This page list all parameters of the current SCIP version. This list can easily be generated by SCIP via the interactive shell using the following command:
SCIP> set save <file name>
or via the function call:
SCIP_CALL( SCIPwriteParams(scip, <file name>, TRUE, FALSE) );
# SCIP version 9.1.0 # branching score function ('s'um, 'p'roduct, 'q'uotient) # [type: char, advanced: TRUE, range: {spq}, default: p] branching/scorefunc = p # branching score factor to weigh downward and upward gain prediction in sum score function # [type: real, advanced: TRUE, range: [0,1], default: 0.167] branching/scorefac = 0.167 # should branching on binary variables be preferred? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] branching/preferbinary = FALSE # minimal relative distance of branching point to bounds when branching on a continuous variable # [type: real, advanced: FALSE, range: [0,0.5], default: 0.2] branching/clamp = 0.2 # fraction by which to move branching point of a continuous variable towards the middle of the domain; a value of 1.0 leads to branching always in the middle of the domain # [type: real, advanced: FALSE, range: [0,1], default: 0.75] branching/midpull = 0.75 # multiply midpull by relative domain width if the latter is below this value # [type: real, advanced: FALSE, range: [0,1], default: 0.5] branching/midpullreldomtrig = 0.5 # strategy for normalization of LP gain when updating pseudocosts of continuous variables (divide by movement of 'l'p value, reduction in 'd'omain width, or reduction in domain width of 's'ibling) # [type: char, advanced: FALSE, range: {dls}, default: s] branching/lpgainnormalize = s # should updating pseudo costs for continuous variables be delayed to the time after separation? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] branching/delaypscostupdate = TRUE # should pseudo costs be updated also in diving and probing mode? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] branching/divingpscost = TRUE # should all strong branching children be regarded even if one is detected to be infeasible? (only with propagation) # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/forceallchildren = FALSE # child node to be regarded first during strong branching (only with propagation): 'u'p child, 'd'own child, 'h'istory-based, or 'a'utomatic # [type: char, advanced: TRUE, range: {aduh}, default: a] branching/firstsbchild = a # should LP solutions during strong branching with propagation be checked for feasibility? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] branching/checksol = TRUE # should LP solutions during strong branching with propagation be rounded? (only when checksbsol=TRUE) # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] branching/roundsbsol = TRUE # score adjustment near zero by adding epsilon (TRUE) or using maximum (FALSE) # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/sumadjustscore = FALSE # should automatic tree compression after the presolving be enabled? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] compression/enable = FALSE # should conflict analysis be enabled? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] conflict/enable = TRUE # should conflicts based on an old cutoff bound be removed from the conflict pool after improving the primal bound? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] conflict/cleanboundexceedings = TRUE # use local rows to construct infeasibility proofs # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] conflict/uselocalrows = TRUE # should propagation conflict analysis be used? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] conflict/useprop = TRUE # should infeasible LP conflict analysis be used? ('o'ff, 'c'onflict graph, 'd'ual ray, 'b'oth conflict graph and dual ray) # [type: char, advanced: FALSE, range: {ocdb}, default: b] conflict/useinflp = b # should bound exceeding LP conflict analysis be used? ('o'ff, 'c'onflict graph, 'd'ual ray, 'b'oth conflict graph and dual ray) # [type: char, advanced: FALSE, range: {ocdb}, default: b] conflict/useboundlp = b # should infeasible/bound exceeding strong branching conflict analysis be used? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] conflict/usesb = TRUE # should pseudo solution conflict analysis be used? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] conflict/usepseudo = TRUE # maximal fraction of variables involved in a conflict constraint # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.15] conflict/maxvarsfac = 0.15 # minimal absolute maximum of variables involved in a conflict constraint # [type: int, advanced: TRUE, range: [0,2147483647], default: 0] conflict/minmaxvars = 0 # maximal number of LP resolving loops during conflict analysis (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 2] conflict/maxlploops = 2 # maximal number of LP iterations in each LP resolving loop (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 10] conflict/lpiterations = 10 # number of depth levels up to which first UIP's are used in conflict analysis (-1: use All-FirstUIP rule) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] conflict/fuiplevels = -1 # maximal number of intermediate conflict constraints generated in conflict graph (-1: use every intermediate constraint) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] conflict/interconss = -1 # number of depth levels up to which UIP reconvergence constraints are generated (-1: generate reconvergence constraints in all depth levels) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] conflict/reconvlevels = -1 # maximal number of conflict constraints accepted at an infeasible node (-1: use all generated conflict constraints) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 10] conflict/maxconss = 10 # maximal size of conflict store (-1: auto, 0: disable storage) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 10000] conflict/maxstoresize = 10000 # should binary conflicts be preferred? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] conflict/preferbinary = FALSE # prefer infeasibility proof to boundexceeding proof # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] conflict/prefinfproof = TRUE # should conflict constraints be generated that are only valid locally? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] conflict/allowlocal = TRUE # should conflict constraints be attached only to the local subtree where they can be useful? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] conflict/settlelocal = FALSE # should earlier nodes be repropagated in order to replace branching decisions by deductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] conflict/repropagate = TRUE # should constraints be kept for repropagation even if they are too long? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] conflict/keepreprop = TRUE # should the conflict constraints be separated? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] conflict/separate = TRUE # should the conflict constraints be subject to aging? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] conflict/dynamic = TRUE # should the conflict's relaxations be subject to LP aging and cleanup? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] conflict/removable = TRUE # score factor for depth level in bound relaxation heuristic # [type: real, advanced: TRUE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 1] conflict/graph/depthscorefac = 1 # score factor for impact on acticity in bound relaxation heuristic # [type: real, advanced: TRUE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 1] conflict/proofscorefac = 1 # score factor for up locks in bound relaxation heuristic # [type: real, advanced: TRUE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 0] conflict/uplockscorefac = 0 # score factor for down locks in bound relaxation heuristic # [type: real, advanced: TRUE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 0] conflict/downlockscorefac = 0 # factor to decrease importance of variables' earlier conflict scores # [type: real, advanced: TRUE, range: [1e-06,1], default: 0.98] conflict/scorefac = 0.98 # number of successful conflict analysis calls that trigger a restart (0: disable conflict restarts) # [type: int, advanced: FALSE, range: [0,2147483647], default: 0] conflict/restartnum = 0 # factor to increase restartnum with after each restart # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 1.5] conflict/restartfac = 1.5 # should relaxed bounds be ignored? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] conflict/ignorerelaxedbd = FALSE # maximal number of variables to try to detect global bound implications and shorten the whole conflict set (0: disabled) # [type: int, advanced: TRUE, range: [0,2147483647], default: 250] conflict/maxvarsdetectimpliedbounds = 250 # try to shorten the whole conflict set or terminate early (depending on the 'maxvarsdetectimpliedbounds' parameter) # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] conflict/fullshortenconflict = TRUE # the weight the VSIDS score is weight by updating the VSIDS for a variable if it is part of a conflict # [type: real, advanced: FALSE, range: [0,1], default: 0] conflict/conflictweight = 0 # the weight the VSIDS score is weight by updating the VSIDS for a variable if it is part of a conflict graph # [type: real, advanced: FALSE, range: [0,1], default: 1] conflict/conflictgraphweight = 1 # minimal improvement of primal bound to remove conflicts based on a previous incumbent # [type: real, advanced: TRUE, range: [0,1], default: 0.05] conflict/minimprove = 0.05 # weight of the size of a conflict used in score calculation # [type: real, advanced: TRUE, range: [0,1], default: 0.001] conflict/weightsize = 0.001 # weight of the repropagation depth of a conflict used in score calculation # [type: real, advanced: TRUE, range: [0,1], default: 0.1] conflict/weightrepropdepth = 0.1 # weight of the valid depth of a conflict used in score calculation # [type: real, advanced: TRUE, range: [0,1], default: 1] conflict/weightvaliddepth = 1 # apply cut generating functions to construct alternative proofs # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] conflict/sepaaltproofs = FALSE # maximum age an unnecessary constraint can reach before it is deleted (0: dynamic, -1: keep all constraints) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 0] constraints/agelimit = 0 # age of a constraint after which it is marked obsolete (0: dynamic, -1 do not mark constraints obsolete) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/obsoleteage = -1 # should enforcement of pseudo solution be disabled? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/disableenfops = FALSE # verbosity level of output # [type: int, advanced: FALSE, range: [0,5], default: 4] display/verblevel = 4 # maximal number of characters in a node information line # [type: int, advanced: FALSE, range: [0,2147483647], default: 143] display/width = 143 # frequency for displaying node information lines # [type: int, advanced: FALSE, range: [-1,2147483647], default: 100] display/freq = 100 # frequency for displaying header lines (every n'th node information line) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 15] display/headerfreq = 15 # should the LP solver display status messages? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] display/lpinfo = FALSE # display all violations for a given start solution / the best solution after the solving process? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] display/allviols = FALSE # should the relevant statistics be displayed at the end of solving? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] display/relevantstats = TRUE # should setting of common subscip parameters include the activation of the UCT node selector? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/useuctsubscip = FALSE # should statistics be collected for variable domain value pairs? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] history/valuebased = FALSE # should variable histories be merged from sub-SCIPs whenever possible? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] history/allowmerge = FALSE # should variable histories be transferred to initialize SCIP copies? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] history/allowtransfer = FALSE # maximal time in seconds to run # [type: real, advanced: FALSE, range: [0,1e+20], default: 1e+20] limits/time = 1e+20 # maximal number of nodes to process (-1: no limit) # [type: longint, advanced: FALSE, range: [-1,9223372036854775807], default: -1] limits/nodes = -1 # maximal number of total nodes (incl. restarts) to process (-1: no limit) # [type: longint, advanced: FALSE, range: [-1,9223372036854775807], default: -1] limits/totalnodes = -1 # solving stops, if the given number of nodes was processed since the last improvement of the primal solution value (-1: no limit) # [type: longint, advanced: FALSE, range: [-1,9223372036854775807], default: -1] limits/stallnodes = -1 # maximal memory usage in MB; reported memory usage is lower than real memory usage! # [type: real, advanced: FALSE, range: [0,8796093022207], default: 8796093022207] limits/memory = 8796093022207 # solving stops, if the relative gap = |primal - dual|/MIN(|dual|,|primal|) is below the given value, the gap is 'Infinity', if primal and dual bound have opposite signs # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0] limits/gap = 0 # solving stops, if the absolute gap = |primalbound - dualbound| is below the given value # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0] limits/absgap = 0 # solving stops, if primal bound is at least as good as given value (alias objectivestop) # [type: real, advanced: FALSE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 1e+99] limits/primal = 1e+99 # solving stops, if dual bound is at least as good as given value # [type: real, advanced: FALSE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 1e+99] limits/dual = 1e+99 # solving stops, if the given number of solutions were found; this limit is first checked in presolving (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] limits/solutions = -1 # solving stops, if the given number of solution improvements were found (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] limits/bestsol = -1 # maximal number of solutions to store in the solution storage # [type: int, advanced: FALSE, range: [1,2147483647], default: 100] limits/maxsol = 100 # maximal number of solutions candidates to store in the solution storage of the original problem # [type: int, advanced: FALSE, range: [0,2147483647], default: 10] limits/maxorigsol = 10 # solving stops, if the given number of restarts was triggered (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] limits/restarts = -1 # if solve exceeds this number of nodes for the first time, an automatic restart is triggered (-1: no automatic restart) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] limits/autorestartnodes = -1 # frequency for solving LP at the nodes (-1: never; 0: only root LP) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] lp/solvefreq = 1 # iteration limit for each single LP solve (-1: no limit) # [type: longint, advanced: TRUE, range: [-1,9223372036854775807], default: -1] lp/iterlim = -1 # iteration limit for initial root LP solve (-1: no limit) # [type: longint, advanced: TRUE, range: [-1,9223372036854775807], default: -1] lp/rootiterlim = -1 # maximal depth for solving LP at the nodes (-1: no depth limit) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] lp/solvedepth = -1 # LP algorithm for solving initial LP relaxations (automatic 's'implex, 'p'rimal simplex, 'd'ual simplex, 'b'arrier, barrier with 'c'rossover) # [type: char, advanced: FALSE, range: {spdbc}, default: s] lp/initalgorithm = s # LP algorithm for resolving LP relaxations if a starting basis exists (automatic 's'implex, 'p'rimal simplex, 'd'ual simplex, 'b'arrier, barrier with 'c'rossover) # [type: char, advanced: FALSE, range: {spdbc}, default: s] lp/resolvealgorithm = s # LP pricing strategy ('l'pi default, 'a'uto, 'f'ull pricing, 'p'artial, 's'teepest edge pricing, 'q'uickstart steepest edge pricing, 'd'evex pricing) # [type: char, advanced: FALSE, range: {lafpsqd}, default: l] lp/pricing = l # should lp state be cleared at the end of probing mode when lp was initially unsolved, e.g., when called right after presolving? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] lp/clearinitialprobinglp = TRUE # should the LP be resolved to restore the state at start of diving (if FALSE we buffer the solution values)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] lp/resolverestore = FALSE # should the buffers for storing LP solution values during diving be freed at end of diving? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] lp/freesolvalbuffers = FALSE # maximum age a dynamic column can reach before it is deleted from the LP (-1: don't delete columns due to aging) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 10] lp/colagelimit = 10 # maximum age a dynamic row can reach before it is deleted from the LP (-1: don't delete rows due to aging) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 10] lp/rowagelimit = 10 # should new non-basic columns be removed after LP solving? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] lp/cleanupcols = FALSE # should new non-basic columns be removed after root LP solving? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] lp/cleanupcolsroot = FALSE # should new basic rows be removed after LP solving? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] lp/cleanuprows = TRUE # should new basic rows be removed after root LP solving? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] lp/cleanuprowsroot = TRUE # should LP solver's return status be checked for stability? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] lp/checkstability = TRUE # maximum condition number of LP basis counted as stable (-1.0: no limit) # [type: real, advanced: TRUE, range: [-1,1.79769313486232e+308], default: -1] lp/conditionlimit = -1 # minimal Markowitz threshold to control sparsity/stability in LU factorization # [type: real, advanced: TRUE, range: [0.0001,0.9999], default: 0.01] lp/minmarkowitz = 0.01 # should LP solutions be checked for primal feasibility, resolving LP when numerical troubles occur? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] lp/checkprimfeas = TRUE # should LP solutions be checked for dual feasibility, resolving LP when numerical troubles occur? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] lp/checkdualfeas = TRUE # should infeasibility proofs from the LP be checked? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] lp/checkfarkas = TRUE # which FASTMIP setting of LP solver should be used? 0: off, 1: low # [type: int, advanced: TRUE, range: [0,1], default: 1] lp/fastmip = 1 # LP scaling (0: none, 1: normal, 2: aggressive) # [type: int, advanced: TRUE, range: [0,2], default: 1] lp/scaling = 1 # should presolving of LP solver be used? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] lp/presolving = TRUE # should the lexicographic dual algorithm be used? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] lp/lexdualalgo = FALSE # should the lexicographic dual algorithm be applied only at the root node # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] lp/lexdualrootonly = TRUE # maximum number of rounds in the lexicographic dual algorithm (-1: unbounded) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 2] lp/lexdualmaxrounds = 2 # choose fractional basic variables in lexicographic dual algorithm? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] lp/lexdualbasic = FALSE # turn on the lex dual algorithm only when stalling? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] lp/lexdualstalling = TRUE # disable the cutoff bound in the LP solver? (0: enabled, 1: disabled, 2: auto) # [type: int, advanced: TRUE, range: [0,2], default: 2] lp/disablecutoff = 2 # simplex algorithm shall use row representation of the basis if number of rows divided by number of columns exceeds this value (-1.0 to disable row representation) # [type: real, advanced: TRUE, range: [-1,1.79769313486232e+308], default: 1.2] lp/rowrepswitch = 1.2 # number of threads used for solving the LP (0: automatic) # [type: int, advanced: TRUE, range: [0,64], default: 0] lp/threads = 0 # factor of average LP iterations that is used as LP iteration limit for LP resolve (-1: unlimited) # [type: real, advanced: TRUE, range: [-1,1.79769313486232e+308], default: -1] lp/resolveiterfac = -1 # minimum number of iterations that are allowed for LP resolve # [type: int, advanced: TRUE, range: [1,2147483647], default: 1000] lp/resolveitermin = 1000 # LP solution polishing method (0: disabled, 1: only root, 2: always, 3: auto) # [type: int, advanced: TRUE, range: [0,3], default: 3] lp/solutionpolishing = 3 # LP refactorization interval (0: auto) # [type: int, advanced: TRUE, range: [0,2147483647], default: 0] lp/refactorinterval = 0 # should the Farkas duals always be collected when an LP is found to be infeasible? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] lp/alwaysgetduals = FALSE # solver to use for solving NLPs; leave empty to select NLPI with highest priority # [type: string, advanced: FALSE, default: ""] nlp/solver = "" # should the NLP relaxation be always disabled (also for NLPs/MINLPs)? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] nlp/disable = FALSE # fraction of maximal memory usage resulting in switch to memory saving mode # [type: real, advanced: FALSE, range: [0,1], default: 0.8] memory/savefac = 0.8 # memory growing factor for dynamically allocated arrays # [type: real, advanced: TRUE, range: [1,10], default: 1.2] memory/arraygrowfac = 1.2 # initial size of dynamically allocated arrays # [type: int, advanced: TRUE, range: [0,2147483647], default: 4] memory/arraygrowinit = 4 # memory growing factor for tree array # [type: real, advanced: TRUE, range: [1,10], default: 2] memory/treegrowfac = 2 # initial size of tree array # [type: int, advanced: TRUE, range: [0,2147483647], default: 65536] memory/treegrowinit = 65536 # memory growing factor for path array # [type: real, advanced: TRUE, range: [1,10], default: 2] memory/pathgrowfac = 2 # initial size of path array # [type: int, advanced: TRUE, range: [0,2147483647], default: 256] memory/pathgrowinit = 256 # should the CTRL-C interrupt be caught by SCIP? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] misc/catchctrlc = TRUE # should a hashtable be used to map from variable names to variables? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] misc/usevartable = TRUE # should a hashtable be used to map from constraint names to constraints? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] misc/useconstable = TRUE # should smaller hashtables be used? yields better performance for small problems with about 100 variables # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] misc/usesmalltables = FALSE # should the statistics be reset if the transformed problem is freed (in case of a Benders' decomposition this parameter should be set to FALSE) # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] misc/resetstat = TRUE # should only solutions be checked which improve the primal bound # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] misc/improvingsols = FALSE # should the reason be printed if a given start solution is infeasible # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] misc/printreason = TRUE # should the usage of external memory be estimated? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] misc/estimexternmem = TRUE # try to avoid running into memory limit by restricting plugins like heuristics? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] misc/avoidmemout = TRUE # should SCIP try to transfer original solutions to the transformed space (after presolving)? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] misc/transorigsols = TRUE # should SCIP try to transfer transformed solutions to the original space (after solving)? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] misc/transsolsorig = TRUE # should SCIP calculate the primal dual integral value? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] misc/calcintegral = TRUE # should SCIP try to remove infinite fixings from solutions copied to the solution store? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] misc/finitesolutionstore = FALSE # should the best solution be transformed to the orignal space and be output in command line run? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] misc/outputorigsol = TRUE # should strong dual reductions be allowed in propagation and presolving? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] misc/allowstrongdualreds = TRUE # should weak dual reductions be allowed in propagation and presolving? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] misc/allowweakdualreds = TRUE # should the objective function be scaled so that it is always integer? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] misc/scaleobj = TRUE # should detailed statistics for diving heuristics be shown? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] misc/showdivingstats = FALSE # objective value for reference purposes # [type: real, advanced: FALSE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 1e+99] misc/referencevalue = 1e+99 # bitset describing used symmetry handling technique: (0: off; 1: constraint-based (orbitopes and/or symresacks); 2: orbital fixing; 3: orbitopes and orbital fixing; 4: Schreier Sims cuts; 5: Schreier Sims cuts and orbitopes; 6: Schreier Sims cuts and orbital fixing; 7: Schreier Sims cuts, orbitopes, and orbital fixing) See type_symmetry.h. # [type: int, advanced: FALSE, range: [0,7], default: 7] misc/usesymmetry = 7 # global shift of all random seeds in the plugins and the LP random seed # [type: int, advanced: FALSE, range: [0,2147483647], default: 0] randomization/randomseedshift = 0 # seed value for permuting the problem after reading/transformation (0: no permutation) # [type: int, advanced: FALSE, range: [0,2147483647], default: 0] randomization/permutationseed = 0 # should order of constraints be permuted (depends on permutationseed)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] randomization/permuteconss = TRUE # should order of variables be permuted (depends on permutationseed)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] randomization/permutevars = FALSE # random seed for LP solver, e.g. for perturbations in the simplex (0: LP default) # [type: int, advanced: FALSE, range: [0,2147483647], default: 0] randomization/lpseed = 0 # child selection rule ('d'own, 'u'p, 'p'seudo costs, 'i'nference, 'l'p value, 'r'oot LP value difference, 'h'ybrid inference/root LP value difference) # [type: char, advanced: FALSE, range: {dupilrh}, default: h] nodeselection/childsel = h # values larger than this are considered infinity # [type: real, advanced: FALSE, range: [10000000000,1e+98], default: 1e+20] numerics/infinity = 1e+20 # absolute values smaller than this are considered zero # [type: real, advanced: FALSE, range: [1e-20,0.001], default: 1e-09] numerics/epsilon = 1e-09 # absolute values of sums smaller than this are considered zero # [type: real, advanced: FALSE, range: [1e-17,0.001], default: 1e-06] numerics/sumepsilon = 1e-06 # feasibility tolerance for constraints # [type: real, advanced: FALSE, range: [1e-17,0.001], default: 1e-06] numerics/feastol = 1e-06 # feasibility tolerance factor; for checking the feasibility of the best solution # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 1] numerics/checkfeastolfac = 1 # factor w.r.t. primal feasibility tolerance that determines default (and maximal) primal feasibility tolerance of LP solver # [type: real, advanced: FALSE, range: [1e-06,1], default: 1] numerics/lpfeastolfactor = 1 # feasibility tolerance for reduced costs in LP solution # [type: real, advanced: FALSE, range: [1e-17,0.001], default: 1e-07] numerics/dualfeastol = 1e-07 # LP convergence tolerance used in barrier algorithm # [type: real, advanced: TRUE, range: [1e-17,0.001], default: 1e-10] numerics/barrierconvtol = 1e-10 # minimal relative improve for strengthening bounds # [type: real, advanced: TRUE, range: [1e-17,1e+98], default: 0.05] numerics/boundstreps = 0.05 # minimal variable distance value to use for branching pseudo cost updates # [type: real, advanced: TRUE, range: [1e-17,1], default: 0.1] numerics/pseudocosteps = 0.1 # minimal objective distance value to use for branching pseudo cost updates # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.0001] numerics/pseudocostdelta = 0.0001 # minimal decrease factor that causes the recomputation of a value (e.g., pseudo objective) instead of an update # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 10000000] numerics/recomputefac = 10000000 # values larger than this are considered huge and should be handled separately (e.g., in activity computation) # [type: real, advanced: TRUE, range: [0,1e+98], default: 1e+15] numerics/hugeval = 1e+15 # maximal number of presolving rounds (-1: unlimited, 0: off) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] presolving/maxrounds = -1 # abort presolve, if at most this fraction of the problem was changed in last presolve round # [type: real, advanced: TRUE, range: [0,1], default: 0.0008] presolving/abortfac = 0.0008 # maximal number of restarts (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] presolving/maxrestarts = -1 # fraction of integer variables that were fixed in the root node triggering a restart with preprocessing after root node evaluation # [type: real, advanced: TRUE, range: [0,1], default: 0.025] presolving/restartfac = 0.025 # limit on number of entries in clique table relative to number of problem nonzeros # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 2] presolving/clqtablefac = 2 # fraction of integer variables that were fixed in the root node triggering an immediate restart with preprocessing # [type: real, advanced: TRUE, range: [0,1], default: 0.1] presolving/immrestartfac = 0.1 # fraction of integer variables that were globally fixed during the solving process triggering a restart with preprocessing # [type: real, advanced: TRUE, range: [0,1], default: 1] presolving/subrestartfac = 1 # minimal fraction of integer variables removed after restart to allow for an additional restart # [type: real, advanced: TRUE, range: [0,1], default: 0.1] presolving/restartminred = 0.1 # should multi-aggregation of variables be forbidden? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] presolving/donotmultaggr = FALSE # should aggregation of variables be forbidden? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] presolving/donotaggr = FALSE # maximal number of variables priced in per pricing round # [type: int, advanced: FALSE, range: [1,2147483647], default: 100] pricing/maxvars = 100 # maximal number of priced variables at the root node # [type: int, advanced: FALSE, range: [1,2147483647], default: 2000] pricing/maxvarsroot = 2000 # pricing is aborted, if fac * pricing/maxvars pricing candidates were found # [type: real, advanced: FALSE, range: [1,1.79769313486232e+308], default: 2] pricing/abortfac = 2 # should variables created at the current node be deleted when the node is solved in case they are not present in the LP anymore? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] pricing/delvars = FALSE # should variables created at the root node be deleted when the root is solved in case they are not present in the LP anymore? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] pricing/delvarsroot = FALSE # should the variables be labelled for the application of Benders' decomposition? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] decomposition/benderslabels = FALSE # if a decomposition exists, should Benders' decomposition be applied? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] decomposition/applybenders = FALSE # maximum number of edges in block graph computation (-1: no limit, 0: disable block graph computation) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 10000] decomposition/maxgraphedge = 10000 # disable expensive measures # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] decomposition/disablemeasures = FALSE # the tolerance used for checking optimality in Benders' decomposition. tol where optimality is given by LB + tol > UB. # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 1e-06] benders/solutiontol = 1e-06 # should Benders' cuts be generated from the solution to the LP relaxation? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] benders/cutlpsol = TRUE # should Benders' decomposition be copied for use in sub-SCIPs? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] benders/copybenders = TRUE # maximal number of propagation rounds per node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 100] propagating/maxrounds = 100 # maximal number of propagation rounds in the root node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1000] propagating/maxroundsroot = 1000 # should propagation be aborted immediately? setting this to FALSE could help conflict analysis to produce more conflict constraints # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] propagating/abortoncutoff = TRUE # should reoptimization used? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] reoptimization/enable = FALSE # maximal number of saved nodes # [type: int, advanced: TRUE, range: [-1,2147483647], default: 2147483647] reoptimization/maxsavednodes = 2147483647 # maximal number of bound changes between two stored nodes on one path # [type: int, advanced: TRUE, range: [0,2147483647], default: 2147483647] reoptimization/maxdiffofnodes = 2147483647 # save global constraints to separate infeasible subtrees. # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] reoptimization/globalcons/sepainfsubtrees = TRUE # separate the optimal solution, i.e., for constrained shortest path # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] reoptimization/sepabestsol = FALSE # use variable history of the previous solve if the objctive function has changed only slightly # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] reoptimization/storevarhistory = FALSE # re-use pseudo costs if the objective function changed only slightly # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] reoptimization/usepscost = FALSE # at which reopttype should the LP be solved? (1: transit, 3: strong branched, 4: w/ added logicor, 5: only leafs). # [type: int, advanced: TRUE, range: [1,5], default: 1] reoptimization/solvelp = 1 # maximal number of bound changes at node to skip solving the LP # [type: int, advanced: TRUE, range: [0,2147483647], default: 1] reoptimization/solvelpdiff = 1 # number of best solutions which should be saved for the following runs. (-1: save all) # [type: int, advanced: TRUE, range: [0,2147483647], default: 2147483647] reoptimization/savesols = 2147483647 # similarity of two sequential objective function to disable solving the root LP. # [type: real, advanced: TRUE, range: [-1,1], default: 0.8] reoptimization/objsimrootLP = 0.8 # similarity of two objective functions to re-use stored solutions # [type: real, advanced: TRUE, range: [-1,1], default: -1] reoptimization/objsimsol = -1 # minimum similarity for using reoptimization of the search tree. # [type: real, advanced: TRUE, range: [-1,1], default: -1] reoptimization/delay = -1 # time limit over all reoptimization rounds?. # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] reoptimization/commontimelimit = FALSE # replace branched inner nodes by their child nodes, if the number of bound changes is not to large # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] reoptimization/shrinkinner = TRUE # try to fix variables at the root node before reoptimizing by probing like strong branching # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] reoptimization/strongbranchinginit = TRUE # delete stored nodes which were not reoptimized # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] reoptimization/reducetofrontier = TRUE # force a restart if the last n optimal solutions were found by heuristic reoptsols # [type: int, advanced: TRUE, range: [1,2147483647], default: 3] reoptimization/forceheurrestart = 3 # save constraint propagations # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] reoptimization/saveconsprop = FALSE # use constraints to reconstruct the subtree pruned be dual reduction when reactivating the node # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] reoptimization/usesplitcons = TRUE # use 'd'efault, 'r'andom or a variable ordering based on 'i'nference score for interdiction branching used during reoptimization # [type: char, advanced: TRUE, range: {dir}, default: d] reoptimization/varorderinterdiction = d # reoptimize cuts found at the root node # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] reoptimization/usecuts = FALSE # maximal age of a cut to be use for reoptimization # [type: int, advanced: TRUE, range: [0,2147483647], default: 0] reoptimization/maxcutage = 0 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separation (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: FALSE, range: [0,1], default: 1] separating/maxbounddist = 1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying local separation (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: FALSE, range: [0,1], default: 0] separating/maxlocalbounddist = 0 # maximal ratio between coefficients in strongcg, cmir, and flowcover cuts # [type: real, advanced: FALSE, range: [1,1e+98], default: 10000] separating/maxcoefratio = 10000 # maximal ratio between coefficients (as factor of 1/feastol) to ensure in rowprep cleanup # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 10] separating/maxcoefratiofacrowprep = 10 # minimal efficacy for a cut to enter the LP # [type: real, advanced: FALSE, range: [0,1e+98], default: 0.0001] separating/minefficacy = 0.0001 # minimal efficacy for a cut to enter the LP in the root node # [type: real, advanced: FALSE, range: [0,1e+98], default: 0.0001] separating/minefficacyroot = 0.0001 # minimum cut activity quotient to convert cuts into constraints during a restart (0.0: all cuts are converted) # [type: real, advanced: FALSE, range: [0,1], default: 0.8] separating/minactivityquot = 0.8 # factor w.r.t. maxcuts for maximal number of cuts generated per separation round (-1.0: no limit, >= 0.0: valid finite limit) # [type: real, advanced: FALSE, range: [-1,1.79769313486232e+308], default: 2] separating/maxcutsgenfactor = 2 # factor w.r.t. maxcutsroot for maximal number of generated cuts per separation round at the root node (-1.0: no limit, >= 0.0: valid finite limit) # [type: real, advanced: FALSE, range: [-1,1.79769313486232e+308], default: 2] separating/maxcutsrootgenfactor = 2 # function used for calc. scalar prod. in orthogonality test ('e'uclidean, 'd'iscrete) # [type: char, advanced: TRUE, range: {ed}, default: e] separating/orthofunc = e # row norm to use for efficacy calculation ('e'uclidean, 'm'aximum, 's'um, 'd'iscrete) # [type: char, advanced: TRUE, range: {emsd}, default: e] separating/efficacynorm = e # cut selection during restart ('a'ge, activity 'q'uotient) # [type: char, advanced: TRUE, range: {aq}, default: a] separating/cutselrestart = a # cut selection for sub SCIPs ('a'ge, activity 'q'uotient) # [type: char, advanced: TRUE, range: {aq}, default: a] separating/cutselsubscip = a # should cutpool separate only cuts with high relative efficacy? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/filtercutpoolrel = FALSE # maximal number of runs for which separation is enabled (-1: unlimited) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] separating/maxruns = -1 # maximal number of separation rounds per node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] separating/maxrounds = -1 # maximal number of separation rounds in the root node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] separating/maxroundsroot = -1 # maximal number of separation rounds in the root node of a subsequent run (-1: unlimited) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] separating/maxroundsrootsubrun = -1 # maximal additional number of separation rounds in subsequent price-and-cut loops (-1: no additional restriction) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 1] separating/maxaddrounds = 1 # maximal number of consecutive separation rounds without objective or integrality improvement in local nodes (-1: no additional restriction) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1] separating/maxstallrounds = 1 # maximal number of consecutive separation rounds without objective or integrality improvement in the root node (-1: no additional restriction) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 10] separating/maxstallroundsroot = 10 # maximal number of cuts separated per separation round (0: disable local separation) # [type: int, advanced: FALSE, range: [0,2147483647], default: 100] separating/maxcuts = 100 # maximal number of separated cuts per separation round at the root node (0: disable root node separation) # [type: int, advanced: FALSE, range: [0,2147483647], default: 2000] separating/maxcutsroot = 2000 # maximum age a cut can reach before it is deleted from the global cut pool, or -1 to keep all cuts # [type: int, advanced: TRUE, range: [-1,2147483647], default: 80] separating/cutagelimit = 80 # separation frequency for the global cut pool (-1: disable global cut pool, 0: only separate pool at the root) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] separating/poolfreq = 10 # parallel optimisation mode, 0: opportunistic or 1: deterministic. # [type: int, advanced: FALSE, range: [0,1], default: 1] parallel/mode = 1 # the minimum number of threads used during parallel solve # [type: int, advanced: FALSE, range: [0,64], default: 1] parallel/minnthreads = 1 # the maximum number of threads used during parallel solve # [type: int, advanced: FALSE, range: [0,64], default: 8] parallel/maxnthreads = 8 # set different random seeds in each concurrent solver? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] concurrent/changeseeds = TRUE # use different child selection rules in each concurrent solver? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] concurrent/changechildsel = TRUE # should the concurrent solvers communicate global variable bound changes? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] concurrent/commvarbnds = TRUE # should the problem be presolved before it is copied to the concurrent solvers? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] concurrent/presolvebefore = TRUE # maximum number of solutions that will be shared in a one synchronization # [type: int, advanced: FALSE, range: [0,2147483647], default: 5131912] concurrent/initseed = 5131912 # initial frequency of synchronization with other threads # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 10] concurrent/sync/freqinit = 10 # maximal frequency of synchronization with other threads # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 10] concurrent/sync/freqmax = 10 # factor by which the frequency of synchronization is changed # [type: real, advanced: FALSE, range: [1,1.79769313486232e+308], default: 1.5] concurrent/sync/freqfactor = 1.5 # when adapting the synchronization frequency this value is the targeted relative difference by which the absolute gap decreases per synchronization # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.001] concurrent/sync/targetprogress = 0.001 # maximum number of solutions that will be shared in a single synchronization # [type: int, advanced: FALSE, range: [0,1000], default: 3] concurrent/sync/maxnsols = 3 # maximum number of synchronizations before reading is enforced regardless of delay # [type: int, advanced: TRUE, range: [0,100], default: 7] concurrent/sync/maxnsyncdelay = 7 # minimum delay before synchronization data is read # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 10] concurrent/sync/minsyncdelay = 10 # how many of the N best solutions should be considered for synchronization? # [type: int, advanced: FALSE, range: [0,2147483647], default: 10] concurrent/sync/nbestsols = 10 # path prefix for parameter setting files of concurrent solvers # [type: string, advanced: FALSE, default: ""] concurrent/paramsetprefix = "" # default clock type (1: CPU user seconds, 2: wall clock time) # [type: int, advanced: FALSE, range: [1,2], default: 2] timing/clocktype = 2 # is timing enabled? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] timing/enabled = TRUE # belongs reading time to solving time? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] timing/reading = FALSE # should clock checks of solving time be performed less frequently (note: time limit could be exceeded slightly) # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] timing/rareclockcheck = FALSE # should timing for statistic output be performed? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] timing/statistictiming = TRUE # should time for evaluation in NLP solves be measured? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] timing/nlpieval = FALSE # name of the VBC tool output file, or - if no VBC tool output should be created # [type: string, advanced: FALSE, default: "-"] visual/vbcfilename = "-" # name of the BAK tool output file, or - if no BAK tool output should be created # [type: string, advanced: FALSE, default: "-"] visual/bakfilename = "-" # should the real solving time be used instead of a time step counter in visualization? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] visual/realtime = TRUE # should the node where solutions are found be visualized? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] visual/dispsols = FALSE # should lower bound information be visualized? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] visual/displb = FALSE # should be output the external value of the objective? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] visual/objextern = TRUE # should model constraints be marked as initial? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] reading/initialconss = TRUE # should model constraints be subject to aging? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] reading/dynamicconss = TRUE # should columns be added and removed dynamically to the LP? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] reading/dynamiccols = FALSE # should rows be added and removed dynamically to the LP? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] reading/dynamicrows = FALSE # should all constraints be written (including the redundant constraints)? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] write/allconss = FALSE # should variables set to zero be printed? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] write/printzeros = FALSE # when writing a generic problem the index for the first variable should start with? # [type: int, advanced: FALSE, range: [0,1073741823], default: 0] write/genericnamesoffset = 0 # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/nonlinear/sepafreq = 1 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/nonlinear/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/nonlinear/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/nonlinear/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/nonlinear/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/nonlinear/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/nonlinear/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 28] constraints/nonlinear/presoltiming = 28 # limit on number of propagation rounds for a set of constraints within one round of SCIP propagation # [type: int, advanced: FALSE, range: [0,2147483647], default: 10] constraints/nonlinear/maxproprounds = 10 # whether to check bounds of all auxiliary variable to seed reverse propagation # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/nonlinear/propauxvars = TRUE # strategy on how to relax variable bounds during bound tightening: relax (n)ot, relax by (a)bsolute value, relax always by a(b)solute value, relax by (r)relative value # [type: char, advanced: TRUE, range: {nabr}, default: r] constraints/nonlinear/varboundrelax = r # by how much to relax variable bounds during bound tightening if strategy 'a', 'b', or 'r' # [type: real, advanced: TRUE, range: [0,1], default: 1e-09] constraints/nonlinear/varboundrelaxamount = 1e-09 # by how much to relax constraint sides during bound tightening # [type: real, advanced: TRUE, range: [0,1], default: 1e-09] constraints/nonlinear/conssiderelaxamount = 1e-09 # maximal relative perturbation of reference point when computing facet of envelope of vertex-polyhedral function (dim>2) # [type: real, advanced: TRUE, range: [0,1], default: 0.001] constraints/nonlinear/vpmaxperturb = 0.001 # adjust computed facet of envelope of vertex-polyhedral function up to a violation of this value times LP feasibility tolerance # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 10] constraints/nonlinear/vpadjfacetthresh = 10 # whether to use dual simplex instead of primal simplex for LP that computes facet of vertex-polyhedral function # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/nonlinear/vpdualsimplex = TRUE # maximal number of auxiliary expressions per bilinear term # [type: int, advanced: FALSE, range: [0,2147483647], default: 10] constraints/nonlinear/bilinmaxnauxexprs = 10 # whether to reformulate products of binary variables during presolving # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/nonlinear/reformbinprods = TRUE # whether to use the AND constraint handler for reformulating binary products # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/nonlinear/reformbinprodsand = TRUE # minimum number of terms to reformulate bilinear binary products by factorizing variables (<= 1: disabled) # [type: int, advanced: FALSE, range: [1,2147483647], default: 50] constraints/nonlinear/reformbinprodsfac = 50 # whether to forbid multiaggregation of nonlinear variables # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/nonlinear/forbidmultaggrnlvar = TRUE # whether to tighten LP feasibility tolerance during enforcement, if it seems useful # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/nonlinear/tightenlpfeastol = TRUE # whether to (re)run propagation in enforcement # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/nonlinear/propinenforce = FALSE # threshold for when to regard a cut from an estimator as weak (lower values allow more weak cuts) # [type: real, advanced: TRUE, range: [0,1], default: 0.2] constraints/nonlinear/weakcutthreshold = 0.2 # "strong" cuts will be scaled to have their maximal coef in [1/strongcutmaxcoef,strongcutmaxcoef] # [type: real, advanced: TRUE, range: [1,1e+20], default: 1000] constraints/nonlinear/strongcutmaxcoef = 1000 # consider efficacy requirement when deciding whether a cut is "strong" # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/nonlinear/strongcutefficacy = FALSE # whether to force "strong" cuts in enforcement # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/nonlinear/forcestrongcut = FALSE # an expression will be enforced if the "auxiliary" violation is at least this factor times the "original" violation # [type: real, advanced: TRUE, range: [0,1], default: 0.01] constraints/nonlinear/enfoauxviolfactor = 0.01 # retry enfo of constraint with weak cuts if violation is least this factor of maximal violated constraints # [type: real, advanced: TRUE, range: [0,2], default: 0.5] constraints/nonlinear/weakcutminviolfactor = 0.5 # whether to make rows to be non-removable in the node where they are added (can prevent some cycling): 'o'ff, in 'e'nforcement only, 'a'lways # [type: char, advanced: TRUE, range: {oea}, default: o] constraints/nonlinear/rownotremovable = o # method how to scale violations to make them comparable (not used for feasibility check): (n)one, (a)ctivity and side, norm of (g)radient # [type: char, advanced: TRUE, range: {nag}, default: n] constraints/nonlinear/violscale = n # whether variables contained in a single constraint should be forced to be at their lower or upper bounds ('d'isable, change 't'ype, add 'b'ound disjunction) # [type: char, advanced: TRUE, range: {bdt}, default: t] constraints/nonlinear/checkvarlocks = t # from which depth on in the tree to allow branching on auxiliary variables (variables added for extended formulation) # [type: int, advanced: FALSE, range: [0,2147483647], default: 2147483647] constraints/nonlinear/branching/aux = 2147483647 # whether to use external branching candidates and branching rules for branching # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] constraints/nonlinear/branching/external = FALSE # consider a constraint highly violated if its violation is >= this factor * maximal violation among all constraints # [type: real, advanced: FALSE, range: [0,1], default: 0] constraints/nonlinear/branching/highviolfactor = 0 # consider a variable branching score high if its branching score >= this factor * maximal branching score among all variables # [type: real, advanced: FALSE, range: [0,1], default: 0.9] constraints/nonlinear/branching/highscorefactor = 0.9 # weight by how much to consider the violation assigned to a variable for its branching score # [type: real, advanced: FALSE, range: [0,1e+20], default: 1] constraints/nonlinear/branching/violweight = 1 # weight by how much to consider fractionality of integer variables in branching score for spatial branching # [type: real, advanced: FALSE, range: [0,1e+20], default: 1] constraints/nonlinear/branching/fracweight = 1 # weight by how much to consider the dual values of rows that contain a variable for its branching score # [type: real, advanced: FALSE, range: [0,1e+20], default: 0] constraints/nonlinear/branching/dualweight = 0 # weight by how much to consider the pseudo cost of a variable for its branching score # [type: real, advanced: FALSE, range: [0,1e+20], default: 1] constraints/nonlinear/branching/pscostweight = 1 # weight by how much to consider the domain width in branching score # [type: real, advanced: FALSE, range: [0,1e+20], default: 0] constraints/nonlinear/branching/domainweight = 0 # weight by how much to consider variable type (continuous: 0, binary: 1, integer: 0.1, impl-integer: 0.01) in branching score # [type: real, advanced: FALSE, range: [0,1e+20], default: 0.5] constraints/nonlinear/branching/vartypeweight = 0.5 # how to aggregate several branching scores given for the same expression: 'a'verage, 'm'aximum, 's'um # [type: char, advanced: FALSE, range: {ams}, default: s] constraints/nonlinear/branching/scoreagg = s # method used to split violation in expression onto variables: 'u'niform, 'm'idness of solution, 'd'omain width, 'l'ogarithmic domain width # [type: char, advanced: FALSE, range: {umdl}, default: m] constraints/nonlinear/branching/violsplit = m # minimum pseudo-cost update count required to consider pseudo-costs reliable # [type: real, advanced: FALSE, range: [0,1e+20], default: 2] constraints/nonlinear/branching/pscostreliable = 2 # minimal average pseudo cost count for discrete variables at which to start considering spatial branching before branching on fractional integer variables # [type: real, advanced: FALSE, range: [0,1e+20], default: 1e+20] constraints/nonlinear/branching/mixfractional = 1e+20 # whether tight linearizations of nonlinear constraints should be added to cutpool when some heuristics finds a new solution ('o'ff, on new 'i'ncumbents, on 'e'very solution) # [type: char, advanced: FALSE, range: {oie}, default: o] constraints/nonlinear/linearizeheursol = o # whether to assume that any constraint is convex # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] constraints/nonlinear/assumeconvex = FALSE # is statistics table <cons_nonlinear> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] table/cons_nonlinear/active = FALSE # is statistics table <nlhdlr> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/nlhdlr/active = TRUE # priority of conflict handler <linear> # [type: int, advanced: TRUE, range: [-2147483648,2147483647], default: -1000000] conflict/linear/priority = -1000000 # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] constraints/linear/sepafreq = 0 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/linear/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/linear/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/linear/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/linear/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/linear/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/linear/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 20] constraints/linear/presoltiming = 20 # enable nonlinear upgrading for constraint handler <linear> # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/nonlinear/upgrade/linear = TRUE # multiplier on propagation frequency, how often the bounds are tightened (-1: never, 0: only at root) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 1] constraints/linear/tightenboundsfreq = 1 # maximal number of separation rounds per node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 5] constraints/linear/maxrounds = 5 # maximal number of separation rounds per node in the root node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] constraints/linear/maxroundsroot = -1 # maximal number of cuts separated per separation round # [type: int, advanced: FALSE, range: [0,2147483647], default: 50] constraints/linear/maxsepacuts = 50 # maximal number of cuts separated per separation round in the root node # [type: int, advanced: FALSE, range: [0,2147483647], default: 200] constraints/linear/maxsepacutsroot = 200 # should pairwise constraint comparison be performed in presolving? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/presolpairwise = TRUE # should hash table be used for detecting redundant constraints in advance # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/presolusehashing = TRUE # number for minimal pairwise presolve comparisons # [type: int, advanced: TRUE, range: [1,2147483647], default: 200000] constraints/linear/nmincomparisons = 200000 # minimal gain per minimal pairwise presolve comparisons to repeat pairwise comparison round # [type: real, advanced: TRUE, range: [0,1], default: 1e-06] constraints/linear/mingainpernmincomparisons = 1e-06 # maximal allowed relative gain in maximum norm for constraint aggregation (0.0: disable constraint aggregation) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] constraints/linear/maxaggrnormscale = 0 # maximum activity delta to run easy propagation on linear constraint (faster, but numerically less stable) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 1000000] constraints/linear/maxeasyactivitydelta = 1000000 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for separating knapsack cardinality cuts # [type: real, advanced: TRUE, range: [0,1], default: 0] constraints/linear/maxcardbounddist = 0 # should all constraints be subject to cardinality cut generation instead of only the ones with non-zero dual value? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] constraints/linear/separateall = FALSE # should presolving search for aggregations in equations # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/aggregatevariables = TRUE # should presolving try to simplify inequalities # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/simplifyinequalities = TRUE # should dual presolving steps be performed? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/dualpresolving = TRUE # should stuffing of singleton continuous variables be performed? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/singletonstuffing = TRUE # should single variable stuffing be performed, which tries to fulfill constraints using the cheapest variable? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/linear/singlevarstuffing = FALSE # apply binaries sorting in decr. order of coeff abs value? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/sortvars = TRUE # should the violation for a constraint with side 0.0 be checked relative to 1.0 (FALSE) or to the maximum absolute value in the activity (TRUE)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/linear/checkrelmaxabs = FALSE # should presolving try to detect constraints parallel to the objective function defining an upper bound and prevent these constraints from entering the LP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/detectcutoffbound = TRUE # should presolving try to detect constraints parallel to the objective function defining a lower bound and prevent these constraints from entering the LP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/detectlowerbound = TRUE # should presolving try to detect subsets of constraints parallel to the objective function? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/detectpartialobjective = TRUE # should presolving and propagation try to improve bounds, detect infeasibility, and extract sub-constraints from ranged rows and equations? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/rangedrowpropagation = TRUE # should presolving and propagation extract sub-constraints from ranged rows and equations? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/rangedrowartcons = TRUE # maximum depth to apply ranged row propagation # [type: int, advanced: TRUE, range: [0,2147483647], default: 2147483647] constraints/linear/rangedrowmaxdepth = 2147483647 # frequency for applying ranged row propagation # [type: int, advanced: TRUE, range: [1,1073741822], default: 1] constraints/linear/rangedrowfreq = 1 # should multi-aggregations only be performed if the constraint can be removed afterwards? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/linear/multaggrremove = FALSE # maximum coefficient dynamism (ie. maxabsval / minabsval) for primal multiaggregation # [type: real, advanced: TRUE, range: [1,1.79769313486232e+308], default: 1000] constraints/linear/maxmultaggrquot = 1000 # maximum coefficient dynamism (ie. maxabsval / minabsval) for dual multiaggregation # [type: real, advanced: TRUE, range: [1,1.79769313486232e+308], default: 1e+20] constraints/linear/maxdualmultaggrquot = 1e+20 # should Cliques be extracted? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/extractcliques = TRUE # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/and/sepafreq = 1 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/and/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/and/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/and/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/and/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/and/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/and/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 20] constraints/and/presoltiming = 20 # should pairwise constraint comparison be performed in presolving? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/and/presolpairwise = TRUE # should hash table be used for detecting redundant constraints in advance # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/and/presolusehashing = TRUE # should the AND-constraint get linearized and removed (in presolving)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/and/linearize = FALSE # should cuts be separated during LP enforcing? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/and/enforcecuts = TRUE # should an aggregated linearization be used? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/and/aggrlinearization = FALSE # should all binary resultant variables be upgraded to implicit binary variables? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/and/upgraderesultant = TRUE # should dual presolving be performed? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/and/dualpresolving = TRUE # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/benders/sepafreq = -1 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/benders/propfreq = -1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/benders/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/benders/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 0] constraints/benders/maxprerounds = 0 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/benders/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/benders/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 4] constraints/benders/presoltiming = 4 # is the Benders' decomposition constraint handler active? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] constraints/benders/active = FALSE # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/benderslp/sepafreq = -1 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/benderslp/propfreq = -1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/benderslp/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/benderslp/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 0] constraints/benderslp/maxprerounds = 0 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/benderslp/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/benderslp/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 28] constraints/benderslp/presoltiming = 28 # depth at which Benders' decomposition cuts are generated from the LP solution (-1: always, 0: only at root) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 0] constraints/benderslp/maxdepth = 0 # the depth frequency for generating LP cuts after the max depth is reached (0: never, 1: all nodes, ...) # [type: int, advanced: TRUE, range: [0,1073741822], default: 0] constraints/benderslp/depthfreq = 0 # the number of nodes processed without a dual bound improvement before enforcing the LP relaxation, 0: no stall count applied # [type: int, advanced: TRUE, range: [0,2147483647], default: 100] constraints/benderslp/stalllimit = 100 # after the root node, only iterlimit fractional LP solutions are used at each node to generate Benders' decomposition cuts. # [type: int, advanced: TRUE, range: [0,2147483647], default: 100] constraints/benderslp/iterlimit = 100 # is the Benders' decomposition LP cut constraint handler active? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] constraints/benderslp/active = FALSE # maximal percantage of continuous variables within a conflict # [type: real, advanced: FALSE, range: [0,1], default: 0.4] conflict/bounddisjunction/continuousfrac = 0.4 # priority of conflict handler <bounddisjunction> # [type: int, advanced: TRUE, range: [-2147483648,2147483647], default: -3000000] conflict/bounddisjunction/priority = -3000000 # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/bounddisjunction/sepafreq = -1 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/bounddisjunction/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/bounddisjunction/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/bounddisjunction/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/bounddisjunction/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/bounddisjunction/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/bounddisjunction/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 4] constraints/bounddisjunction/presoltiming = 4 # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] constraints/cardinality/sepafreq = 10 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/cardinality/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/cardinality/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/cardinality/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/cardinality/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/cardinality/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/cardinality/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 4] constraints/cardinality/presoltiming = 4 # whether to use balanced instead of unbalanced branching # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/cardinality/branchbalanced = FALSE # maximum depth for using balanced branching (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 20] constraints/cardinality/balanceddepth = 20 # determines that balanced branching is only used if the branching cut off value w.r.t. the current LP solution is greater than a given value # [type: real, advanced: TRUE, range: [0.01,1.79769313486232e+308], default: 2] constraints/cardinality/balancedcutoff = 2 # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/conjunction/sepafreq = -1 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/conjunction/propfreq = -1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/conjunction/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/conjunction/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/conjunction/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/conjunction/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/conjunction/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 4] constraints/conjunction/presoltiming = 4 # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/countsols/sepafreq = -1 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/countsols/propfreq = -1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/countsols/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/countsols/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 0] constraints/countsols/maxprerounds = 0 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/countsols/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/countsols/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 28] constraints/countsols/presoltiming = 28 # is the constraint handler active? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] constraints/countsols/active = FALSE # should the sparse solution test be turned on? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/countsols/sparsetest = TRUE # is it allowed to discard solutions? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/countsols/discardsols = TRUE # should the solutions be collected? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] constraints/countsols/collect = FALSE # counting stops, if the given number of solutions were found (-1: no limit) # [type: longint, advanced: FALSE, range: [-1,9223372036854775807], default: -1] constraints/countsols/sollimit = -1 # display activation status of display column <sols> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 0] display/sols/active = 0 # display activation status of display column <feasST> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 0] display/feasST/active = 0 # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/cumulative/sepafreq = 1 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/cumulative/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/cumulative/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/cumulative/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/cumulative/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/cumulative/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/cumulative/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 28] constraints/cumulative/presoltiming = 28 # should time-table (core-times) propagator be used to infer bounds? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/cumulative/ttinfer = TRUE # should edge-finding be used to detect an overload? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] constraints/cumulative/efcheck = FALSE # should edge-finding be used to infer bounds? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] constraints/cumulative/efinfer = FALSE # should edge-finding be executed? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/cumulative/useadjustedjobs = FALSE # should time-table edge-finding be used to detect an overload? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/cumulative/ttefcheck = TRUE # should time-table edge-finding be used to infer bounds? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/cumulative/ttefinfer = TRUE # should the binary representation be used? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] constraints/cumulative/usebinvars = FALSE # should cuts be added only locally? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] constraints/cumulative/localcuts = FALSE # should covering cuts be added every node? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/cumulative/usecovercuts = TRUE # should the cumulative constraint create cuts as knapsack constraints? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/cumulative/cutsasconss = TRUE # shall old sepa algo be applied? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/cumulative/sepaold = TRUE # should branching candidates be added to storage? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] constraints/cumulative/fillbranchcands = FALSE # should dual presolving be applied? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/cumulative/dualpresolve = TRUE # should coefficient tightening be applied? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] constraints/cumulative/coeftightening = FALSE # should demands and capacity be normalized? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/cumulative/normalize = TRUE # should pairwise constraint comparison be performed in presolving? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/cumulative/presolpairwise = TRUE # extract disjunctive constraints? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/cumulative/disjunctive = TRUE # number of branch-and-bound nodes to solve an independent cumulative constraint (-1: no limit)? # [type: longint, advanced: FALSE, range: [-1,9223372036854775807], default: 10000] constraints/cumulative/maxnodes = 10000 # search for conflict set via maximal cliques to detect disjunctive constraints # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/cumulative/detectdisjunctive = TRUE # search for conflict set via maximal cliques to detect variable bound constraints # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/cumulative/detectvarbounds = TRUE # should bound widening be used during the conflict analysis? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/cumulative/usebdwidening = TRUE # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/disjunction/sepafreq = -1 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/disjunction/propfreq = -1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/disjunction/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/disjunction/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/disjunction/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/disjunction/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/disjunction/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 4] constraints/disjunction/presoltiming = 4 # alawys perform branching if one of the constraints is violated, otherwise only if all integers are fixed # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/disjunction/alwaysbranch = TRUE # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/fixedvar/sepafreq = -1 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/fixedvar/propfreq = -1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/fixedvar/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] constraints/fixedvar/eagerfreq = -1 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 0] constraints/fixedvar/maxprerounds = 0 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/fixedvar/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/fixedvar/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 28] constraints/fixedvar/presoltiming = 28 # whether to check and enforce bounds on fixed variables # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/fixedvar/enabled = TRUE # whether to act on subSCIPs # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/fixedvar/subscips = TRUE # whether to prefer separation over tightening LP feastol in enforcement # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/fixedvar/prefercut = TRUE # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] constraints/indicator/sepafreq = 10 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/indicator/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/indicator/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/indicator/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/indicator/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 4] constraints/indicator/presoltiming = 4 # enable linear upgrading for constraint handler <indicator> # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/upgrade/indicator = TRUE # priority of conflict handler <indicatorconflict> # [type: int, advanced: TRUE, range: [-2147483648,2147483647], default: 200000] conflict/indicatorconflict/priority = 200000 # Branch on indicator constraints in enforcing? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/branchindicators = FALSE # Generate logicor constraints instead of cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/genlogicor = FALSE # Add coupling constraints or rows if big-M is small enough? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/indicator/addcoupling = TRUE # maximum coefficient for binary variable in coupling constraint # [type: real, advanced: TRUE, range: [0,1000000000], default: 10000] constraints/indicator/maxcouplingvalue = 10000 # Add initial variable upper bound constraints, if 'addcoupling' is true? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/addcouplingcons = FALSE # Should the coupling inequalities be separated dynamically? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/indicator/sepacouplingcuts = TRUE # Allow to use local bounds in order to separate coupling inequalities? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/sepacouplinglocal = FALSE # maximum coefficient for binary variable in separated coupling constraint # [type: real, advanced: TRUE, range: [0,1000000000], default: 10000] constraints/indicator/sepacouplingvalue = 10000 # Separate cuts based on perspective formulation? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/sepaperspective = FALSE # Allow to use local bounds in order to separate perspective cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/indicator/sepapersplocal = TRUE # maximal number of separated non violated IISs, before separation is stopped # [type: int, advanced: FALSE, range: [0,2147483647], default: 3] constraints/indicator/maxsepanonviolated = 3 # Update bounds of original variables for separation? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/updatebounds = FALSE # maximum estimated condition of the solution basis matrix of the alternative LP to be trustworthy (0.0 to disable check) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] constraints/indicator/maxconditionaltlp = 0 # maximal number of cuts separated per separation round # [type: int, advanced: FALSE, range: [0,2147483647], default: 100] constraints/indicator/maxsepacuts = 100 # maximal number of cuts separated per separation round in the root node # [type: int, advanced: FALSE, range: [0,2147483647], default: 2000] constraints/indicator/maxsepacutsroot = 2000 # Remove indicator constraint if corresponding variable bound constraint has been added? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/removeindicators = FALSE # Do not generate indicator constraint, but a bilinear constraint instead? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/generatebilinear = FALSE # Scale slack variable coefficient at construction time? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/scaleslackvar = FALSE # Try to make solutions feasible by setting indicator variables? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/indicator/trysolutions = TRUE # In enforcing try to generate cuts (only if sepaalternativelp is true)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/enforcecuts = FALSE # Should dual reduction steps be performed? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/indicator/dualreductions = TRUE # Add opposite inequality in nodes in which the binary variable has been fixed to 0? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/addopposite = FALSE # Try to upgrade bounddisjunction conflicts by replacing slack variables? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/conflictsupgrade = FALSE # fraction of binary variables that need to be fixed before restart occurs (in forcerestart) # [type: real, advanced: TRUE, range: [0,1], default: 0.9] constraints/indicator/restartfrac = 0.9 # Collect other constraints to alternative LP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/useotherconss = FALSE # Use objective cut with current best solution to alternative LP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/useobjectivecut = FALSE # Try to construct a feasible solution from a cover? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/trysolfromcover = FALSE # Try to upgrade linear constraints to indicator constraints? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/upgradelinear = FALSE # Separate using the alternative LP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/sepaalternativelp = FALSE # Force restart if absolute gap is 1 or enough binary variables have been fixed? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/forcerestart = FALSE # Decompose problem (do not generate linear constraint if all variables are continuous)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/indicator/nolinconscont = FALSE # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/integral/sepafreq = -1 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/integral/propfreq = -1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/integral/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] constraints/integral/eagerfreq = -1 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 0] constraints/integral/maxprerounds = 0 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/integral/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/integral/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 28] constraints/integral/presoltiming = 28 # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] constraints/knapsack/sepafreq = 0 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/knapsack/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/knapsack/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/knapsack/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/knapsack/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/knapsack/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/knapsack/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 28] constraints/knapsack/presoltiming = 28 # enable linear upgrading for constraint handler <knapsack> # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/upgrade/knapsack = TRUE # multiplier on separation frequency, how often knapsack cuts are separated (-1: never, 0: only at root) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 1] constraints/knapsack/sepacardfreq = 1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for separating knapsack cuts # [type: real, advanced: TRUE, range: [0,1], default: 0] constraints/knapsack/maxcardbounddist = 0 # lower clique size limit for greedy clique extraction algorithm (relative to largest clique) # [type: real, advanced: TRUE, range: [0,1], default: 0.5] constraints/knapsack/cliqueextractfactor = 0.5 # maximal number of separation rounds per node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 5] constraints/knapsack/maxrounds = 5 # maximal number of separation rounds per node in the root node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] constraints/knapsack/maxroundsroot = -1 # maximal number of cuts separated per separation round # [type: int, advanced: FALSE, range: [0,2147483647], default: 50] constraints/knapsack/maxsepacuts = 50 # maximal number of cuts separated per separation round in the root node # [type: int, advanced: FALSE, range: [0,2147483647], default: 200] constraints/knapsack/maxsepacutsroot = 200 # should disaggregation of knapsack constraints be allowed in preprocessing? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/knapsack/disaggregation = TRUE # should presolving try to simplify knapsacks # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/knapsack/simplifyinequalities = TRUE # should negated clique information be used in solving process # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/knapsack/negatedclique = TRUE # should pairwise constraint comparison be performed in presolving? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/knapsack/presolpairwise = TRUE # should hash table be used for detecting redundant constraints in advance # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/knapsack/presolusehashing = TRUE # should dual presolving steps be performed? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/knapsack/dualpresolving = TRUE # should GUB information be used for separation? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/knapsack/usegubs = FALSE # should presolving try to detect constraints parallel to the objective function defining an upper bound and prevent these constraints from entering the LP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/knapsack/detectcutoffbound = TRUE # should presolving try to detect constraints parallel to the objective function defining a lower bound and prevent these constraints from entering the LP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/knapsack/detectlowerbound = TRUE # should clique partition information be updated when old partition seems outdated? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/knapsack/updatecliquepartitions = FALSE # factor on the growth of global cliques to decide when to update a previous (negated) clique partition (used only if updatecliquepartitions is set to TRUE) # [type: real, advanced: TRUE, range: [1,10], default: 1.5] constraints/knapsack/clqpartupdatefac = 1.5 # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/linking/sepafreq = 1 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/linking/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/linking/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/linking/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/linking/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/linking/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/linking/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 8] constraints/linking/presoltiming = 8 # this constraint will not propagate or separate, linear and setppc are used? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] constraints/linking/linearize = FALSE # priority of conflict handler <logicor> # [type: int, advanced: TRUE, range: [-2147483648,2147483647], default: 800000] conflict/logicor/priority = 800000 # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] constraints/logicor/sepafreq = 0 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/logicor/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/logicor/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/logicor/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/logicor/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/logicor/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/logicor/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 28] constraints/logicor/presoltiming = 28 # enable linear upgrading for constraint handler <logicor> # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/upgrade/logicor = TRUE # should pairwise constraint comparison be performed in presolving? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/logicor/presolpairwise = TRUE # should hash table be used for detecting redundant constraints in advance # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/logicor/presolusehashing = TRUE # should dual presolving steps be performed? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/logicor/dualpresolving = TRUE # should negated clique information be used in presolving # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/logicor/negatedclique = TRUE # should implications/cliques be used in presolving # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/logicor/implications = TRUE # should pairwise constraint comparison try to strengthen constraints by removing superflous non-zeros? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/logicor/strengthen = TRUE # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] constraints/or/sepafreq = 0 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/or/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/or/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/or/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/or/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/or/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/or/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 8] constraints/or/presoltiming = 8 # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 5] constraints/orbisack/sepafreq = 5 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 5] constraints/orbisack/propfreq = 5 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/orbisack/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] constraints/orbisack/eagerfreq = -1 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/orbisack/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/orbisack/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/orbisack/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 16] constraints/orbisack/presoltiming = 16 # Separate cover inequalities for orbisacks? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/orbisack/coverseparation = TRUE # Separate orbisack inequalities? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/orbisack/orbiSeparation = FALSE # Maximum size of coefficients for orbisack inequalities # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 1000000] constraints/orbisack/coeffbound = 1000000 # Upgrade orbisack constraints to packing/partioning orbisacks? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/orbisack/checkpporbisack = TRUE # Whether orbisack constraints should be forced to be copied to sub SCIPs. # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/orbisack/forceconscopy = FALSE # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/orbitope/sepafreq = -1 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/orbitope/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/orbitope/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] constraints/orbitope/eagerfreq = -1 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/orbitope/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/orbitope/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/orbitope/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 8] constraints/orbitope/presoltiming = 8 # Strengthen orbitope constraints to packing/partioning orbitopes? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/orbitope/checkpporbitope = TRUE # Whether we separate inequalities for full orbitopes? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/orbitope/sepafullorbitope = FALSE # Whether orbitope constraints should be forced to be copied to sub SCIPs. # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/orbitope/forceconscopy = FALSE # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/pseudoboolean/sepafreq = -1 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/pseudoboolean/propfreq = -1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/pseudoboolean/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/pseudoboolean/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/pseudoboolean/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/pseudoboolean/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/pseudoboolean/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 8] constraints/pseudoboolean/presoltiming = 8 # decompose all normal pseudo boolean constraint into a "linear" constraint and "and" constraints # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/pseudoboolean/decomposenormal = FALSE # decompose all indicator pseudo boolean constraint into a "linear" constraint and "and" constraints # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/pseudoboolean/decomposeindicator = TRUE # should the nonlinear constraints be separated during LP processing? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/pseudoboolean/nlcseparate = TRUE # should the nonlinear constraints be propagated during node processing? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/pseudoboolean/nlcpropagate = TRUE # should the nonlinear constraints be removable? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/pseudoboolean/nlcremovable = TRUE # priority of conflict handler <setppc> # [type: int, advanced: TRUE, range: [-2147483648,2147483647], default: 700000] conflict/setppc/priority = 700000 # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] constraints/setppc/sepafreq = 0 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/setppc/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/setppc/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/setppc/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/setppc/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/setppc/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/setppc/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 28] constraints/setppc/presoltiming = 28 # enable linear upgrading for constraint handler <setppc> # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/upgrade/setppc = TRUE # enable nonlinear upgrading for constraint handler <setppc> # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/nonlinear/upgrade/setppc = TRUE # number of children created in pseudo branching (0: disable pseudo branching) # [type: int, advanced: TRUE, range: [0,2147483647], default: 2] constraints/setppc/npseudobranches = 2 # should pairwise constraint comparison be performed in presolving? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/setppc/presolpairwise = TRUE # should hash table be used for detecting redundant constraints in advance # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/setppc/presolusehashing = TRUE # should dual presolving steps be performed? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/setppc/dualpresolving = TRUE # should we try to lift variables into other clique constraints, fix variables, aggregate them, and also shrink the amount of variables in clique constraints # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/setppc/cliquelifting = FALSE # should we try to generate extra cliques out of all binary variables to maybe fasten redundant constraint detection # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/setppc/addvariablesascliques = FALSE # should we try to shrink the number of variables in a clique constraints, by replacing more than one variable by only one # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/setppc/cliqueshrinking = TRUE # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] constraints/SOS1/sepafreq = 10 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/SOS1/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/SOS1/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/SOS1/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/SOS1/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/SOS1/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/SOS1/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 8] constraints/SOS1/presoltiming = 8 # do not create an adjacency matrix if number of SOS1 variables is larger than predefined value (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 10000] constraints/SOS1/maxsosadjacency = 10000 # maximal number of extensions that will be computed for each SOS1 constraint (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 1] constraints/SOS1/maxextensions = 1 # maximal number of bound tightening rounds per presolving round (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 5] constraints/SOS1/maxtightenbds = 5 # if TRUE then perform implication graph analysis (might add additional SOS1 constraints) # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/SOS1/perfimplanalysis = FALSE # number of recursive calls of implication graph analysis (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/SOS1/depthimplanalysis = -1 # whether to use conflict graph propagation # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/SOS1/conflictprop = TRUE # whether to use implication graph propagation # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/SOS1/implprop = TRUE # whether to use SOS1 constraint propagation # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/SOS1/sosconsprop = FALSE # which branching rule should be applied ? ('n': neighborhood, 'b': bipartite, 's': SOS1/clique) (note: in some cases an automatic switching to SOS1 branching is possible) # [type: char, advanced: TRUE, range: {nbs}, default: n] constraints/SOS1/branchingrule = n # if TRUE then automatically switch to SOS1 branching if the SOS1 constraints do not overlap # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/SOS1/autosos1branch = TRUE # if neighborhood branching is used, then fix the branching variable (if positive in sign) to the value of the feasibility tolerance # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/SOS1/fixnonzero = FALSE # if TRUE then add complementarity constraints to the branching nodes (can be used in combination with neighborhood or bipartite branching) # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/SOS1/addcomps = FALSE # maximal number of complementarity constraints added per branching node (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/SOS1/maxaddcomps = -1 # minimal feasibility value for complementarity constraints in order to be added to the branching node # [type: real, advanced: TRUE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: -0.6] constraints/SOS1/addcompsfeas = -0.6 # minimal feasibility value for bound inequalities in order to be added to the branching node # [type: real, advanced: TRUE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 1] constraints/SOS1/addbdsfeas = 1 # should added complementarity constraints be extended to SOS1 constraints to get tighter bound inequalities # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/SOS1/addextendedbds = TRUE # Use SOS1 branching in enforcing (otherwise leave decision to branching rules)? This value can only be set to false if all SOS1 variables are binary # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/SOS1/branchsos = TRUE # Branch on SOS constraint with most number of nonzeros? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] constraints/SOS1/branchnonzeros = FALSE # Branch on SOS cons. with highest nonzero-variable weight for branching (needs branchnonzeros = false)? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] constraints/SOS1/branchweight = FALSE # only add complementarity constraints to branching nodes for predefined depth (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 30] constraints/SOS1/addcompsdepth = 30 # maximal number of strong branching rounds to perform for each node (-1: auto); only available for neighborhood and bipartite branching # [type: int, advanced: TRUE, range: [-1,2147483647], default: 0] constraints/SOS1/nstrongrounds = 0 # maximal number LP iterations to perform for each strong branching round (-2: auto, -1: no limit) # [type: int, advanced: TRUE, range: [-2,2147483647], default: 10000] constraints/SOS1/nstrongiter = 10000 # if TRUE separate bound inequalities from initial SOS1 constraints # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/SOS1/boundcutsfromsos1 = FALSE # if TRUE separate bound inequalities from the conflict graph # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/SOS1/boundcutsfromgraph = TRUE # if TRUE then automatically switch to separating initial SOS1 constraints if the SOS1 constraints do not overlap # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/SOS1/autocutsfromsos1 = TRUE # frequency for separating bound cuts; zero means to separate only in the root node # [type: int, advanced: TRUE, range: [-1,1073741822], default: 10] constraints/SOS1/boundcutsfreq = 10 # node depth of separating bound cuts (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 40] constraints/SOS1/boundcutsdepth = 40 # maximal number of bound cuts separated per branching node # [type: int, advanced: TRUE, range: [0,2147483647], default: 50] constraints/SOS1/maxboundcuts = 50 # maximal number of bound cuts separated per iteration in the root node # [type: int, advanced: TRUE, range: [0,2147483647], default: 150] constraints/SOS1/maxboundcutsroot = 150 # if TRUE then bound cuts are strengthened in case bound variables are available # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/SOS1/strthenboundcuts = TRUE # frequency for separating implied bound cuts; zero means to separate only in the root node # [type: int, advanced: TRUE, range: [-1,1073741822], default: 0] constraints/SOS1/implcutsfreq = 0 # node depth of separating implied bound cuts (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 40] constraints/SOS1/implcutsdepth = 40 # maximal number of implied bound cuts separated per branching node # [type: int, advanced: TRUE, range: [0,2147483647], default: 50] constraints/SOS1/maximplcuts = 50 # maximal number of implied bound cuts separated per iteration in the root node # [type: int, advanced: TRUE, range: [0,2147483647], default: 150] constraints/SOS1/maximplcutsroot = 150 # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] constraints/SOS2/sepafreq = 0 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/SOS2/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/SOS2/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/SOS2/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/SOS2/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/SOS2/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/SOS2/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 4] constraints/SOS2/presoltiming = 4 # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/superindicator/sepafreq = -1 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/superindicator/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/superindicator/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/superindicator/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/superindicator/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/superindicator/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/superindicator/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 8] constraints/superindicator/presoltiming = 8 # should type of slack constraint be checked when creating superindicator constraint? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/superindicator/checkslacktype = TRUE # maximum big-M coefficient of binary variable in upgrade to a linear constraint (relative to smallest coefficient) # [type: real, advanced: TRUE, range: [0,1e+15], default: 10000] constraints/superindicator/maxupgdcoeflinear = 10000 # priority for upgrading to an indicator constraint (-1: never) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 1] constraints/superindicator/upgdprioindicator = 1 # priority for upgrading to an indicator constraint (-1: never) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 2] constraints/superindicator/upgdpriolinear = 2 # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 5] constraints/symresack/sepafreq = 5 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 5] constraints/symresack/propfreq = 5 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/symresack/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] constraints/symresack/eagerfreq = -1 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/symresack/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/symresack/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/symresack/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 16] constraints/symresack/presoltiming = 16 # Upgrade symresack constraints to packing/partioning symresacks? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/symresack/ppsymresack = TRUE # Check whether permutation is monotone when upgrading to packing/partioning symresacks? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/symresack/checkmonotonicity = TRUE # Whether symresack constraints should be forced to be copied to sub SCIPs. # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/symresack/forceconscopy = FALSE # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] constraints/varbound/sepafreq = 0 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/varbound/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/varbound/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/varbound/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/varbound/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/varbound/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/varbound/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 12] constraints/varbound/presoltiming = 12 # enable linear upgrading for constraint handler <varbound> # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/upgrade/varbound = TRUE # should pairwise constraint comparison be performed in presolving? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/varbound/presolpairwise = TRUE # maximum coefficient in varbound constraint to be added as a row into LP # [type: real, advanced: TRUE, range: [0,1e+20], default: 1000000000] constraints/varbound/maxlpcoef = 1000000000 # should bound widening be used in conflict analysis? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/varbound/usebdwidening = TRUE # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] constraints/xor/sepafreq = 0 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/xor/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/xor/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 100] constraints/xor/eagerfreq = 100 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/xor/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/xor/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/xor/delayprop = FALSE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 28] constraints/xor/presoltiming = 28 # enable linear upgrading for constraint handler <xor> # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] constraints/linear/upgrade/xor = TRUE # should pairwise constraint comparison be performed in presolving? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/xor/presolpairwise = TRUE # should hash table be used for detecting redundant constraints in advance? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/xor/presolusehashing = TRUE # should the extended formulation be added in presolving? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/xor/addextendedform = FALSE # should the extended flow formulation be added (nonsymmetric formulation otherwise)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/xor/addflowextended = FALSE # should parity inequalities be separated? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/xor/separateparity = FALSE # frequency for applying the Gauss propagator # [type: int, advanced: TRUE, range: [-1,1073741822], default: 5] constraints/xor/gausspropfreq = 5 # frequency for separating cuts (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] constraints/components/sepafreq = -1 # frequency for propagating domains (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] constraints/components/propfreq = 1 # timing when constraint propagation should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS) # [type: int, advanced: TRUE, range: [1,15], default: 1] constraints/components/proptiming = 1 # frequency for using all instead of only the useful constraints in separation, propagation and enforcement (-1: never, 0: only in first evaluation) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] constraints/components/eagerfreq = -1 # maximal number of presolving rounds the constraint handler participates in (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] constraints/components/maxprerounds = -1 # should separation method be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] constraints/components/delaysepa = FALSE # should propagation method be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] constraints/components/delayprop = TRUE # timing mask of the constraint handler's presolving method (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 32] constraints/components/presoltiming = 32 # maximum depth of a node to run components detection (-1: disable component detection during solving) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] constraints/components/maxdepth = -1 # maximum number of integer (or binary) variables to solve a subproblem during presolving (-1: unlimited) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 500] constraints/components/maxintvars = 500 # minimum absolute size (in terms of variables) to solve a component individually during branch-and-bound # [type: int, advanced: TRUE, range: [0,2147483647], default: 50] constraints/components/minsize = 50 # minimum relative size (in terms of variables) to solve a component individually during branch-and-bound # [type: real, advanced: TRUE, range: [0,1], default: 0.1] constraints/components/minrelsize = 0.1 # maximum number of nodes to be solved in subproblems during presolving # [type: longint, advanced: FALSE, range: [-1,9223372036854775807], default: 10000] constraints/components/nodelimit = 10000 # the weight of an integer variable compared to binary variables # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 1] constraints/components/intfactor = 1 # factor to increase the feasibility tolerance of the main SCIP in all sub-SCIPs, default value 1.0 # [type: real, advanced: TRUE, range: [0,1000000], default: 1] constraints/components/feastolfactor = 1 # should possible "and" constraint be linearized when writing the mps file? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] reading/mpsreader/linearize-and-constraints = TRUE # should an aggregated linearization for and constraints be used? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] reading/mpsreader/aggrlinearization-ands = TRUE # should possible "and" constraint be linearized when writing the lp file? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] reading/lpreader/linearize-and-constraints = TRUE # should an aggregated linearization for and constraints be used? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] reading/lpreader/aggrlinearization-ands = TRUE # should the current directory be changed to that of the ZIMPL file before parsing? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] reading/zplreader/changedir = TRUE # should ZIMPL starting solutions be forwarded to SCIP? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] reading/zplreader/usestartsol = TRUE # additional parameter string passed to the ZIMPL parser (or - for no additional parameters) # [type: string, advanced: FALSE, default: "-"] reading/zplreader/parameters = "-" # shall characters '#', '*', '+', '/', and '-' in variable and constraint names be replaced by '_'? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] reading/gmsreader/replaceforbiddenchars = FALSE # default M value for big-M reformulation of indicator constraints in case no bound on slack variable is given # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 1000000] reading/gmsreader/bigmdefault = 1000000 # which reformulation to use for indicator constraints: 'b'ig-M, 's'os1 # [type: char, advanced: FALSE, range: {bs}, default: s] reading/gmsreader/indicatorreform = s # is it allowed to use the gams function signpower(x,a)? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] reading/gmsreader/signpower = FALSE # should model constraints be subject to aging? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] reading/opbreader/dynamicconss = FALSE # use '*' between coefficients and variables by writing to problem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] reading/opbreader/multisymbol = FALSE # should an artificial objective, depending on the number of clauses a variable appears in, be used? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] reading/cnfreader/useobj = FALSE # should fixed and aggregated variables be printed (if not, re-parsing might fail) # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] reading/cipreader/writefixedvars = TRUE # should Benders' decomposition be used? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] reading/storeader/usebenders = FALSE # only use improving bounds # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] reading/bndreader/improveonly = FALSE # should the coloring values be relativ or absolute # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] reading/ppmreader/rgbrelativ = TRUE # should the output format be binary(P6) (otherwise plain(P3) format) # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] reading/ppmreader/rgbascii = TRUE # splitting coefficients in this number of intervals # [type: int, advanced: FALSE, range: [3,16], default: 3] reading/ppmreader/coefficientlimit = 3 # maximal color value # [type: int, advanced: FALSE, range: [0,255], default: 160] reading/ppmreader/rgblimit = 160 # should the output format be binary(P4) (otherwise plain(P1) format) # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] reading/pbmreader/binary = TRUE # maximum number of rows in the scaled picture (-1 for no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1000] reading/pbmreader/maxrows = 1000 # maximum number of columns in the scaled picture (-1 for no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1000] reading/pbmreader/maxcols = 1000 # priority of presolver <boundshift> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 7900000] presolving/boundshift/priority = 7900000 # maximal number of presolving rounds the presolver participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 0] presolving/boundshift/maxrounds = 0 # timing mask of presolver <boundshift> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 4] presolving/boundshift/timing = 4 # absolute value of maximum shift # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 9223372036854775807] presolving/boundshift/maxshift = 9223372036854775807 # is flipping allowed (multiplying with -1)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] presolving/boundshift/flipping = TRUE # shift only integer ranges? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] presolving/boundshift/integer = TRUE # priority of presolver <convertinttobin> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 6000000] presolving/convertinttobin/priority = 6000000 # maximal number of presolving rounds the presolver participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 0] presolving/convertinttobin/maxrounds = 0 # timing mask of presolver <convertinttobin> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 4] presolving/convertinttobin/timing = 4 # absolute value of maximum domain size for converting an integer variable to binaries variables # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 9223372036854775807] presolving/convertinttobin/maxdomainsize = 9223372036854775807 # should only integer variables with a domain size of 2^p - 1 be converted(, there we don't need an knapsack-constraint for restricting the sum of the binaries) # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] presolving/convertinttobin/onlypoweroftwo = FALSE # should only integer variables with uplocks equals downlocks be converted # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] presolving/convertinttobin/samelocksinbothdirections = FALSE # priority of presolver <domcol> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1000] presolving/domcol/priority = -1000 # maximal number of presolving rounds the presolver participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] presolving/domcol/maxrounds = -1 # timing mask of presolver <domcol> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 16] presolving/domcol/timing = 16 # minimal number of pair comparisons # [type: int, advanced: FALSE, range: [100,1048576], default: 1024] presolving/domcol/numminpairs = 1024 # maximal number of pair comparisons # [type: int, advanced: FALSE, range: [1024,1000000000], default: 1048576] presolving/domcol/nummaxpairs = 1048576 # should predictive bound strengthening be applied? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] presolving/domcol/predbndstr = FALSE # should reductions for continuous variables be performed? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] presolving/domcol/continuousred = TRUE # priority of presolver <dualagg> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -12000] presolving/dualagg/priority = -12000 # maximal number of presolving rounds the presolver participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 0] presolving/dualagg/maxrounds = 0 # timing mask of presolver <dualagg> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 16] presolving/dualagg/timing = 16 # priority of presolver <dualcomp> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -50] presolving/dualcomp/priority = -50 # maximal number of presolving rounds the presolver participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] presolving/dualcomp/maxrounds = -1 # timing mask of presolver <dualcomp> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 16] presolving/dualcomp/timing = 16 # should only discrete variables be compensated? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] presolving/dualcomp/componlydisvars = FALSE # priority of presolver <dualinfer> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -3000] presolving/dualinfer/priority = -3000 # maximal number of presolving rounds the presolver participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 0] presolving/dualinfer/maxrounds = 0 # timing mask of presolver <dualinfer> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 16] presolving/dualinfer/timing = 16 # use convex combination of columns for determining dual bounds # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] presolving/dualinfer/twocolcombine = TRUE # maximal number of dual bound strengthening loops # [type: int, advanced: FALSE, range: [-1,2147483647], default: 12] presolving/dualinfer/maxdualbndloops = 12 # maximal number of considered non-zeros within one column (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 100] presolving/dualinfer/maxconsiderednonzeros = 100 # maximal number of consecutive useless hashtable retrieves # [type: int, advanced: TRUE, range: [-1,2147483647], default: 1000] presolving/dualinfer/maxretrievefails = 1000 # maximal number of consecutive useless column combines # [type: int, advanced: TRUE, range: [-1,2147483647], default: 1000] presolving/dualinfer/maxcombinefails = 1000 # Maximum number of hashlist entries as multiple of number of columns in the problem (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 10] presolving/dualinfer/maxhashfac = 10 # Maximum number of processed column pairs as multiple of the number of columns in the problem (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 1] presolving/dualinfer/maxpairfac = 1 # Maximum number of row's non-zeros for changing inequality to equality # [type: int, advanced: FALSE, range: [2,2147483647], default: 3] presolving/dualinfer/maxrowsupport = 3 # priority of presolver <gateextraction> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 1000000] presolving/gateextraction/priority = 1000000 # maximal number of presolving rounds the presolver participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] presolving/gateextraction/maxrounds = -1 # timing mask of presolver <gateextraction> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 16] presolving/gateextraction/timing = 16 # should we only try to extract set-partitioning constraints and no and-constraints # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] presolving/gateextraction/onlysetpart = FALSE # should we try to extract set-partitioning constraint out of one logicor and one corresponding set-packing constraint # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] presolving/gateextraction/searchequations = TRUE # order logicor contraints to extract big-gates before smaller ones (-1), do not order them (0) or order them to extract smaller gates at first (1) # [type: int, advanced: TRUE, range: [-1,1], default: 1] presolving/gateextraction/sorting = 1 # priority of presolver <implics> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -10000] presolving/implics/priority = -10000 # maximal number of presolving rounds the presolver participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] presolving/implics/maxrounds = -1 # timing mask of presolver <implics> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 8] presolving/implics/timing = 8 # priority of presolver <inttobinary> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 7000000] presolving/inttobinary/priority = 7000000 # maximal number of presolving rounds the presolver participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] presolving/inttobinary/maxrounds = -1 # timing mask of presolver <inttobinary> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 4] presolving/inttobinary/timing = 4 # priority of presolver <milp> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 9999999] presolving/milp/priority = 9999999 # maximal number of presolving rounds the presolver participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] presolving/milp/maxrounds = -1 # timing mask of presolver <milp> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 8] presolving/milp/timing = 8 # maximum number of threads presolving may use (0: automatic) # [type: int, advanced: FALSE, range: [0,2147483647], default: 1] presolving/milp/threads = 1 # maximal possible fillin for substitutions to be considered # [type: int, advanced: FALSE, range: [-2147483648,2147483647], default: 3] presolving/milp/maxfillinpersubstitution = 3 # maximal amount of nonzeros allowed to be shifted to make space for substitutions # [type: int, advanced: TRUE, range: [0,2147483647], default: 10] presolving/milp/maxshiftperrow = 10 # the random seed used for randomization of tie breaking # [type: int, advanced: FALSE, range: [-2147483648,2147483647], default: 0] presolving/milp/randomseed = 0 # should linear dependent equations and free columns be removed? (0: never, 1: for LPs, 2: always) # [type: int, advanced: TRUE, range: [0,2], default: 0] presolving/milp/detectlineardependency = 0 # modify SCIP constraints when the number of nonzeros or rows is at most this factor times the number of nonzeros or rows before presolving # [type: real, advanced: FALSE, range: [0,1], default: 0.8] presolving/milp/modifyconsfac = 0.8 # the markowitz tolerance used for substitutions # [type: real, advanced: FALSE, range: [0,1], default: 0.01] presolving/milp/markowitztolerance = 0.01 # absolute bound value that is considered too huge for activitity based calculations # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 100000000] presolving/milp/hugebound = 100000000 # maximal badge size in Probing in PaPILO if PaPILO is executed in sequential mode # [type: int, advanced: FALSE, range: [-1,2147483647], default: 15000] presolving/milp/maxbadgesizeseq = 15000 # maximal badge size in Probing in PaPILO if PaPILO is executed in parallel mode # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] presolving/milp/maxbadgesizepar = -1 # internal maxrounds for each milp presolving (-1: no limit, 0: model cleanup) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] presolving/milp/internalmaxrounds = -1 # should the parallel rows presolver be enabled within the presolve library? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] presolving/milp/enableparallelrows = TRUE # should the dominated column presolver be enabled within the presolve library? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] presolving/milp/enabledomcol = TRUE # should the dualinfer presolver be enabled within the presolve library? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] presolving/milp/enabledualinfer = TRUE # should the multi-aggregation presolver be enabled within the presolve library? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] presolving/milp/enablemultiaggr = TRUE # should the probing presolver be enabled within the presolve library? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] presolving/milp/enableprobing = TRUE # should the sparsify presolver be enabled within the presolve library? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] presolving/milp/enablesparsify = FALSE # filename to store the problem before MILP presolving starts (only enforced constraints) # [type: string, advanced: TRUE, default: "-"] presolving/milp/probfilename = "-" # verbosity level of PaPILO (0: quiet, 1: errors, 2: warnings, 3: normal, 4: detailed) # [type: int, advanced: FALSE, range: [0,4], default: 0] presolving/milp/verbosity = 0 # priority of presolver <qpkktref> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1] presolving/qpkktref/priority = -1 # maximal number of presolving rounds the presolver participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 0] presolving/qpkktref/maxrounds = 0 # timing mask of presolver <qpkktref> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 8] presolving/qpkktref/timing = 8 # if TRUE then allow binary variables for KKT update # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] presolving/qpkktref/addkktbinary = FALSE # if TRUE then only apply the update to QPs with bounded variables; if the variables are not bounded then a finite optimal solution might not exist and the KKT conditions would then be invalid # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] presolving/qpkktref/updatequadbounded = TRUE # if TRUE then apply quadratic constraint update even if the quadratic constraint matrix is known to be indefinite # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] presolving/qpkktref/updatequadindef = FALSE # priority of presolver <redvub> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -9000000] presolving/redvub/priority = -9000000 # maximal number of presolving rounds the presolver participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 0] presolving/redvub/maxrounds = 0 # timing mask of presolver <redvub> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 16] presolving/redvub/timing = 16 # priority of presolver <trivial> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 9000000] presolving/trivial/priority = 9000000 # maximal number of presolving rounds the presolver participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] presolving/trivial/maxrounds = -1 # timing mask of presolver <trivial> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 4] presolving/trivial/timing = 4 # priority of presolver <tworowbnd> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -2000] presolving/tworowbnd/priority = -2000 # maximal number of presolving rounds the presolver participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 0] presolving/tworowbnd/maxrounds = 0 # timing mask of presolver <tworowbnd> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 16] presolving/tworowbnd/timing = 16 # should tworowbnd presolver be copied to sub-SCIPs? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] presolving/tworowbnd/enablecopy = TRUE # maximal number of considered non-zeros within one row (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 100] presolving/tworowbnd/maxconsiderednonzeros = 100 # maximal number of consecutive useless hashtable retrieves # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1000] presolving/tworowbnd/maxretrievefails = 1000 # maximal number of consecutive useless row combines # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1000] presolving/tworowbnd/maxcombinefails = 1000 # Maximum number of hashlist entries as multiple of number of rows in the problem (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 10] presolving/tworowbnd/maxhashfac = 10 # Maximum number of processed row pairs as multiple of the number of rows in the problem (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1] presolving/tworowbnd/maxpairfac = 1 # priority of presolver <sparsify> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -24000] presolving/sparsify/priority = -24000 # maximal number of presolving rounds the presolver participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] presolving/sparsify/maxrounds = -1 # timing mask of presolver <sparsify> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 16] presolving/sparsify/timing = 16 # should sparsify presolver be copied to sub-SCIPs? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] presolving/sparsify/enablecopy = TRUE # should we cancel nonzeros in constraints of the linear constraint handler? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] presolving/sparsify/cancellinear = TRUE # should we forbid cancellations that destroy integer coefficients? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] presolving/sparsify/preserveintcoefs = TRUE # maximal fillin for continuous variables (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 0] presolving/sparsify/maxcontfillin = 0 # maximal fillin for binary variables (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 0] presolving/sparsify/maxbinfillin = 0 # maximal fillin for integer variables including binaries (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 0] presolving/sparsify/maxintfillin = 0 # maximal support of one equality to be used for cancelling (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] presolving/sparsify/maxnonzeros = -1 # maximal number of considered non-zeros within one row (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 70] presolving/sparsify/maxconsiderednonzeros = 70 # order in which to process inequalities ('n'o sorting, 'i'ncreasing nonzeros, 'd'ecreasing nonzeros) # [type: char, advanced: TRUE, range: {nid}, default: d] presolving/sparsify/rowsort = d # limit on the number of useless vs. useful hashtable retrieves as a multiple of the number of constraints # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 100] presolving/sparsify/maxretrievefac = 100 # number of calls to wait until next execution as a multiple of the number of useless calls # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 2] presolving/sparsify/waitingfac = 2 # priority of presolver <dualsparsify> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -240000] presolving/dualsparsify/priority = -240000 # maximal number of presolving rounds the presolver participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] presolving/dualsparsify/maxrounds = -1 # timing mask of presolver <dualsparsify> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 16] presolving/dualsparsify/timing = 16 # should dualsparsify presolver be copied to sub-SCIPs? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] presolving/dualsparsify/enablecopy = TRUE # should we forbid cancellations that destroy integer coefficients? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] presolving/dualsparsify/preserveintcoefs = FALSE # should we preserve good locked properties of variables (at most one lock in one direction)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] presolving/dualsparsify/preservegoodlocks = FALSE # maximal fillin for continuous variables (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1] presolving/dualsparsify/maxcontfillin = 1 # maximal fillin for binary variables (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1] presolving/dualsparsify/maxbinfillin = 1 # maximal fillin for integer variables including binaries (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1] presolving/dualsparsify/maxintfillin = 1 # maximal number of considered nonzeros within one column (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 70] presolving/dualsparsify/maxconsiderednonzeros = 70 # minimal eliminated nonzeros within one column if we need to add a constraint to the problem # [type: int, advanced: FALSE, range: [0,2147483647], default: 100] presolving/dualsparsify/mineliminatednonzeros = 100 # limit on the number of useless vs. useful hashtable retrieves as a multiple of the number of constraints # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 100] presolving/dualsparsify/maxretrievefac = 100 # number of calls to wait until next execution as a multiple of the number of useless calls # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 2] presolving/dualsparsify/waitingfac = 2 # priority of presolver <stuffing> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -100] presolving/stuffing/priority = -100 # maximal number of presolving rounds the presolver participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 0] presolving/stuffing/maxrounds = 0 # timing mask of presolver <stuffing> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [4,60], default: 16] presolving/stuffing/timing = 16 # priority of node selection rule <bfs> in standard mode # [type: int, advanced: FALSE, range: [-536870912,1073741823], default: 100000] nodeselection/bfs/stdpriority = 100000 # priority of node selection rule <bfs> in memory saving mode # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 0] nodeselection/bfs/memsavepriority = 0 # minimal plunging depth, before new best node may be selected (-1 for dynamic setting) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] nodeselection/bfs/minplungedepth = -1 # maximal plunging depth, before new best node is forced to be selected (-1 for dynamic setting) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] nodeselection/bfs/maxplungedepth = -1 # maximal quotient (curlowerbound - lowerbound)/(cutoffbound - lowerbound) where plunging is performed # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.25] nodeselection/bfs/maxplungequot = 0.25 # priority of node selection rule <breadthfirst> in standard mode # [type: int, advanced: FALSE, range: [-536870912,1073741823], default: -10000] nodeselection/breadthfirst/stdpriority = -10000 # priority of node selection rule <breadthfirst> in memory saving mode # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1000000] nodeselection/breadthfirst/memsavepriority = -1000000 # priority of node selection rule <dfs> in standard mode # [type: int, advanced: FALSE, range: [-536870912,1073741823], default: 0] nodeselection/dfs/stdpriority = 0 # priority of node selection rule <dfs> in memory saving mode # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 100000] nodeselection/dfs/memsavepriority = 100000 # priority of node selection rule <estimate> in standard mode # [type: int, advanced: FALSE, range: [-536870912,1073741823], default: 200000] nodeselection/estimate/stdpriority = 200000 # priority of node selection rule <estimate> in memory saving mode # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 100] nodeselection/estimate/memsavepriority = 100 # minimal plunging depth, before new best node may be selected (-1 for dynamic setting) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] nodeselection/estimate/minplungedepth = -1 # maximal plunging depth, before new best node is forced to be selected (-1 for dynamic setting) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] nodeselection/estimate/maxplungedepth = -1 # maximal quotient (estimate - lowerbound)/(cutoffbound - lowerbound) where plunging is performed # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.25] nodeselection/estimate/maxplungequot = 0.25 # frequency at which the best node instead of the best estimate is selected (0: never) # [type: int, advanced: FALSE, range: [0,2147483647], default: 10] nodeselection/estimate/bestnodefreq = 10 # depth until breadth-first search is applied # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] nodeselection/estimate/breadthfirstdepth = -1 # number of nodes before doing plunging the first time # [type: int, advanced: FALSE, range: [0,2147483647], default: 0] nodeselection/estimate/plungeoffset = 0 # priority of node selection rule <hybridestim> in standard mode # [type: int, advanced: FALSE, range: [-536870912,1073741823], default: 50000] nodeselection/hybridestim/stdpriority = 50000 # priority of node selection rule <hybridestim> in memory saving mode # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 50] nodeselection/hybridestim/memsavepriority = 50 # minimal plunging depth, before new best node may be selected (-1 for dynamic setting) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] nodeselection/hybridestim/minplungedepth = -1 # maximal plunging depth, before new best node is forced to be selected (-1 for dynamic setting) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] nodeselection/hybridestim/maxplungedepth = -1 # maximal quotient (estimate - lowerbound)/(cutoffbound - lowerbound) where plunging is performed # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.25] nodeselection/hybridestim/maxplungequot = 0.25 # frequency at which the best node instead of the hybrid best estimate / best bound is selected (0: never) # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] nodeselection/hybridestim/bestnodefreq = 1000 # weight of estimate value in node selection score (0: pure best bound search, 1: pure best estimate search) # [type: real, advanced: TRUE, range: [0,1], default: 0.1] nodeselection/hybridestim/estimweight = 0.1 # priority of node selection rule <restartdfs> in standard mode # [type: int, advanced: FALSE, range: [-536870912,1073741823], default: 10000] nodeselection/restartdfs/stdpriority = 10000 # priority of node selection rule <restartdfs> in memory saving mode # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 50000] nodeselection/restartdfs/memsavepriority = 50000 # frequency for selecting the best node instead of the deepest one # [type: int, advanced: FALSE, range: [0,2147483647], default: 100] nodeselection/restartdfs/selectbestfreq = 100 # count only leaf nodes (otherwise all nodes)? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nodeselection/restartdfs/countonlyleaves = TRUE # priority of node selection rule <uct> in standard mode # [type: int, advanced: FALSE, range: [-536870912,1073741823], default: 10] nodeselection/uct/stdpriority = 10 # priority of node selection rule <uct> in memory saving mode # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 0] nodeselection/uct/memsavepriority = 0 # maximum number of nodes before switching to default rule # [type: int, advanced: TRUE, range: [0,1000000], default: 31] nodeselection/uct/nodelimit = 31 # weight for visit quotient of node selection rule # [type: real, advanced: TRUE, range: [0,1], default: 0.1] nodeselection/uct/weight = 0.1 # should the estimate (TRUE) or lower bound of a node be used for UCT score? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] nodeselection/uct/useestimate = FALSE # priority of branching rule <allfullstrong> # [type: int, advanced: FALSE, range: [-536870912,536870911], default: -1000] branching/allfullstrong/priority = -1000 # maximal depth level, up to which branching rule <allfullstrong> should be used (-1 for no limit) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] branching/allfullstrong/maxdepth = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying branching rule (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: FALSE, range: [0,1], default: 1] branching/allfullstrong/maxbounddist = 1 # priority of branching rule <cloud> # [type: int, advanced: FALSE, range: [-536870912,536870911], default: 0] branching/cloud/priority = 0 # maximal depth level, up to which branching rule <cloud> should be used (-1 for no limit) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] branching/cloud/maxdepth = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying branching rule (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: FALSE, range: [0,1], default: 1] branching/cloud/maxbounddist = 1 # should a cloud of points be used? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] branching/cloud/usecloud = TRUE # should only F2 be used? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] branching/cloud/onlyF2 = FALSE # should the union of candidates be used? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] branching/cloud/useunion = FALSE # maximum number of points for the cloud (-1 means no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] branching/cloud/maxpoints = -1 # minimum success rate for the cloud # [type: real, advanced: FALSE, range: [0,1], default: 0] branching/cloud/minsuccessrate = 0 # minimum success rate for the union # [type: real, advanced: FALSE, range: [0,1], default: 0] branching/cloud/minsuccessunion = 0 # maximum depth for the union # [type: int, advanced: FALSE, range: [0,65000], default: 65000] branching/cloud/maxdepthunion = 65000 # priority of branching rule <distribution> # [type: int, advanced: FALSE, range: [-536870912,536870911], default: 0] branching/distribution/priority = 0 # maximal depth level, up to which branching rule <distribution> should be used (-1 for no limit) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] branching/distribution/maxdepth = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying branching rule (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: FALSE, range: [0,1], default: 1] branching/distribution/maxbounddist = 1 # the score;largest 'd'ifference, 'l'owest cumulative probability,'h'ighest c.p., 'v'otes lowest c.p., votes highest c.p.('w') # [type: char, advanced: TRUE, range: {dhlvw}, default: v] branching/distribution/scoreparam = v # should only rows which are active at the current node be considered? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/distribution/onlyactiverows = FALSE # should the branching score weigh up- and down-scores of a variable # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/distribution/weightedscore = FALSE # priority of branching rule <fullstrong> # [type: int, advanced: FALSE, range: [-536870912,536870911], default: 0] branching/fullstrong/priority = 0 # maximal depth level, up to which branching rule <fullstrong> should be used (-1 for no limit) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] branching/fullstrong/maxdepth = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying branching rule (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: FALSE, range: [0,1], default: 1] branching/fullstrong/maxbounddist = 1 # number of intermediate LPs solved to trigger reevaluation of strong branching value for a variable that was already evaluated at the current node # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 10] branching/fullstrong/reevalage = 10 # maximum number of propagation rounds to be performed during strong branching before solving the LP (-1: no limit, -2: parameter settings) # [type: int, advanced: TRUE, range: [-3,2147483647], default: -2] branching/fullstrong/maxproprounds = -2 # should valid bounds be identified in a probing-like fashion during strong branching (only with propagation)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] branching/fullstrong/probingbounds = TRUE # should strong branching be applied even if there is just a single candidate? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/fullstrong/forcestrongbranch = FALSE # priority of branching rule <gomory> # [type: int, advanced: FALSE, range: [-536870912,536870911], default: -1000] branching/gomory/priority = -1000 # maximal depth level, up to which branching rule <gomory> should be used (-1 for no limit) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] branching/gomory/maxdepth = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying branching rule (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: FALSE, range: [0,1], default: 1] branching/gomory/maxbounddist = 1 # maximum amount of branching candidates to generate Gomory cuts for (-1: all candidates) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] branching/gomory/maxncands = -1 # weight of efficacy in the weighted sum cut scoring rule # [type: real, advanced: FALSE, range: [-1,1], default: 1] branching/gomory/efficacyweight = 1 # weight of objective parallelism in the weighted sum cut scoring rule # [type: real, advanced: FALSE, range: [-1,1], default: 0] branching/gomory/objparallelweight = 0 # weight of integer support in the weighted sum cut scoring rule # [type: real, advanced: FALSE, range: [-1,1], default: 0] branching/gomory/intsupportweight = 0 # whether relpscost branching should be called without branching (used for bound inferences and conflicts) # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] branching/gomory/performrelpscost = FALSE # use weaker cuts that are exactly derived from the branching split disjunction # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] branching/gomory/useweakercuts = TRUE # priority of branching rule <inference> # [type: int, advanced: FALSE, range: [-536870912,536870911], default: 1000] branching/inference/priority = 1000 # maximal depth level, up to which branching rule <inference> should be used (-1 for no limit) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] branching/inference/maxdepth = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying branching rule (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: FALSE, range: [0,1], default: 1] branching/inference/maxbounddist = 1 # weight in score calculations for conflict score # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 1000] branching/inference/conflictweight = 1000 # weight in score calculations for inference score # [type: real, advanced: TRUE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 1] branching/inference/inferenceweight = 1 # weight in score calculations for cutoff score # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 1] branching/inference/cutoffweight = 1 # should branching on LP solution be restricted to the fractional variables? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] branching/inference/fractionals = TRUE # should a weighted sum of inference, conflict and cutoff weights be used? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] branching/inference/useweightedsum = TRUE # weight in score calculations for conflict score # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.001] branching/inference/reliablescore = 0.001 # priority value for using conflict weights in lex. order # [type: int, advanced: FALSE, range: [0,2147483647], default: 1] branching/inference/conflictprio = 1 # priority value for using cutoff weights in lex. order # [type: int, advanced: FALSE, range: [0,2147483647], default: 1] branching/inference/cutoffprio = 1 # priority of branching rule <leastinf> # [type: int, advanced: FALSE, range: [-536870912,536870911], default: 50] branching/leastinf/priority = 50 # maximal depth level, up to which branching rule <leastinf> should be used (-1 for no limit) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] branching/leastinf/maxdepth = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying branching rule (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: FALSE, range: [0,1], default: 1] branching/leastinf/maxbounddist = 1 # priority of branching rule <lookahead> # [type: int, advanced: FALSE, range: [-536870912,536870911], default: 0] branching/lookahead/priority = 0 # maximal depth level, up to which branching rule <lookahead> should be used (-1 for no limit) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] branching/lookahead/maxdepth = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying branching rule (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: FALSE, range: [0,1], default: 1] branching/lookahead/maxbounddist = 1 # should binary constraints be collected and applied? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/lookahead/useimpliedbincons = FALSE # should binary constraints be added as rows to the base LP? (0: no, 1: separate, 2: as initial rows) # [type: int, advanced: TRUE, range: [0,2], default: 0] branching/lookahead/addbinconsrow = 0 # how many constraints that are violated by the base lp solution should be gathered until the rule is stopped and they are added? [0 for unrestricted] # [type: int, advanced: TRUE, range: [0,2147483647], default: 1] branching/lookahead/maxnviolatedcons = 1 # how many binary constraints that are violated by the base lp solution should be gathered until the rule is stopped and they are added? [0 for unrestricted] # [type: int, advanced: TRUE, range: [0,2147483647], default: 0] branching/lookahead/maxnviolatedbincons = 0 # how many domain reductions that are violated by the base lp solution should be gathered until the rule is stopped and they are added? [0 for unrestricted] # [type: int, advanced: TRUE, range: [0,2147483647], default: 1] branching/lookahead/maxnviolateddomreds = 1 # max number of LPs solved after which a previous prob branching results are recalculated # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 10] branching/lookahead/reevalage = 10 # max number of LPs solved after which a previous FSB scoring results are recalculated # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 10] branching/lookahead/reevalagefsb = 10 # the max depth of LAB. # [type: int, advanced: TRUE, range: [1,2147483647], default: 2] branching/lookahead/recursiondepth = 2 # should domain reductions be collected and applied? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] branching/lookahead/usedomainreduction = TRUE # should domain reductions of feasible siblings should be merged? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/lookahead/mergedomainreductions = FALSE # should domain reductions only be applied if there are simple bound changes? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/lookahead/prefersimplebounds = FALSE # should only domain reductions that violate the LP solution be applied? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/lookahead/onlyvioldomreds = FALSE # should binary constraints, that are not violated by the base LP, be collected and added? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/lookahead/addnonviocons = FALSE # toggles the abbreviated LAB. # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] branching/lookahead/abbreviated = TRUE # if abbreviated: The max number of candidates to consider at the node. # [type: int, advanced: TRUE, range: [1,2147483647], default: 4] branching/lookahead/maxncands = 4 # if abbreviated: The max number of candidates to consider per deeper node. # [type: int, advanced: TRUE, range: [0,2147483647], default: 2] branching/lookahead/maxndeepercands = 2 # if abbreviated: Should the information gathered to obtain the best candidates be reused? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] branching/lookahead/reusebasis = TRUE # if only non violating constraints are added, should the branching decision be stored till the next call? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] branching/lookahead/storeunviolatedsol = TRUE # if abbreviated: Use pseudo costs to estimate the score of a candidate. # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/lookahead/abbrevpseudo = FALSE # should the average score be used for uninitialized scores in level 2? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/lookahead/level2avgscore = FALSE # should uninitialized scores in level 2 be set to 0? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/lookahead/level2zeroscore = FALSE # add binary constraints with two variables found at the root node also as a clique # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/lookahead/addclique = FALSE # should domain propagation be executed before each temporary node is solved? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] branching/lookahead/propagate = TRUE # should branching data generated at depth level 2 be stored for re-using it? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] branching/lookahead/uselevel2data = TRUE # should bounds known for child nodes be applied? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/lookahead/applychildbounds = FALSE # should the maximum number of domain reductions maxnviolateddomreds be enforced? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/lookahead/enforcemaxdomreds = FALSE # should branching results (and scores) be updated w.r.t. proven dual bounds? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/lookahead/updatebranchingresults = FALSE # maximum number of propagation rounds to perform at each temporary node (-1: unlimited, 0: SCIP default) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 0] branching/lookahead/maxproprounds = 0 # scoring function to be used at the base level # [type: char, advanced: TRUE, range: {dfswplcra}, default: a] branching/lookahead/scoringfunction = a # scoring function to be used at deeper levels # [type: char, advanced: TRUE, range: {dfswlcrx}, default: x] branching/lookahead/deeperscoringfunction = x # scoring function to be used during FSB scoring # [type: char, advanced: TRUE, range: {dfswlcr}, default: d] branching/lookahead/scoringscoringfunction = d # if scoringfunction is 's', this value is used to weight the min of the gains of two child problems in the convex combination # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.8] branching/lookahead/minweight = 0.8 # if the FSB score is of a candidate is worse than the best by this factor, skip this candidate (-1: disable) # [type: real, advanced: TRUE, range: [-1,1.79769313486232e+308], default: -1] branching/lookahead/worsefactor = -1 # should lookahead branching only be applied if the max gain in level 1 is not uniquely that of the best candidate? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/lookahead/filterbymaxgain = FALSE # priority of branching rule <mostinf> # [type: int, advanced: FALSE, range: [-536870912,536870911], default: 100] branching/mostinf/priority = 100 # maximal depth level, up to which branching rule <mostinf> should be used (-1 for no limit) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] branching/mostinf/maxdepth = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying branching rule (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: FALSE, range: [0,1], default: 1] branching/mostinf/maxbounddist = 1 # priority of branching rule <multaggr> # [type: int, advanced: FALSE, range: [-536870912,536870911], default: 0] branching/multaggr/priority = 0 # maximal depth level, up to which branching rule <multaggr> should be used (-1 for no limit) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] branching/multaggr/maxdepth = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying branching rule (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: FALSE, range: [0,1], default: 1] branching/multaggr/maxbounddist = 1 # number of intermediate LPs solved to trigger reevaluation of strong branching value for a variable that was already evaluated at the current node # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 0] branching/multaggr/reevalage = 0 # maximum number of propagation rounds to be performed during multaggr branching before solving the LP (-1: no limit, -2: parameter settings) # [type: int, advanced: TRUE, range: [-2,2147483647], default: 0] branching/multaggr/maxproprounds = 0 # should valid bounds be identified in a probing-like fashion during multaggr branching (only with propagation)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] branching/multaggr/probingbounds = TRUE # priority of branching rule <nodereopt> # [type: int, advanced: FALSE, range: [-536870912,536870911], default: -9000000] branching/nodereopt/priority = -9000000 # maximal depth level, up to which branching rule <nodereopt> should be used (-1 for no limit) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] branching/nodereopt/maxdepth = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying branching rule (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: FALSE, range: [0,1], default: 1] branching/nodereopt/maxbounddist = 1 # priority of branching rule <pscost> # [type: int, advanced: FALSE, range: [-536870912,536870911], default: 2000] branching/pscost/priority = 2000 # maximal depth level, up to which branching rule <pscost> should be used (-1 for no limit) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] branching/pscost/maxdepth = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying branching rule (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: FALSE, range: [0,1], default: 1] branching/pscost/maxbounddist = 1 # strategy for utilizing pseudo-costs of external branching candidates (multiply as in pseudo costs 'u'pdate rule, or by 'd'omain reduction, or by domain reduction of 's'ibling, or by 'v'ariable score) # [type: char, advanced: FALSE, range: {dsuv}, default: u] branching/pscost/strategy = u # weight for minimum of scores of a branching candidate when building weighted sum of min/max/sum of scores # [type: real, advanced: TRUE, range: [-1e+20,1e+20], default: 0.8] branching/pscost/minscoreweight = 0.8 # weight for maximum of scores of a branching candidate when building weighted sum of min/max/sum of scores # [type: real, advanced: TRUE, range: [-1e+20,1e+20], default: 1.3] branching/pscost/maxscoreweight = 1.3 # weight for sum of scores of a branching candidate when building weighted sum of min/max/sum of scores # [type: real, advanced: TRUE, range: [-1e+20,1e+20], default: 0.1] branching/pscost/sumscoreweight = 0.1 # number of children to create in n-ary branching # [type: int, advanced: FALSE, range: [2,2147483647], default: 2] branching/pscost/nchildren = 2 # maximal depth where to do n-ary branching, -1 to turn off # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] branching/pscost/narymaxdepth = -1 # minimal domain width in children when doing n-ary branching, relative to global bounds # [type: real, advanced: FALSE, range: [0,1], default: 0.001] branching/pscost/naryminwidth = 0.001 # factor of domain width in n-ary branching when creating nodes with increasing distance from branching value # [type: real, advanced: FALSE, range: [1,1.79769313486232e+308], default: 2] branching/pscost/narywidthfactor = 2 # priority of branching rule <random> # [type: int, advanced: FALSE, range: [-536870912,536870911], default: -100000] branching/random/priority = -100000 # maximal depth level, up to which branching rule <random> should be used (-1 for no limit) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] branching/random/maxdepth = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying branching rule (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: FALSE, range: [0,1], default: 1] branching/random/maxbounddist = 1 # initial random seed value # [type: int, advanced: FALSE, range: [0,2147483647], default: 41] branching/random/seed = 41 # priority of branching rule <relpscost> # [type: int, advanced: FALSE, range: [-536870912,536870911], default: 10000] branching/relpscost/priority = 10000 # maximal depth level, up to which branching rule <relpscost> should be used (-1 for no limit) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] branching/relpscost/maxdepth = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying branching rule (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: FALSE, range: [0,1], default: 1] branching/relpscost/maxbounddist = 1 # weight in score calculations for conflict score # [type: real, advanced: TRUE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 0.01] branching/relpscost/conflictweight = 0.01 # weight in score calculations for conflict length score # [type: real, advanced: TRUE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 0] branching/relpscost/conflictlengthweight = 0 # weight in score calculations for inference score # [type: real, advanced: TRUE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 0.0001] branching/relpscost/inferenceweight = 0.0001 # weight in score calculations for cutoff score # [type: real, advanced: TRUE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 0.0001] branching/relpscost/cutoffweight = 0.0001 # weight in score calculations for average GMI cuts normalized efficacy # [type: real, advanced: TRUE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 0] branching/relpscost/gmiavgeffweight = 0 # weight in score calculations for last GMI cuts normalized efficacy # [type: real, advanced: TRUE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 1e-05] branching/relpscost/gmilasteffweight = 1e-05 # weight in score calculations for pseudo cost score # [type: real, advanced: TRUE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 1] branching/relpscost/pscostweight = 1 # weight in score calculations for nlcount score # [type: real, advanced: TRUE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 0.1] branching/relpscost/nlscoreweight = 0.1 # minimal value for minimum pseudo cost size to regard pseudo cost value as reliable # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 1] branching/relpscost/minreliable = 1 # maximal value for minimum pseudo cost size to regard pseudo cost value as reliable # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 5] branching/relpscost/maxreliable = 5 # maximal fraction of strong branching LP iterations compared to node relaxation LP iterations # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.5] branching/relpscost/sbiterquot = 0.5 # additional number of allowed strong branching LP iterations # [type: int, advanced: FALSE, range: [0,2147483647], default: 100000] branching/relpscost/sbiterofs = 100000 # maximal number of further variables evaluated without better score # [type: int, advanced: TRUE, range: [1,2147483647], default: 9] branching/relpscost/maxlookahead = 9 # maximal number of candidates initialized with strong branching per node # [type: int, advanced: FALSE, range: [0,2147483647], default: 100] branching/relpscost/initcand = 100 # iteration limit for strong branching initializations of pseudo cost entries (0: auto) # [type: int, advanced: FALSE, range: [0,2147483647], default: 0] branching/relpscost/inititer = 0 # maximal number of bound tightenings before the node is reevaluated (-1: unlimited) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 5] branching/relpscost/maxbdchgs = 5 # maximum number of propagation rounds to be performed during strong branching before solving the LP (-1: no limit, -2: parameter settings) # [type: int, advanced: TRUE, range: [-2,2147483647], default: -2] branching/relpscost/maxproprounds = -2 # should valid bounds be identified in a probing-like fashion during strong branching (only with propagation)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] branching/relpscost/probingbounds = TRUE # should reliability be based on relative errors? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/relpscost/userelerrorreliability = FALSE # low relative error tolerance for reliability # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.05] branching/relpscost/lowerrortol = 0.05 # high relative error tolerance for reliability # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 1] branching/relpscost/higherrortol = 1 # should strong branching result be considered for pseudo costs if the other direction was infeasible? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/relpscost/storesemiinitcosts = FALSE # should the scoring function use only local cutoff and inference information obtained for strong branching candidates? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/relpscost/usesblocalinfo = FALSE # should the strong branching decision be based on a hypothesis test? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/relpscost/usehyptestforreliability = FALSE # should the confidence level be adjusted dynamically? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/relpscost/usedynamicconfidence = FALSE # should branching rule skip candidates that have a low probability to be better than the best strong-branching or pseudo-candidate? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] branching/relpscost/skipbadinitcands = TRUE # the confidence level for statistical methods, between 0 (Min) and 4 (Max). # [type: int, advanced: TRUE, range: [0,4], default: 2] branching/relpscost/confidencelevel = 2 # should candidates be initialized in randomized order? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/relpscost/randinitorder = FALSE # should smaller weights be used for pseudo cost updates after hitting the LP iteration limit? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/relpscost/usesmallweightsitlim = FALSE # should the weights of the branching rule be adjusted dynamically during solving based on objective and infeasible leaf counters? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] branching/relpscost/dynamicweights = TRUE # should degeneracy be taken into account to update weights and skip strong branching? (0: off, 1: after root, 2: always) # [type: int, advanced: TRUE, range: [0,2], default: 1] branching/relpscost/degeneracyaware = 1 # start seed for random number generation # [type: int, advanced: TRUE, range: [0,2147483647], default: 5] branching/relpscost/startrandseed = 5 # Use symmetry to filter branching candidates? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/relpscost/filtercandssym = FALSE # Transfer pscost information to symmetric variables? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/relpscost/transsympscost = FALSE # should candidate branching variables be scored using the Treemodel branching rules? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] branching/treemodel/enable = FALSE # scoring function to use at nodes predicted to be high in the tree ('d'efault, 's'vts, 'r'atio, 't'ree sample) # [type: char, advanced: FALSE, range: {dsrt}, default: r] branching/treemodel/highrule = r # scoring function to use at nodes predicted to be low in the tree ('d'efault, 's'vts, 'r'atio, 't'ree sample) # [type: char, advanced: FALSE, range: {dsrt}, default: r] branching/treemodel/lowrule = r # estimated tree height at which we switch from using the low rule to the high rule # [type: int, advanced: FALSE, range: [0,2147483647], default: 10] branching/treemodel/height = 10 # should dominated candidates be filtered before using the high scoring function? ('a'uto, 't'rue, 'f'alse) # [type: char, advanced: TRUE, range: {atf}, default: a] branching/treemodel/filterhigh = a # should dominated candidates be filtered before using the low scoring function? ('a'uto, 't'rue, 'f'alse) # [type: char, advanced: TRUE, range: {atf}, default: a] branching/treemodel/filterlow = a # maximum number of fixed-point iterations when computing the ratio # [type: int, advanced: TRUE, range: [1,2147483647], default: 24] branching/treemodel/maxfpiter = 24 # maximum height to compute the SVTS score exactly before approximating # [type: int, advanced: TRUE, range: [0,2147483647], default: 100] branching/treemodel/maxsvtsheight = 100 # which method should be used as a fallback if the tree size estimates are infinite? ('d'efault, 'r'atio) # [type: char, advanced: TRUE, range: {dr}, default: r] branching/treemodel/fallbackinf = r # which method should be used as a fallback if there is no primal bound available? ('d'efault, 'r'atio) # [type: char, advanced: TRUE, range: {dr}, default: r] branching/treemodel/fallbacknoprim = r # threshold at which pseudocosts are considered small, making hybrid scores more likely to be the deciding factor in branching # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.1] branching/treemodel/smallpscost = 0.1 # priority of branching rule <vanillafullstrong> # [type: int, advanced: FALSE, range: [-536870912,536870911], default: -2000] branching/vanillafullstrong/priority = -2000 # maximal depth level, up to which branching rule <vanillafullstrong> should be used (-1 for no limit) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] branching/vanillafullstrong/maxdepth = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying branching rule (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: FALSE, range: [0,1], default: 1] branching/vanillafullstrong/maxbounddist = 1 # should integral variables in the current LP solution be considered as branching candidates? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] branching/vanillafullstrong/integralcands = FALSE # should strong branching side-effects be prevented (e.g., domain changes, stat updates etc.)? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] branching/vanillafullstrong/idempotent = FALSE # should strong branching scores be computed for all candidates, or can we early stop when a variable has infinite score? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/vanillafullstrong/scoreall = FALSE # should strong branching scores be collected? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/vanillafullstrong/collectscores = FALSE # should candidates only be scored, but no branching be performed? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] branching/vanillafullstrong/donotbranch = FALSE # restart policy: (a)lways, (c)ompletion, (e)stimation, (n)ever # [type: char, advanced: FALSE, range: {acen}, default: e] estimation/restarts/restartpolicy = e # tree size estimation method: (c)ompletion, (e)nsemble, time series forecasts on either (g)ap, (l)eaf frequency, (o)open nodes, tree (w)eight, (s)sg, or (t)ree profile or w(b)e # [type: char, advanced: FALSE, range: {bceglostw}, default: w] estimation/method = w # restart limit # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1] estimation/restarts/restartlimit = 1 # minimum number of nodes before restart # [type: longint, advanced: FALSE, range: [-1,9223372036854775807], default: 1000] estimation/restarts/minnodes = 1000 # should only leaves count for the minnodes parameter? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] estimation/restarts/countonlyleaves = FALSE # factor by which the estimated number of nodes should exceed the current number of nodes # [type: real, advanced: FALSE, range: [1,1.79769313486232e+308], default: 50] estimation/restarts/restartfactor = 50 # whether to apply a restart when nonlinear constraints are present # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] estimation/restarts/restartnonlinear = FALSE # whether to apply a restart when active pricers are used # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] estimation/restarts/restartactpricers = FALSE # coefficient of tree weight in monotone approximation of search completion # [type: real, advanced: FALSE, range: [0,1], default: 0.3667] estimation/coefmonoweight = 0.3667 # coefficient of 1 - SSG in monotone approximation of search completion # [type: real, advanced: FALSE, range: [0,1], default: 0.6333] estimation/coefmonossg = 0.6333 # limit on the number of successive samples to really trigger a restart # [type: int, advanced: FALSE, range: [1,2147483647], default: 50] estimation/restarts/hitcounterlim = 50 # report frequency on estimation: -1: never, 0:always, k >= 1: k times evenly during search # [type: int, advanced: TRUE, range: [-1,1073741823], default: -1] estimation/reportfreq = -1 # user regression forest in RFCSV format # [type: string, advanced: FALSE, default: "-"] estimation/regforestfilename = "-" # approximation of search tree completion: (a)uto, (g)ap, tree (w)eight, (m)onotone regression, (r)egression forest, (s)sg # [type: char, advanced: FALSE, range: {agmrsw}, default: a] estimation/completiontype = a # should the event handler collect data? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] estimation/treeprofile/enabled = FALSE # minimum average number of nodes at each depth before producing estimations # [type: real, advanced: FALSE, range: [1,1.79769313486232e+308], default: 20] estimation/treeprofile/minnodesperdepth = 20 # use leaf nodes as basic observations for time series, or all nodes? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] estimation/useleafts = TRUE # should statistics be shown at the end? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] estimation/showstats = FALSE # the maximum number of individual SSG subtrees; -1: no limit # [type: int, advanced: FALSE, range: [-1,1073741823], default: -1] estimation/ssg/nmaxsubtrees = -1 # minimum number of nodes to process between two consecutive SSG splits # [type: longint, advanced: FALSE, range: [0,9223372036854775807], default: 0] estimation/ssg/nminnodeslastsplit = 0 # is statistics table <estim> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/estim/active = TRUE # display activation status of display column <completed> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/completed/active = 1 # display activation status of display column <nrank1nodes> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 0] display/nrank1nodes/active = 0 # display activation status of display column <nnodesbelowinc> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 0] display/nnodesbelowinc/active = 0 # should the event handler adapt the solver behavior? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] solvingphases/enabled = FALSE # should the event handler test all phase transitions? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] solvingphases/testmode = FALSE # settings file for feasibility phase -- precedence over emphasis settings # [type: string, advanced: FALSE, default: "-"] solvingphases/feassetname = "-" # settings file for improvement phase -- precedence over emphasis settings # [type: string, advanced: FALSE, default: "-"] solvingphases/improvesetname = "-" # settings file for proof phase -- precedence over emphasis settings # [type: string, advanced: FALSE, default: "-"] solvingphases/proofsetname = "-" # node offset for rank-1 and estimate transitions # [type: longint, advanced: FALSE, range: [1,9223372036854775807], default: 50] solvingphases/nodeoffset = 50 # should the event handler fall back from optimal phase? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] solvingphases/fallback = FALSE # transition method: Possible options are 'e'stimate,'l'ogarithmic regression,'o'ptimal-value based,'r'ank-1 # [type: char, advanced: FALSE, range: {elor}, default: r] solvingphases/transitionmethod = r # should the event handler interrupt the solving process after optimal solution was found? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] solvingphases/interruptoptimal = FALSE # should a restart be applied between the feasibility and improvement phase? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] solvingphases/userestart1to2 = FALSE # should a restart be applied between the improvement and the proof phase? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] solvingphases/userestart2to3 = FALSE # optimal solution value for problem # [type: real, advanced: FALSE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 1e+99] solvingphases/optimalvalue = 1e+99 # x-type for logarithmic regression - (t)ime, (n)odes, (l)p iterations # [type: char, advanced: FALSE, range: {lnt}, default: n] solvingphases/xtype = n # should emphasis settings for the solving phases be used, or settings files? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] solvingphases/useemphsettings = TRUE # priority of compression <largestrepr> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 2000] compression/largestrepr/priority = 2000 # minimal number of leave nodes for calling tree compression <largestrepr> # [type: int, advanced: FALSE, range: [1,2147483647], default: 20] compression/largestrepr/minnleaves = 20 # number of runs in the constrained part. # [type: int, advanced: FALSE, range: [1,2147483647], default: 5] compression/largestrepr/iterations = 5 # minimal number of common variables. # [type: int, advanced: FALSE, range: [1,2147483647], default: 3] compression/largestrepr/mincommonvars = 3 # priority of compression <weakcompr> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 1000] compression/weakcompr/priority = 1000 # minimal number of leave nodes for calling tree compression <weakcompr> # [type: int, advanced: FALSE, range: [1,2147483647], default: 50] compression/weakcompr/minnleaves = 50 # convert constraints into nodes # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] compression/weakcompr/convertconss = FALSE # priority of heuristic <actconsdiving> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1003700] heuristics/actconsdiving/priority = -1003700 # frequency for calling primal heuristic <actconsdiving> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] heuristics/actconsdiving/freq = -1 # frequency offset for calling primal heuristic <actconsdiving> # [type: int, advanced: FALSE, range: [0,1073741822], default: 5] heuristics/actconsdiving/freqofs = 5 # maximal depth level to call primal heuristic <actconsdiving> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/actconsdiving/maxdepth = -1 # minimal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/actconsdiving/minreldepth = 0 # maximal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/actconsdiving/maxreldepth = 1 # maximal fraction of diving LP iterations compared to node LP iterations # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.05] heuristics/actconsdiving/maxlpiterquot = 0.05 # additional number of allowed LP iterations # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] heuristics/actconsdiving/maxlpiterofs = 1000 # maximal quotient (curlowerbound - lowerbound)/(cutoffbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.8] heuristics/actconsdiving/maxdiveubquot = 0.8 # maximal quotient (curlowerbound - lowerbound)/(avglowerbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/actconsdiving/maxdiveavgquot = 0 # maximal UBQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/actconsdiving/maxdiveubquotnosol = 1 # maximal AVGQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 1] heuristics/actconsdiving/maxdiveavgquotnosol = 1 # use one level of backtracking if infeasibility is encountered? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/actconsdiving/backtrack = TRUE # percentage of immediate domain changes during probing to trigger LP resolve # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.15] heuristics/actconsdiving/lpresolvedomchgquot = 0.15 # LP solve frequency for diving heuristics (0: only after enough domain changes have been found) # [type: int, advanced: FALSE, range: [0,2147483647], default: 0] heuristics/actconsdiving/lpsolvefreq = 0 # should only LP branching candidates be considered instead of the slower but more general constraint handler diving variable selection? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/actconsdiving/onlylpbranchcands = TRUE # priority of heuristic <adaptivediving> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -70000] heuristics/adaptivediving/priority = -70000 # frequency for calling primal heuristic <adaptivediving> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 5] heuristics/adaptivediving/freq = 5 # frequency offset for calling primal heuristic <adaptivediving> # [type: int, advanced: FALSE, range: [0,1073741822], default: 3] heuristics/adaptivediving/freqofs = 3 # maximal depth level to call primal heuristic <adaptivediving> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/adaptivediving/maxdepth = -1 # parameter that increases probability of exploration among divesets (only active if seltype is 'e') # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 1] heuristics/adaptivediving/epsilon = 1 # score parameter for selection: minimize either average 'n'odes, LP 'i'terations,backtrack/'c'onflict ratio, 'd'epth, 1 / 's'olutions, or 1 / solutions'u'ccess # [type: char, advanced: FALSE, range: {cdinsu}, default: c] heuristics/adaptivediving/scoretype = c # selection strategy: (e)psilon-greedy, (w)eighted distribution, (n)ext diving # [type: char, advanced: FALSE, range: {enw}, default: w] heuristics/adaptivediving/seltype = w # should the heuristic use its own statistics, or shared statistics? