SCIP_RETCODE SCIPhashmapCreate(SCIP_HASHMAP **hashmap, BMS_BLKMEM *blkmem, int mapsize)
Definition: misc.c:3024
const char * SCIPeventhdlrGetName(SCIP_EVENTHDLR *eventhdlr)
Definition: event.c:324
#define HEUR_MAXDEPTH
SCIP_RETCODE SCIPaddLongintParam(SCIP *scip, const char *name, const char *desc, SCIP_Longint *valueptr, SCIP_Bool isadvanced, SCIP_Longint defaultvalue, SCIP_Longint minvalue, SCIP_Longint maxvalue, SCIP_DECL_PARAMCHGD((*paramchgd)), SCIP_PARAMDATA *paramdata)
Definition: scip_param.c:111
SCIP_RETCODE SCIPcopyLimits(SCIP *sourcescip, SCIP *targetscip)
Definition: scip_copy.c:3287
#define DEFAULT_USEFINALLP
SCIP_Real SCIPinfinity(SCIP *scip)
#define TRUE
Definition: def.h:95
#define SCIPdebug(x)
Definition: pub_message.h:93
enum SCIP_Retcode SCIP_RETCODE
Definition: type_retcode.h:63
#define SCIPstatisticMessage
Definition: pub_message.h:123
methods commonly used by primal heuristics
SCIP_RETCODE SCIPsetPresolving(SCIP *scip, SCIP_PARAMSETTING paramsetting, SCIP_Bool quiet)
Definition: scip_param.c:932
SCIP_BRANCHRULE * SCIPfindBranchrule(SCIP *scip, const char *name)
Definition: scip_branch.c:297
SCIP_RETCODE SCIPchgVarLbDive(SCIP *scip, SCIP_VAR *var, SCIP_Real newbound)
Definition: scip_lp.c:2416
SCIP_Longint SCIPgetNRootFirstLPIterations(SCIP *scip)
struct SCIP_HeurData SCIP_HEURDATA
Definition: type_heur.h:76
public methods for problem variables
#define SCIPfreeBlockMemory(scip, ptr)
Definition: scip_mem.h:108
SCIP_RETCODE SCIPincludeHeurBasic(SCIP *scip, SCIP_HEUR **heur, const char *name, const char *desc, char dispchar, int priority, int freq, int freqofs, int maxdepth, SCIP_HEURTIMING timingmask, SCIP_Bool usessubscip, SCIP_DECL_HEUREXEC((*heurexec)), SCIP_HEURDATA *heurdata)
Definition: scip_heur.c:117
SCIP_EVENTHDLR * SCIPfindEventhdlr(SCIP *scip, const char *name)
Definition: scip_event.c:234
#define HEUR_FREQ
void * SCIPhashmapGetImage(SCIP_HASHMAP *hashmap, void *origin)
Definition: misc.c:3211
#define SCIP_LONGINT_MAX
Definition: def.h:172
#define SCIPfreeBufferArray(scip, ptr)
Definition: scip_mem.h:136
SCIP_RETCODE SCIPcreate(SCIP **scip)
Definition: scip_general.c:292
void SCIPheurSetData(SCIP_HEUR *heur, SCIP_HEURDATA *heurdata)
Definition: heur.c:1371
#define SCIPallocBlockMemory(scip, ptr)
Definition: scip_mem.h:89
public methods for SCIP variables
void SCIPwarningMessage(SCIP *scip, const char *formatstr,...)
Definition: scip_message.c:120
#define HEUR_PRIORITY
SCIP_RETCODE SCIPchgVarUbGlobal(SCIP *scip, SCIP_VAR *var, SCIP_Real newbound)
Definition: scip_var.c:5032
#define SCIPdebugMsg
Definition: scip_message.h:78
SCIP_Real SCIPgetRhsLinear(SCIP *scip, SCIP_CONS *cons)
SCIP_Real SCIPgetPresolvingTime(SCIP *scip)
Definition: scip_timing.c:442
SCIP_RETCODE SCIPprintStatistics(SCIP *scip, FILE *file)
SCIP_RETCODE SCIPaddCoefLinear(SCIP *scip, SCIP_CONS *cons, SCIP_VAR *var, SCIP_Real val)
SCIP_Real SCIPgetObjNorm(SCIP *scip)
Definition: scip_prob.c:1649
SCIP_Real SCIPfeasCeil(SCIP *scip, SCIP_Real val)
public methods for numerical tolerances
SCIP_Real SCIPfeasFloor(SCIP *scip, SCIP_Real val)
public methods for querying solving statistics
int SCIPgetNFixedVars(SCIP *scip)
Definition: scip_prob.c:2317
SCIP_Bool SCIPisLPConstructed(SCIP *scip)
Definition: scip_lp.c:101
SCIP_Real SCIPvarGetUbGlobal(SCIP_VAR *var)
Definition: var.c:17933
#define HEUR_NAME
#define DEFAULT_MINLPITERS
SCIP_RETCODE SCIPsolve(SCIP *scip)
Definition: scip_solve.c:2624
const char * SCIPheurGetName(SCIP_HEUR *heur)
Definition: heur.c:1450
SCIP_HEUR * SCIPfindHeur(SCIP *scip, const char *name)
Definition: scip_heur.c:258
#define DEFAULT_LPITERSQUOT
#define SCIPerrorMessage
Definition: pub_message.h:64
SCIP_Bool SCIPisParamFixed(SCIP *scip, const char *name)
Definition: scip_param.c:219
SCIP_RETCODE SCIPdeleteSubproblemProximity(SCIP *scip)
SCIP_RETCODE SCIPaddCons(SCIP *scip, SCIP_CONS *cons)
Definition: scip_prob.c:2778
#define HEUR_FREQOFS
SCIP_RETCODE SCIPsolveDiveLP(SCIP *scip, int itlim, SCIP_Bool *lperror, SCIP_Bool *cutoff)
Definition: scip_lp.c:2675
SCIP_RETCODE SCIPsetHeurFree(SCIP *scip, SCIP_HEUR *heur, SCIP_DECL_HEURFREE((*heurfree)))
Definition: scip_heur.c:178
#define HEUR_TIMING
public methods for event handler plugins and event handlers
SCIP_RETCODE SCIPsetBoolParam(SCIP *scip, const char *name, SCIP_Bool value)
Definition: scip_param.c:429
SCIP_RETCODE SCIPpresolve(SCIP *scip)
Definition: scip_solve.c:2454
BMS_BLKMEM * SCIPblkmem(SCIP *scip)
Definition: scip_mem.c:57
static SCIP_RETCODE createNewSol(SCIP *scip, SCIP *subscip, SCIP_VAR **subvars, SCIP_HEUR *heur, SCIP_SOL *subsol, SCIP_Bool usefinallp, SCIP_Bool *success)
struct SCIP_EventData SCIP_EVENTDATA
Definition: type_event.h:173
void SCIPhashmapFree(SCIP_HASHMAP **hashmap)
Definition: misc.c:3058
#define NULL
Definition: lpi_spx1.cpp:164
SCIP_HEUR * SCIPsolGetHeur(SCIP_SOL *sol)
Definition: sol.c:2613
SCIP_Real SCIPgetSolTransObj(SCIP *scip, SCIP_SOL *sol)
Definition: scip_sol.c:1491
#define DEFAULT_MAXNODES
#define REALABS(x)
Definition: def.h:210
public methods for problem copies
public methods for primal CIP solutions
#define EVENTHDLR_NAME
#define SCIP_CALL(x)
Definition: def.h:394
SCIP_Real SCIPgetLowerbound(SCIP *scip)
static SCIP_DECL_HEUREXEC(heurExecProximity)
#define DEFAULT_NODESOFS
SCIP_Bool SCIPisFeasLE(SCIP *scip, SCIP_Real val1, SCIP_Real val2)
#define DEFAULT_NODESQUOT
public methods for primal heuristic plugins and divesets
public methods for constraint handler plugins and constraints
#define DEFAULT_USEUCT
SCIP_RETCODE SCIPchgVarObj(SCIP *scip, SCIP_VAR *var, SCIP_Real newobj)
Definition: scip_var.c:4513
#define SCIPallocBufferArray(scip, ptr, num)
Definition: scip_mem.h:124
SCIP_RETCODE SCIPsetSolVal(SCIP *scip, SCIP_SOL *sol, SCIP_VAR *var, SCIP_Real val)
Definition: scip_sol.c:1221
#define DEFAULT_MAXLPITERS
public data structures and miscellaneous methods
#define DEFAULT_MINGAP
SCIP_RETCODE SCIPfreeTransform(SCIP *scip)
Definition: scip_solve.c:3469
#define SCIP_Bool
Definition: def.h:93
SCIP_RETCODE SCIPcatchEvent(SCIP *scip, SCIP_EVENTTYPE eventtype, SCIP_EVENTHDLR *eventhdlr, SCIP_EVENTDATA *eventdata, int *filterpos)
Definition: scip_event.c:286
SCIP_LPSOLSTAT SCIPgetLPSolstat(SCIP *scip)
Definition: scip_lp.c:168
SCIP_EVENTTYPE SCIPeventGetType(SCIP_EVENT *event)
Definition: event.c:1030
SCIP_Longint SCIPsolGetNodenum(SCIP_SOL *sol)
Definition: sol.c:2593
SCIP_Real SCIPgetGap(SCIP *scip)
SCIP_RETCODE SCIPchgVarUbDive(SCIP *scip, SCIP_VAR *var, SCIP_Real newbound)
Definition: scip_lp.c:2448
#define MAX(x, y)
Definition: tclique_def.h:92
SCIP_RETCODE SCIPsetIntParam(SCIP *scip, const char *name, int value)
Definition: scip_param.c:487
SCIP_RETCODE SCIPdropEvent(SCIP *scip, SCIP_EVENTTYPE eventtype, SCIP_EVENTHDLR *eventhdlr, SCIP_EVENTDATA *eventdata, int filterpos)
Definition: scip_event.c:320
SCIP_RETCODE SCIPfreeSol(SCIP *scip, SCIP_SOL **sol)
Definition: scip_sol.c:985
SCIP_Real SCIPvarGetObj(SCIP_VAR *var)
Definition: var.c:17771
int SCIPgetNSols(SCIP *scip)
Definition: scip_sol.c:2214
#define DEFAULT_MINNODES
SCIP_Real SCIPgetSolOrigObj(SCIP *scip, SCIP_SOL *sol)
Definition: scip_sol.c:1444
Constraint handler for linear constraints in their most general form, .
int SCIPgetNObjVars(SCIP *scip)
Definition: scip_prob.c:2228
SCIP_RETCODE SCIPchgRhsLinear(SCIP *scip, SCIP_CONS *cons, SCIP_Real rhs)
SCIP_Bool SCIPisInfinity(SCIP *scip, SCIP_Real val)
SCIP_RETCODE SCIPtrySol(SCIP *scip, SCIP_SOL *sol, SCIP_Bool printreason, SCIP_Bool completely, SCIP_Bool checkbounds, SCIP_Bool checkintegrality, SCIP_Bool checklprows, SCIP_Bool *stored)
Definition: scip_sol.c:3135
int SCIPgetNBinVars(SCIP *scip)
Definition: scip_prob.c:2045
public methods for the LP relaxation, rows and columns
#define SCIP_EVENTTYPE_NODESOLVED
Definition: type_event.h:136
int SCIPgetNVars(SCIP *scip)
Definition: scip_prob.c:2000
public methods for nonlinear relaxation
public methods for branching rule plugins and branching
SCIP_Bool SCIPisObjIntegral(SCIP *scip)
Definition: scip_prob.c:1570
public methods for managing events
general public methods
#define HEUR_DESC
SCIP_SOL * SCIPgetBestSol(SCIP *scip)
Definition: scip_sol.c:2313
#define DEFAULT_WAITINGNODES
public methods for solutions
static SCIP_DECL_HEURFREE(heurFreeProximity)
SCIP_RETCODE SCIPreleaseCons(SCIP *scip, SCIP_CONS **cons)
Definition: scip_cons.c:1119
public methods for message output
SCIP_RETCODE SCIPsetHeurInit(SCIP *scip, SCIP_HEUR *heur, SCIP_DECL_HEURINIT((*heurinit)))
Definition: scip_heur.c:194
SCIP_NODESEL * SCIPfindNodesel(SCIP *scip, const char *name)
Definition: scip_nodesel.c:234
SCIP_RETCODE SCIPcopyLargeNeighborhoodSearch(SCIP *sourcescip, SCIP *subscip, SCIP_HASHMAP *varmap, const char *suffix, SCIP_VAR **fixedvars, SCIP_Real *fixedvals, int nfixedvars, SCIP_Bool uselprows, SCIP_Bool copycuts, SCIP_Bool *success, SCIP_Bool *valid)
Definition: heuristics.c:925
SCIP_VARSTATUS SCIPvarGetStatus(SCIP_VAR *var)
Definition: var.c:17383
#define SCIP_Real
Definition: def.h:186
SCIP_Bool SCIPisStopped(SCIP *scip)
Definition: scip_general.c:703
public methods for message handling
static SCIP_DECL_HEUREXITSOL(heurExitsolProximity)
#define SCIP_Longint
Definition: def.h:171
SCIP_RETCODE SCIPcheckCopyLimits(SCIP *sourcescip, SCIP_Bool *success)
Definition: scip_copy.c:3244
SCIP_RETCODE SCIPstartDive(SCIP *scip)
Definition: scip_lp.c:2239
SCIP_RETCODE SCIPsetSolVals(SCIP *scip, SCIP_SOL *sol, int nvars, SCIP_VAR **vars, SCIP_Real *vals)
Definition: scip_sol.c:1263
SCIP_RETCODE SCIPtransformProb(SCIP *scip)
Definition: scip_solve.c:367
SCIP_Bool SCIPisZero(SCIP *scip, SCIP_Real val)
SCIP_RETCODE SCIPsetHeurCopy(SCIP *scip, SCIP_HEUR *heur, SCIP_DECL_HEURCOPY((*heurcopy)))
Definition: scip_heur.c:162
#define nnodes
Definition: gastrans.c:74
SCIP_Real SCIPvarGetUbLocal(SCIP_VAR *var)
Definition: var.c:17989
SCIP_Bool SCIPisFeasIntegral(SCIP *scip, SCIP_Real val)
#define HEUR_DISPCHAR
SCIP_RETCODE SCIPinterruptSolve(SCIP *scip)
Definition: scip_solve.c:3553
public methods for primal heuristics
SCIP_RETCODE SCIPapplyProximity(SCIP *scip, SCIP_HEUR *heur, SCIP_RESULT *result, SCIP_Real minimprove, SCIP_Longint nnodes, SCIP_Longint nlpiters, SCIP_Longint *nusednodes, SCIP_Longint *nusedlpiters, SCIP_Bool freesubscip)
SCIP_RETCODE SCIPendDive(SCIP *scip)
Definition: scip_lp.c:2288
#define SCIP_CALL_ABORT(x)
Definition: def.h:373
SCIP_HEURDATA * SCIPheurGetData(SCIP_HEUR *heur)
Definition: heur.c:1361
SCIP_Longint SCIPgetNNodes(SCIP *scip)
#define HEUR_USESSUBSCIP
public methods for global and local (sub)problems
SCIP_Bool SCIPvarIsIntegral(SCIP_VAR *var)
Definition: var.c:17455
SCIP_Real SCIPgetSolVal(SCIP *scip, SCIP_SOL *sol, SCIP_VAR *var)
Definition: scip_sol.c:1361
#define DEFAULT_RESTART
int SCIPsolGetIndex(SCIP_SOL *sol)
Definition: sol.c:2644
improvement heuristic which uses an auxiliary objective instead of the original objective function wh...
SCIP_RETCODE SCIPaddRealParam(SCIP *scip, const char *name, const char *desc, SCIP_Real *valueptr, SCIP_Bool isadvanced, SCIP_Real defaultvalue, SCIP_Real minvalue, SCIP_Real maxvalue, SCIP_DECL_PARAMCHGD((*paramchgd)), SCIP_PARAMDATA *paramdata)
Definition: scip_param.c:139
SCIP_RETCODE SCIPsetSubscipsOff(SCIP *scip, SCIP_Bool quiet)
Definition: scip_param.c:883
SCIP_RETCODE SCIPsetLongintParam(SCIP *scip, const char *name, SCIP_Longint value)
Definition: scip_param.c:545
SCIP_RETCODE SCIPaddBoolParam(SCIP *scip, const char *name, const char *desc, SCIP_Bool *valueptr, SCIP_Bool isadvanced, SCIP_Bool defaultvalue, SCIP_DECL_PARAMCHGD((*paramchgd)), SCIP_PARAMDATA *paramdata)
Definition: scip_param.c:57
static SCIP_RETCODE setupSubproblem(SCIP_HEURDATA *heurdata, SCIP *subscip)
SCIP_RETCODE SCIPfree(SCIP **scip)
Definition: scip_general.c:324
SCIP_RETCODE SCIPcreateSol(SCIP *scip, SCIP_SOL **sol, SCIP_HEUR *heur)
Definition: scip_sol.c:328
memory allocation routines