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/adaptivediving/useadaptivecontext = FALSE # coefficient c to decrease initial confidence (calls + 1.0) / (calls + c) in scores # [type: real, advanced: FALSE, range: [1,2147483647], default: 10] heuristics/adaptivediving/selconfidencecoeff = 10 # maximal fraction of diving LP iterations compared to node LP iterations # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.1] heuristics/adaptivediving/maxlpiterquot = 0.1 # additional number of allowed LP iterations # [type: longint, advanced: FALSE, range: [0,2147483647], default: 1500] heuristics/adaptivediving/maxlpiterofs = 1500 # weight of incumbent solutions compared to other solutions in computation of LP iteration limit # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 10] heuristics/adaptivediving/bestsolweight = 10 # priority of heuristic <bound> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1107000] heuristics/bound/priority = -1107000 # frequency for calling primal heuristic <bound> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] heuristics/bound/freq = -1 # frequency offset for calling primal heuristic <bound> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/bound/freqofs = 0 # maximal depth level to call primal heuristic <bound> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/bound/maxdepth = -1 # Should heuristic only be executed if no primal solution was found, yet? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/bound/onlywithoutsol = TRUE # maximum number of propagation rounds during probing (-1 infinity, -2 parameter settings) # [type: int, advanced: TRUE, range: [-1,536870911], default: 0] heuristics/bound/maxproprounds = 0 # to which bound should integer variables be fixed? ('l'ower, 'u'pper, or 'b'oth) # [type: char, advanced: FALSE, range: {lub}, default: l] heuristics/bound/bound = l # priority of heuristic <clique> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 5000] heuristics/clique/priority = 5000 # frequency for calling primal heuristic <clique> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] heuristics/clique/freq = 0 # frequency offset for calling primal heuristic <clique> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/clique/freqofs = 0 # maximal depth level to call primal heuristic <clique> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/clique/maxdepth = -1 # minimum percentage of integer variables that have to be fixable # [type: real, advanced: FALSE, range: [0,1], default: 0.65] heuristics/clique/minintfixingrate = 0.65 # minimum percentage of fixed variables in the sub-MIP # [type: real, advanced: FALSE, range: [0,1], default: 0.65] heuristics/clique/minmipfixingrate = 0.65 # maximum number of nodes to regard in the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 5000] heuristics/clique/maxnodes = 5000 # number of nodes added to the contingent of the total nodes # [type: longint, advanced: FALSE, range: [0,9223372036854775807], default: 500] heuristics/clique/nodesofs = 500 # minimum number of nodes required to start the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 500] heuristics/clique/minnodes = 500 # contingent of sub problem nodes in relation to the number of nodes of the original problem # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/clique/nodesquot = 0.1 # factor by which clique heuristic should at least improve the incumbent # [type: real, advanced: TRUE, range: [0,1], default: 0.01] heuristics/clique/minimprove = 0.01 # maximum number of propagation rounds during probing (-1 infinity) # [type: int, advanced: TRUE, range: [-1,536870911], default: 2] heuristics/clique/maxproprounds = 2 # should all active cuts from cutpool be copied to constraints in subproblem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/clique/copycuts = TRUE # should more variables be fixed based on variable locks if the fixing rate was not reached? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/clique/uselockfixings = FALSE # maximum number of backtracks during the fixing process # [type: int, advanced: TRUE, range: [-1,536870911], default: 10] heuristics/clique/maxbacktracks = 10 # priority of heuristic <coefdiving> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1001000] heuristics/coefdiving/priority = -1001000 # frequency for calling primal heuristic <coefdiving> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] heuristics/coefdiving/freq = -1 # frequency offset for calling primal heuristic <coefdiving> # [type: int, advanced: FALSE, range: [0,1073741822], default: 1] heuristics/coefdiving/freqofs = 1 # maximal depth level to call primal heuristic <coefdiving> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/coefdiving/maxdepth = -1 # minimal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/coefdiving/minreldepth = 0 # maximal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/coefdiving/maxreldepth = 1 # maximal fraction of diving LP iterations compared to node LP iterations # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.05] heuristics/coefdiving/maxlpiterquot = 0.05 # additional number of allowed LP iterations # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] heuristics/coefdiving/maxlpiterofs = 1000 # maximal quotient (curlowerbound - lowerbound)/(cutoffbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.8] heuristics/coefdiving/maxdiveubquot = 0.8 # maximal quotient (curlowerbound - lowerbound)/(avglowerbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/coefdiving/maxdiveavgquot = 0 # maximal UBQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.1] heuristics/coefdiving/maxdiveubquotnosol = 0.1 # maximal AVGQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/coefdiving/maxdiveavgquotnosol = 0 # use one level of backtracking if infeasibility is encountered? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/coefdiving/backtrack = TRUE # percentage of immediate domain changes during probing to trigger LP resolve # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.15] heuristics/coefdiving/lpresolvedomchgquot = 0.15 # LP solve frequency for diving heuristics (0: only after enough domain changes have been found) # [type: int, advanced: FALSE, range: [0,2147483647], default: 0] heuristics/coefdiving/lpsolvefreq = 0 # should only LP branching candidates be considered instead of the slower but more general constraint handler diving variable selection? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/coefdiving/onlylpbranchcands = FALSE # priority of heuristic <completesol> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 0] heuristics/completesol/priority = 0 # frequency for calling primal heuristic <completesol> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] heuristics/completesol/freq = 0 # frequency offset for calling primal heuristic <completesol> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/completesol/freqofs = 0 # maximal depth level to call primal heuristic <completesol> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 0] heuristics/completesol/maxdepth = 0 # maximum number of nodes to regard in the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 5000] heuristics/completesol/maxnodes = 5000 # minimum number of nodes required to start the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 50] heuristics/completesol/minnodes = 50 # maximal rate of unknown solution values # [type: real, advanced: FALSE, range: [0,1], default: 0.85] heuristics/completesol/maxunknownrate = 0.85 # should all subproblem solutions be added to the original SCIP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/completesol/addallsols = FALSE # number of nodes added to the contingent of the total nodes # [type: longint, advanced: FALSE, range: [0,9223372036854775807], default: 500] heuristics/completesol/nodesofs = 500 # contingent of sub problem nodes in relation to the number of nodes of the original problem # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/completesol/nodesquot = 0.1 # factor by which the limit on the number of LP depends on the node limit # [type: real, advanced: TRUE, range: [1,1.79769313486232e+308], default: 2] heuristics/completesol/lplimfac = 2 # weight of the original objective function (1: only original objective) # [type: real, advanced: TRUE, range: [0.001,1], default: 1] heuristics/completesol/objweight = 1 # bound widening factor applied to continuous variables (0: fix variables to given solution values, 1: relax to global bounds) # [type: real, advanced: TRUE, range: [0,1], default: 0.1] heuristics/completesol/boundwidening = 0.1 # factor by which the incumbent should be improved at least # [type: real, advanced: TRUE, range: [0,1], default: 0.01] heuristics/completesol/minimprove = 0.01 # should number of continuous variables be ignored? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/completesol/ignorecont = FALSE # heuristic stops, if the given number of improving solutions were found (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 5] heuristics/completesol/solutions = 5 # maximal number of iterations in propagation (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 10] heuristics/completesol/maxproprounds = 10 # should the heuristic run before presolving? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/completesol/beforepresol = TRUE # maximal number of LP iterations (-1: no limit) # [type: longint, advanced: FALSE, range: [-1,9223372036854775807], default: -1] heuristics/completesol/maxlpiter = -1 # maximal number of continuous variables after presolving # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] heuristics/completesol/maxcontvars = -1 # priority of heuristic <conflictdiving> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1000100] heuristics/conflictdiving/priority = -1000100 # frequency for calling primal heuristic <conflictdiving> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] heuristics/conflictdiving/freq = 10 # frequency offset for calling primal heuristic <conflictdiving> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/conflictdiving/freqofs = 0 # maximal depth level to call primal heuristic <conflictdiving> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/conflictdiving/maxdepth = -1 # minimal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/conflictdiving/minreldepth = 0 # maximal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/conflictdiving/maxreldepth = 1 # maximal fraction of diving LP iterations compared to node LP iterations # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.15] heuristics/conflictdiving/maxlpiterquot = 0.15 # additional number of allowed LP iterations # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] heuristics/conflictdiving/maxlpiterofs = 1000 # maximal quotient (curlowerbound - lowerbound)/(cutoffbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.8] heuristics/conflictdiving/maxdiveubquot = 0.8 # maximal quotient (curlowerbound - lowerbound)/(avglowerbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/conflictdiving/maxdiveavgquot = 0 # maximal UBQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.1] heuristics/conflictdiving/maxdiveubquotnosol = 0.1 # maximal AVGQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/conflictdiving/maxdiveavgquotnosol = 0 # use one level of backtracking if infeasibility is encountered? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/conflictdiving/backtrack = TRUE # percentage of immediate domain changes during probing to trigger LP resolve # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.15] heuristics/conflictdiving/lpresolvedomchgquot = 0.15 # LP solve frequency for diving heuristics (0: only after enough domain changes have been found) # [type: int, advanced: FALSE, range: [0,2147483647], default: 0] heuristics/conflictdiving/lpsolvefreq = 0 # should only LP branching candidates be considered instead of the slower but more general constraint handler diving variable selection? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/conflictdiving/onlylpbranchcands = FALSE # try to maximize the violation # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/conflictdiving/maxviol = TRUE # perform rounding like coefficient diving # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/conflictdiving/likecoef = FALSE # minimal number of conflict locks per variable # [type: int, advanced: TRUE, range: [0,2147483647], default: 5] heuristics/conflictdiving/minconflictlocks = 5 # weight used in a convex combination of conflict and variable locks # [type: real, advanced: TRUE, range: [0,1], default: 0.75] heuristics/conflictdiving/lockweight = 0.75 # priority of heuristic <crossover> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1104000] heuristics/crossover/priority = -1104000 # frequency for calling primal heuristic <crossover> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 30] heuristics/crossover/freq = 30 # frequency offset for calling primal heuristic <crossover> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/crossover/freqofs = 0 # maximal depth level to call primal heuristic <crossover> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/crossover/maxdepth = -1 # number of nodes added to the contingent of the total nodes # [type: longint, advanced: FALSE, range: [0,9223372036854775807], default: 500] heuristics/crossover/nodesofs = 500 # maximum number of nodes to regard in the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 5000] heuristics/crossover/maxnodes = 5000 # minimum number of nodes required to start the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 50] heuristics/crossover/minnodes = 50 # number of solutions to be taken into account # [type: int, advanced: FALSE, range: [2,2147483647], default: 3] heuristics/crossover/nusedsols = 3 # number of nodes without incumbent change that heuristic should wait # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 200] heuristics/crossover/nwaitingnodes = 200 # contingent of sub problem nodes in relation to the number of nodes of the original problem # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/crossover/nodesquot = 0.1 # minimum percentage of integer variables that have to be fixed # [type: real, advanced: FALSE, range: [0,1], default: 0.666] heuristics/crossover/minfixingrate = 0.666 # factor by which Crossover should at least improve the incumbent # [type: real, advanced: TRUE, range: [0,1], default: 0.01] heuristics/crossover/minimprove = 0.01 # factor by which the limit on the number of LP depends on the node limit # [type: real, advanced: TRUE, range: [1,1.79769313486232e+308], default: 2] heuristics/crossover/lplimfac = 2 # should the choice which sols to take be randomized? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/crossover/randomization = TRUE # should the nwaitingnodes parameter be ignored at the root node? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/crossover/dontwaitatroot = FALSE # should subproblem be created out of the rows in the LP rows? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/crossover/uselprows = FALSE # if uselprows == FALSE, should all active cuts from cutpool be copied to constraints in subproblem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/crossover/copycuts = TRUE # should the subproblem be permuted to increase diversification? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/crossover/permute = FALSE # limit on number of improving incumbent solutions in sub-CIP # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] heuristics/crossover/bestsollimit = -1 # should uct node selection be used at the beginning of the search? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/crossover/useuct = FALSE # priority of heuristic <dins> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1105000] heuristics/dins/priority = -1105000 # frequency for calling primal heuristic <dins> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] heuristics/dins/freq = -1 # frequency offset for calling primal heuristic <dins> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/dins/freqofs = 0 # maximal depth level to call primal heuristic <dins> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/dins/maxdepth = -1 # number of nodes added to the contingent of the total nodes # [type: longint, advanced: FALSE, range: [0,9223372036854775807], default: 5000] heuristics/dins/nodesofs = 5000 # contingent of sub problem nodes in relation to the number of nodes of the original problem # [type: real, advanced: FALSE, range: [0,1], default: 0.05] heuristics/dins/nodesquot = 0.05 # minimum number of nodes required to start the subproblem # [type: longint, advanced: FALSE, range: [0,9223372036854775807], default: 50] heuristics/dins/minnodes = 50 # number of pool-solutions to be checked for flag array update (for hard fixing of binary variables) # [type: int, advanced: FALSE, range: [1,2147483647], default: 5] heuristics/dins/solnum = 5 # radius (using Manhattan metric) of the incumbent's neighborhood to be searched # [type: int, advanced: FALSE, range: [1,2147483647], default: 18] heuristics/dins/neighborhoodsize = 18 # maximum number of nodes to regard in the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 5000] heuristics/dins/maxnodes = 5000 # factor by which dins should at least improve the incumbent # [type: real, advanced: TRUE, range: [0,1], default: 0.01] heuristics/dins/minimprove = 0.01 # number of nodes without incumbent change that heuristic should wait # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 200] heuristics/dins/nwaitingnodes = 200 # factor by which the limit on the number of LP depends on the node limit # [type: real, advanced: TRUE, range: [1,1.79769313486232e+308], default: 1.5] heuristics/dins/lplimfac = 1.5 # minimum percentage of integer variables that have to be fixable # [type: real, advanced: FALSE, range: [0,1], default: 0.3] heuristics/dins/minfixingrate = 0.3 # should subproblem be created out of the rows in the LP rows? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/dins/uselprows = FALSE # if uselprows == FALSE, should all active cuts from cutpool be copied to constraints in subproblem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/dins/copycuts = TRUE # should uct node selection be used at the beginning of the search? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/dins/useuct = FALSE # limit on number of improving incumbent solutions in sub-CIP # [type: int, advanced: FALSE, range: [-1,2147483647], default: 3] heuristics/dins/bestsollimit = 3 # priority of heuristic <distributiondiving> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1003300] heuristics/distributiondiving/priority = -1003300 # frequency for calling primal heuristic <distributiondiving> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] heuristics/distributiondiving/freq = 10 # frequency offset for calling primal heuristic <distributiondiving> # [type: int, advanced: FALSE, range: [0,1073741822], default: 3] heuristics/distributiondiving/freqofs = 3 # maximal depth level to call primal heuristic <distributiondiving> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/distributiondiving/maxdepth = -1 # minimal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/distributiondiving/minreldepth = 0 # maximal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/distributiondiving/maxreldepth = 1 # maximal fraction of diving LP iterations compared to node LP iterations # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.05] heuristics/distributiondiving/maxlpiterquot = 0.05 # additional number of allowed LP iterations # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] heuristics/distributiondiving/maxlpiterofs = 1000 # maximal quotient (curlowerbound - lowerbound)/(cutoffbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.8] heuristics/distributiondiving/maxdiveubquot = 0.8 # maximal quotient (curlowerbound - lowerbound)/(avglowerbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/distributiondiving/maxdiveavgquot = 0 # maximal UBQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.1] heuristics/distributiondiving/maxdiveubquotnosol = 0.1 # maximal AVGQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/distributiondiving/maxdiveavgquotnosol = 0 # use one level of backtracking if infeasibility is encountered? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/distributiondiving/backtrack = TRUE # percentage of immediate domain changes during probing to trigger LP resolve # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.15] heuristics/distributiondiving/lpresolvedomchgquot = 0.15 # LP solve frequency for diving heuristics (0: only after enough domain changes have been found) # [type: int, advanced: FALSE, range: [0,2147483647], default: 0] heuristics/distributiondiving/lpsolvefreq = 0 # should only LP branching candidates be considered instead of the slower but more general constraint handler diving variable selection? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/distributiondiving/onlylpbranchcands = TRUE # the score;largest 'd'ifference, 'l'owest cumulative probability,'h'ighest c.p., 'v'otes lowest c.p., votes highest c.p.('w'), 'r'evolving # [type: char, advanced: TRUE, range: {lvdhwr}, default: r] heuristics/distributiondiving/scoreparam = r # priority of heuristic <dps> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 75000] heuristics/dps/priority = 75000 # frequency for calling primal heuristic <dps> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] heuristics/dps/freq = -1 # frequency offset for calling primal heuristic <dps> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/dps/freqofs = 0 # maximal depth level to call primal heuristic <dps> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/dps/maxdepth = -1 # maximal number of iterations # [type: int, advanced: FALSE, range: [1,2147483647], default: 50] heuristics/dps/maxiterations = 50 # maximal linking score of used decomposition (equivalent to percentage of linking constraints) # [type: real, advanced: FALSE, range: [0,1], default: 1] heuristics/dps/maxlinkscore = 1 # multiplier for absolute increase of penalty parameters (0: no increase) # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 100] heuristics/dps/penalty = 100 # should the problem get reoptimized with the original objective function? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/dps/reoptimize = FALSE # should solutions get reused in subproblems? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/dps/reuse = FALSE # should strict limits for reoptimization be set? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/dps/reoptlimits = TRUE # should the heuristic run before or after the processing of the node? (0: before, 1: after, 2: both) # [type: int, advanced: FALSE, range: [0,2], default: 0] heuristics/dps/timing = 0 # priority of heuristic <dualval> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -10] heuristics/dualval/priority = -10 # frequency for calling primal heuristic <dualval> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] heuristics/dualval/freq = -1 # frequency offset for calling primal heuristic <dualval> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/dualval/freqofs = 0 # maximal depth level to call primal heuristic <dualval> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/dualval/maxdepth = -1 # exit if objective doesn't improve # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/dualval/forceimprovements = FALSE # add constraint to ensure that discrete vars are improving # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/dualval/onlycheaper = TRUE # disable the heuristic if it was not called at a leaf of the B&B tree # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/dualval/onlyleaves = FALSE # relax the indicator variables by introducing continuous copies # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/dualval/relaxindicators = FALSE # relax the continous variables # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/dualval/relaxcontvars = FALSE # verblevel of the heuristic, default is 0 to display nothing # [type: int, advanced: FALSE, range: [0,4], default: 0] heuristics/dualval/heurverblevel = 0 # verblevel of the nlp solver, can be 0 or 1 # [type: int, advanced: FALSE, range: [0,1], default: 0] heuristics/dualval/nlpverblevel = 0 # number of ranks that should be displayed when the heuristic is called # [type: int, advanced: FALSE, range: [0,2147483647], default: 10] heuristics/dualval/rankvalue = 10 # maximal number of recursive calls of the heuristic (if dynamicdepth is off) # [type: int, advanced: FALSE, range: [0,2147483647], default: 25] heuristics/dualval/maxcalls = 25 # says if and how the recursion depth is computed at runtime # [type: int, advanced: FALSE, range: [0,1], default: 0] heuristics/dualval/dynamicdepth = 0 # maximal number of variables that may have maximal rank, quit if there are more, turn off by setting -1 # [type: int, advanced: FALSE, range: [-1,2147483647], default: 50] heuristics/dualval/maxequalranks = 50 # minimal gap for which we still run the heuristic, if gap is less we return without doing anything # [type: real, advanced: FALSE, range: [0,100], default: 5] heuristics/dualval/mingap = 5 # value added to objective of slack variables, must not be zero # [type: real, advanced: FALSE, range: [0.1,1e+20], default: 1] heuristics/dualval/lambdaslack = 1 # scaling factor for the objective function # [type: real, advanced: FALSE, range: [0,1], default: 0] heuristics/dualval/lambdaobj = 0 # priority of heuristic <farkasdiving> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -900000] heuristics/farkasdiving/priority = -900000 # frequency for calling primal heuristic <farkasdiving> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] heuristics/farkasdiving/freq = 10 # frequency offset for calling primal heuristic <farkasdiving> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/farkasdiving/freqofs = 0 # maximal depth level to call primal heuristic <farkasdiving> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/farkasdiving/maxdepth = -1 # minimal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/farkasdiving/minreldepth = 0 # maximal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/farkasdiving/maxreldepth = 1 # maximal fraction of diving LP iterations compared to node LP iterations # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.05] heuristics/farkasdiving/maxlpiterquot = 0.05 # additional number of allowed LP iterations # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] heuristics/farkasdiving/maxlpiterofs = 1000 # maximal quotient (curlowerbound - lowerbound)/(cutoffbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.8] heuristics/farkasdiving/maxdiveubquot = 0.8 # maximal quotient (curlowerbound - lowerbound)/(avglowerbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/farkasdiving/maxdiveavgquot = 0 # maximal UBQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.1] heuristics/farkasdiving/maxdiveubquotnosol = 0.1 # maximal AVGQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/farkasdiving/maxdiveavgquotnosol = 0 # use one level of backtracking if infeasibility is encountered? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/farkasdiving/backtrack = TRUE # percentage of immediate domain changes during probing to trigger LP resolve # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.15] heuristics/farkasdiving/lpresolvedomchgquot = 0.15 # LP solve frequency for diving heuristics (0: only after enough domain changes have been found) # [type: int, advanced: FALSE, range: [0,2147483647], default: 1] heuristics/farkasdiving/lpsolvefreq = 1 # should only LP branching candidates be considered instead of the slower but more general constraint handler diving variable selection? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/farkasdiving/onlylpbranchcands = FALSE # should diving candidates be checked before running? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/farkasdiving/checkcands = FALSE # should the score be scaled? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/farkasdiving/scalescore = TRUE # should the heuristic only run within the tree if at least one solution was found at the root node? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/farkasdiving/rootsuccess = TRUE # maximal occurance factor of an objective coefficient # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/farkasdiving/maxobjocc = 1 # minimal objective dynamism (log) to run # [type: real, advanced: TRUE, range: [0,1e+20], default: 0.0001] heuristics/farkasdiving/objdynamism = 0.0001 # scale score by [f]ractionality or [i]mpact on farkasproof # [type: char, advanced: TRUE, range: {fi}, default: i] heuristics/farkasdiving/scaletype = i # priority of heuristic <feaspump> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1000000] heuristics/feaspump/priority = -1000000 # frequency for calling primal heuristic <feaspump> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 20] heuristics/feaspump/freq = 20 # frequency offset for calling primal heuristic <feaspump> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/feaspump/freqofs = 0 # maximal depth level to call primal heuristic <feaspump> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/feaspump/maxdepth = -1 # maximal fraction of diving LP iterations compared to node LP iterations # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.01] heuristics/feaspump/maxlpiterquot = 0.01 # factor by which the regard of the objective is decreased in each round, 1.0 for dynamic # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/feaspump/objfactor = 0.1 # initial weight of the objective function in the convex combination # [type: real, advanced: FALSE, range: [0,1], default: 1] heuristics/feaspump/alpha = 1 # threshold difference for the convex parameter to perform perturbation # [type: real, advanced: FALSE, range: [0,1], default: 1] heuristics/feaspump/alphadiff = 1 # additional number of allowed LP iterations # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] heuristics/feaspump/maxlpiterofs = 1000 # total number of feasible solutions found up to which heuristic is called (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 10] heuristics/feaspump/maxsols = 10 # maximal number of pumping loops (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 10000] heuristics/feaspump/maxloops = 10000 # maximal number of pumping rounds without fractionality improvement (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 10] heuristics/feaspump/maxstallloops = 10 # minimum number of random variables to flip, if a 1-cycle is encountered # [type: int, advanced: TRUE, range: [1,2147483647], default: 10] heuristics/feaspump/minflips = 10 # maximum length of cycles to be checked explicitly in each round # [type: int, advanced: TRUE, range: [1,100], default: 3] heuristics/feaspump/cyclelength = 3 # number of iterations until a random perturbation is forced # [type: int, advanced: TRUE, range: [1,2147483647], default: 100] heuristics/feaspump/perturbfreq = 100 # radius (using Manhattan metric) of the neighborhood to be searched in stage 3 # [type: int, advanced: FALSE, range: [1,2147483647], default: 18] heuristics/feaspump/neighborhoodsize = 18 # should the feasibility pump be called at root node before cut separation? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/feaspump/beforecuts = TRUE # should an iterative round-and-propagate scheme be used to find the integral points? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/feaspump/usefp20 = FALSE # should a random perturbation be performed if a feasible solution was found? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/feaspump/pertsolfound = TRUE # should we solve a local branching sub-MIP if no solution could be found? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/feaspump/stage3 = FALSE # should all active cuts from cutpool be copied to constraints in subproblem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/feaspump/copycuts = TRUE # priority of heuristic <fixandinfer> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -500000] heuristics/fixandinfer/priority = -500000 # frequency for calling primal heuristic <fixandinfer> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] heuristics/fixandinfer/freq = -1 # frequency offset for calling primal heuristic <fixandinfer> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/fixandinfer/freqofs = 0 # maximal depth level to call primal heuristic <fixandinfer> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/fixandinfer/maxdepth = -1 # maximal number of propagation rounds in probing subproblems (-1: no limit, 0: auto) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 0] heuristics/fixandinfer/proprounds = 0 # minimal number of fixings to apply before dive may be aborted # [type: int, advanced: TRUE, range: [0,2147483647], default: 100] heuristics/fixandinfer/minfixings = 100 # priority of heuristic <fracdiving> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1003000] heuristics/fracdiving/priority = -1003000 # frequency for calling primal heuristic <fracdiving> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] heuristics/fracdiving/freq = 10 # frequency offset for calling primal heuristic <fracdiving> # [type: int, advanced: FALSE, range: [0,1073741822], default: 3] heuristics/fracdiving/freqofs = 3 # maximal depth level to call primal heuristic <fracdiving> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/fracdiving/maxdepth = -1 # minimal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/fracdiving/minreldepth = 0 # maximal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/fracdiving/maxreldepth = 1 # maximal fraction of diving LP iterations compared to node LP iterations # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.05] heuristics/fracdiving/maxlpiterquot = 0.05 # additional number of allowed LP iterations # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] heuristics/fracdiving/maxlpiterofs = 1000 # maximal quotient (curlowerbound - lowerbound)/(cutoffbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.8] heuristics/fracdiving/maxdiveubquot = 0.8 # maximal quotient (curlowerbound - lowerbound)/(avglowerbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/fracdiving/maxdiveavgquot = 0 # maximal UBQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.1] heuristics/fracdiving/maxdiveubquotnosol = 0.1 # maximal AVGQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/fracdiving/maxdiveavgquotnosol = 0 # use one level of backtracking if infeasibility is encountered? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/fracdiving/backtrack = TRUE # percentage of immediate domain changes during probing to trigger LP resolve # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.15] heuristics/fracdiving/lpresolvedomchgquot = 0.15 # LP solve frequency for diving heuristics (0: only after enough domain changes have been found) # [type: int, advanced: FALSE, range: [0,2147483647], default: 0] heuristics/fracdiving/lpsolvefreq = 0 # should only LP branching candidates be considered instead of the slower but more general constraint handler diving variable selection? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/fracdiving/onlylpbranchcands = FALSE # priority of heuristic <gins> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1103000] heuristics/gins/priority = -1103000 # frequency for calling primal heuristic <gins> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 20] heuristics/gins/freq = 20 # frequency offset for calling primal heuristic <gins> # [type: int, advanced: FALSE, range: [0,1073741822], default: 8] heuristics/gins/freqofs = 8 # maximal depth level to call primal heuristic <gins> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/gins/maxdepth = -1 # number of nodes added to the contingent of the total nodes # [type: int, advanced: FALSE, range: [0,2147483647], default: 500] heuristics/gins/nodesofs = 500 # maximum number of nodes to regard in the subproblem # [type: int, advanced: TRUE, range: [0,2147483647], default: 5000] heuristics/gins/maxnodes = 5000 # minimum number of nodes required to start the subproblem # [type: int, advanced: TRUE, range: [0,2147483647], default: 50] heuristics/gins/minnodes = 50 # number of nodes without incumbent change that heuristic should wait # [type: int, advanced: TRUE, range: [0,2147483647], default: 100] heuristics/gins/nwaitingnodes = 100 # contingent of sub problem nodes in relation to the number of nodes of the original problem # [type: real, advanced: FALSE, range: [0,1], default: 0.15] heuristics/gins/nodesquot = 0.15 # percentage of integer variables that have to be fixed # [type: real, advanced: FALSE, range: [1e-06,0.999999], default: 0.66] heuristics/gins/minfixingrate = 0.66 # factor by which gins should at least improve the incumbent # [type: real, advanced: TRUE, range: [0,1], default: 0.01] heuristics/gins/minimprove = 0.01 # should subproblem be created out of the rows in the LP rows? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/gins/uselprows = FALSE # if uselprows == FALSE, should all active cuts from cutpool be copied to constraints in subproblem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/gins/copycuts = TRUE # should continuous variables outside the neighborhoods be fixed? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/gins/fixcontvars = FALSE # limit on number of improving incumbent solutions in sub-CIP # [type: int, advanced: FALSE, range: [-1,2147483647], default: 3] heuristics/gins/bestsollimit = 3 # maximum distance to selected variable to enter the subproblem, or -1 to select the distance that best approximates the minimum fixing rate from below # [type: int, advanced: FALSE, range: [-1,2147483647], default: 3] heuristics/gins/maxdistance = 3 # the reference point to compute the neighborhood potential: (r)oot, (l)ocal lp, or (p)seudo solution # [type: char, advanced: TRUE, range: {lpr}, default: r] heuristics/gins/potential = r # should the heuristic solve a sequence of sub-MIP's around the first selected variable # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/gins/userollinghorizon = TRUE # should dense constraints (at least as dense as 1 - minfixingrate) be ignored by connectivity graph? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/gins/relaxdenseconss = FALSE # limiting percentage for variables already used in sub-SCIPs to terminate rolling horizon approach # [type: real, advanced: TRUE, range: [0,1], default: 0.4] heuristics/gins/rollhorizonlimfac = 0.4 # overlap of blocks between runs - 0.0: no overlap, 1.0: shift by only 1 block # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/gins/overlap = 0 # should user decompositions be considered, if available? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/gins/usedecomp = TRUE # should user decompositions be considered for initial selection in rolling horizon, if available? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/gins/usedecomprollhorizon = FALSE # should random initial variable selection be used if decomposition was not successful? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/gins/useselfallback = TRUE # should blocks be treated consecutively (sorted by ascending label?) # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/gins/consecutiveblocks = TRUE # priority of heuristic <guideddiving> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1007000] heuristics/guideddiving/priority = -1007000 # frequency for calling primal heuristic <guideddiving> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] heuristics/guideddiving/freq = 10 # frequency offset for calling primal heuristic <guideddiving> # [type: int, advanced: FALSE, range: [0,1073741822], default: 7] heuristics/guideddiving/freqofs = 7 # maximal depth level to call primal heuristic <guideddiving> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/guideddiving/maxdepth = -1 # minimal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/guideddiving/minreldepth = 0 # maximal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/guideddiving/maxreldepth = 1 # maximal fraction of diving LP iterations compared to node LP iterations # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.05] heuristics/guideddiving/maxlpiterquot = 0.05 # additional number of allowed LP iterations # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] heuristics/guideddiving/maxlpiterofs = 1000 # maximal quotient (curlowerbound - lowerbound)/(cutoffbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.8] heuristics/guideddiving/maxdiveubquot = 0.8 # maximal quotient (curlowerbound - lowerbound)/(avglowerbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/guideddiving/maxdiveavgquot = 0 # maximal UBQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/guideddiving/maxdiveubquotnosol = 1 # maximal AVGQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 1] heuristics/guideddiving/maxdiveavgquotnosol = 1 # use one level of backtracking if infeasibility is encountered? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/guideddiving/backtrack = TRUE # percentage of immediate domain changes during probing to trigger LP resolve # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.15] heuristics/guideddiving/lpresolvedomchgquot = 0.15 # LP solve frequency for diving heuristics (0: only after enough domain changes have been found) # [type: int, advanced: FALSE, range: [0,2147483647], default: 0] heuristics/guideddiving/lpsolvefreq = 0 # should only LP branching candidates be considered instead of the slower but more general constraint handler diving variable selection? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/guideddiving/onlylpbranchcands = FALSE # priority of heuristic <zeroobj> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 100] heuristics/zeroobj/priority = 100 # frequency for calling primal heuristic <zeroobj> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] heuristics/zeroobj/freq = -1 # frequency offset for calling primal heuristic <zeroobj> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/zeroobj/freqofs = 0 # maximal depth level to call primal heuristic <zeroobj> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 0] heuristics/zeroobj/maxdepth = 0 # maximum number of nodes to regard in the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 1000] heuristics/zeroobj/maxnodes = 1000 # number of nodes added to the contingent of the total nodes # [type: longint, advanced: FALSE, range: [0,9223372036854775807], default: 100] heuristics/zeroobj/nodesofs = 100 # minimum number of nodes required to start the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 100] heuristics/zeroobj/minnodes = 100 # maximum number of LP iterations to be performed in the subproblem # [type: longint, advanced: TRUE, range: [-1,9223372036854775807], default: 5000] heuristics/zeroobj/maxlpiters = 5000 # contingent of sub problem nodes in relation to the number of nodes of the original problem # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/zeroobj/nodesquot = 0.1 # factor by which zeroobj should at least improve the incumbent # [type: real, advanced: TRUE, range: [0,1], default: 0.01] heuristics/zeroobj/minimprove = 0.01 # should all subproblem solutions be added to the original SCIP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/zeroobj/addallsols = FALSE # should heuristic only be executed if no primal solution was found, yet? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/zeroobj/onlywithoutsol = TRUE # should uct node selection be used at the beginning of the search? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/zeroobj/useuct = FALSE # priority of heuristic <indicator> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -20200] heuristics/indicator/priority = -20200 # frequency for calling primal heuristic <indicator> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] heuristics/indicator/freq = 1 # frequency offset for calling primal heuristic <indicator> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/indicator/freqofs = 0 # maximal depth level to call primal heuristic <indicator> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/indicator/maxdepth = -1 # whether the one-opt heuristic should be started # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/indicator/oneopt = FALSE # Try to improve other solutions by one-opt? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/indicator/improvesols = FALSE # priority of heuristic <indicatordiving> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -150000] heuristics/indicatordiving/priority = -150000 # frequency for calling primal heuristic <indicatordiving> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] heuristics/indicatordiving/freq = 0 # frequency offset for calling primal heuristic <indicatordiving> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/indicatordiving/freqofs = 0 # maximal depth level to call primal heuristic <indicatordiving> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/indicatordiving/maxdepth = -1 # minimal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/indicatordiving/minreldepth = 0 # maximal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/indicatordiving/maxreldepth = 1 # maximal fraction of diving LP iterations compared to node LP iterations # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.05] heuristics/indicatordiving/maxlpiterquot = 0.05 # additional number of allowed LP iterations # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] heuristics/indicatordiving/maxlpiterofs = 1000 # maximal quotient (curlowerbound - lowerbound)/(cutoffbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.8] heuristics/indicatordiving/maxdiveubquot = 0.8 # maximal quotient (curlowerbound - lowerbound)/(avglowerbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/indicatordiving/maxdiveavgquot = 0 # maximal UBQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.1] heuristics/indicatordiving/maxdiveubquotnosol = 0.1 # maximal AVGQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/indicatordiving/maxdiveavgquotnosol = 0 # use one level of backtracking if infeasibility is encountered? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/indicatordiving/backtrack = TRUE # percentage of immediate domain changes during probing to trigger LP resolve # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.15] heuristics/indicatordiving/lpresolvedomchgquot = 0.15 # LP solve frequency for diving heuristics (0: only after enough domain changes have been found) # [type: int, advanced: FALSE, range: [0,2147483647], default: 30] heuristics/indicatordiving/lpsolvefreq = 30 # should only LP branching candidates be considered instead of the slower but more general constraint handler diving variable selection? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/indicatordiving/onlylpbranchcands = FALSE # in violation case all fractional below this value are fixed to constant # [type: real, advanced: FALSE, range: [0,1], default: 0.5] heuristics/indicatordiving/roundingfrac = 0.5 # decides which roundingmode is selected (0: conservative, 1: aggressive) # [type: int, advanced: FALSE, range: [0,1], default: 0] heuristics/indicatordiving/roundingmode = 0 # which values of semi-continuous variables should get a high score? (0: low, 1: middle, 2: high) # [type: int, advanced: FALSE, range: [0,2], default: 0] heuristics/indicatordiving/semicontscoremode = 0 # should varbound constraints be considered? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/indicatordiving/usevarbounds = TRUE # should heur run if there are no indicator constraints modeling semicont. vars? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/indicatordiving/runwithoutscinds = FALSE # priority of heuristic <intdiving> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1003500] heuristics/intdiving/priority = -1003500 # frequency for calling primal heuristic <intdiving> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] heuristics/intdiving/freq = -1 # frequency offset for calling primal heuristic <intdiving> # [type: int, advanced: FALSE, range: [0,1073741822], default: 9] heuristics/intdiving/freqofs = 9 # maximal depth level to call primal heuristic <intdiving> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/intdiving/maxdepth = -1 # minimal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/intdiving/minreldepth = 0 # maximal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/intdiving/maxreldepth = 1 # maximal fraction of diving LP iterations compared to node LP iterations # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.05] heuristics/intdiving/maxlpiterquot = 0.05 # additional number of allowed LP iterations # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] heuristics/intdiving/maxlpiterofs = 1000 # maximal quotient (curlowerbound - lowerbound)/(cutoffbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.8] heuristics/intdiving/maxdiveubquot = 0.8 # maximal quotient (curlowerbound - lowerbound)/(avglowerbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/intdiving/maxdiveavgquot = 0 # maximal UBQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.1] heuristics/intdiving/maxdiveubquotnosol = 0.1 # maximal AVGQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/intdiving/maxdiveavgquotnosol = 0 # use one level of backtracking if infeasibility is encountered? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/intdiving/backtrack = TRUE # priority of heuristic <intshifting> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -10000] heuristics/intshifting/priority = -10000 # frequency for calling primal heuristic <intshifting> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] heuristics/intshifting/freq = 10 # frequency offset for calling primal heuristic <intshifting> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/intshifting/freqofs = 0 # maximal depth level to call primal heuristic <intshifting> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/intshifting/maxdepth = -1 # priority of heuristic <linesearchdiving> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1006000] heuristics/linesearchdiving/priority = -1006000 # frequency for calling primal heuristic <linesearchdiving> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] heuristics/linesearchdiving/freq = 10 # frequency offset for calling primal heuristic <linesearchdiving> # [type: int, advanced: FALSE, range: [0,1073741822], default: 6] heuristics/linesearchdiving/freqofs = 6 # maximal depth level to call primal heuristic <linesearchdiving> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/linesearchdiving/maxdepth = -1 # minimal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/linesearchdiving/minreldepth = 0 # maximal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/linesearchdiving/maxreldepth = 1 # maximal fraction of diving LP iterations compared to node LP iterations # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.05] heuristics/linesearchdiving/maxlpiterquot = 0.05 # additional number of allowed LP iterations # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] heuristics/linesearchdiving/maxlpiterofs = 1000 # maximal quotient (curlowerbound - lowerbound)/(cutoffbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.8] heuristics/linesearchdiving/maxdiveubquot = 0.8 # maximal quotient (curlowerbound - lowerbound)/(avglowerbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/linesearchdiving/maxdiveavgquot = 0 # maximal UBQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.1] heuristics/linesearchdiving/maxdiveubquotnosol = 0.1 # maximal AVGQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/linesearchdiving/maxdiveavgquotnosol = 0 # use one level of backtracking if infeasibility is encountered? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/linesearchdiving/backtrack = TRUE # percentage of immediate domain changes during probing to trigger LP resolve # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.15] heuristics/linesearchdiving/lpresolvedomchgquot = 0.15 # LP solve frequency for diving heuristics (0: only after enough domain changes have been found) # [type: int, advanced: FALSE, range: [0,2147483647], default: 0] heuristics/linesearchdiving/lpsolvefreq = 0 # should only LP branching candidates be considered instead of the slower but more general constraint handler diving variable selection? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/linesearchdiving/onlylpbranchcands = FALSE # priority of heuristic <localbranching> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1102000] heuristics/localbranching/priority = -1102000 # frequency for calling primal heuristic <localbranching> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] heuristics/localbranching/freq = -1 # frequency offset for calling primal heuristic <localbranching> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/localbranching/freqofs = 0 # maximal depth level to call primal heuristic <localbranching> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/localbranching/maxdepth = -1 # number of nodes added to the contingent of the total nodes # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] heuristics/localbranching/nodesofs = 1000 # radius (using Manhattan metric) of the incumbent's neighborhood to be searched # [type: int, advanced: FALSE, range: [1,2147483647], default: 18] heuristics/localbranching/neighborhoodsize = 18 # contingent of sub problem nodes in relation to the number of nodes of the original problem # [type: real, advanced: FALSE, range: [0,1], default: 0.05] heuristics/localbranching/nodesquot = 0.05 # factor by which the limit on the number of LP depends on the node limit # [type: real, advanced: TRUE, range: [1,1.79769313486232e+308], default: 1.5] heuristics/localbranching/lplimfac = 1.5 # minimum number of nodes required to start the subproblem # [type: int, advanced: TRUE, range: [0,2147483647], default: 1000] heuristics/localbranching/minnodes = 1000 # maximum number of nodes to regard in the subproblem # [type: int, advanced: TRUE, range: [0,2147483647], default: 10000] heuristics/localbranching/maxnodes = 10000 # number of nodes without incumbent change that heuristic should wait # [type: int, advanced: TRUE, range: [0,2147483647], default: 200] heuristics/localbranching/nwaitingnodes = 200 # factor by which localbranching should at least improve the incumbent # [type: real, advanced: TRUE, range: [0,1], default: 0.01] heuristics/localbranching/minimprove = 0.01 # should subproblem be created out of the rows in the LP rows? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/localbranching/uselprows = FALSE # if uselprows == FALSE, should all active cuts from cutpool be copied to constraints in subproblem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/localbranching/copycuts = TRUE # limit on number of improving incumbent solutions in sub-CIP # [type: int, advanced: FALSE, range: [-1,2147483647], default: 3] heuristics/localbranching/bestsollimit = 3 # priority of heuristic <locks> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 3000] heuristics/locks/priority = 3000 # frequency for calling primal heuristic <locks> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] heuristics/locks/freq = 0 # frequency offset for calling primal heuristic <locks> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/locks/freqofs = 0 # maximal depth level to call primal heuristic <locks> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/locks/maxdepth = -1 # maximum number of propagation rounds to be performed in each propagation call (-1: no limit, -2: parameter settings) # [type: int, advanced: TRUE, range: [-2,2147483647], default: 2] heuristics/locks/maxproprounds = 2 # minimum percentage of integer variables that have to be fixable # [type: real, advanced: FALSE, range: [0,1], default: 0.65] heuristics/locks/minfixingrate = 0.65 # probability for rounding a variable up in case of ties # [type: real, advanced: FALSE, range: [0,1], default: 0.67] heuristics/locks/roundupprobability = 0.67 # should a final sub-MIP be solved to costruct a feasible solution if the LP was not roundable? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/locks/usefinalsubmip = TRUE # maximum number of nodes to regard in the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 5000] heuristics/locks/maxnodes = 5000 # number of nodes added to the contingent of the total nodes # [type: longint, advanced: FALSE, range: [0,9223372036854775807], default: 500] heuristics/locks/nodesofs = 500 # minimum number of nodes required to start the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 500] heuristics/locks/minnodes = 500 # contingent of sub problem nodes in relation to the number of nodes of the original problem # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/locks/nodesquot = 0.1 # factor by which locks heuristic should at least improve the incumbent # [type: real, advanced: TRUE, range: [0,1], default: 0.01] heuristics/locks/minimprove = 0.01 # should all active cuts from cutpool be copied to constraints in subproblem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/locks/copycuts = TRUE # should the locks be updated based on LP rows? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/locks/updatelocks = TRUE # minimum fixing rate over all variables (including continuous) to solve LP # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/locks/minfixingratelp = 0 # priority of heuristic <lpface> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1104010] heuristics/lpface/priority = -1104010 # frequency for calling primal heuristic <lpface> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 15] heuristics/lpface/freq = 15 # frequency offset for calling primal heuristic <lpface> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/lpface/freqofs = 0 # maximal depth level to call primal heuristic <lpface> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/lpface/maxdepth = -1 # number of nodes added to the contingent of the total nodes # [type: longint, advanced: FALSE, range: [0,9223372036854775807], default: 200] heuristics/lpface/nodesofs = 200 # maximum number of nodes to regard in the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 5000] heuristics/lpface/maxnodes = 5000 # minimum number of nodes required to start the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 50] heuristics/lpface/minnodes = 50 # contingent of sub problem nodes in relation to the number of nodes of the original problem # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/lpface/nodesquot = 0.1 # required percentage of fixed integer variables in sub-MIP to run # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/lpface/minfixingrate = 0.1 # factor by which the limit on the number of LP depends on the node limit # [type: real, advanced: TRUE, range: [1,1.79769313486232e+308], default: 2] heuristics/lpface/lplimfac = 2 # should subproblem be created out of the rows in the LP rows? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/lpface/uselprows = TRUE # should dually nonbasic rows be turned into equations? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/lpface/dualbasisequations = FALSE # should the heuristic continue solving the same sub-SCIP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/lpface/keepsubscip = FALSE # if uselprows == FALSE, should all active cuts from cutpool be copied to constraints in subproblem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/lpface/copycuts = TRUE # objective function in the sub-SCIP: (z)ero, (r)oot-LP-difference, (i)nference, LP (f)ractionality, (o)riginal # [type: char, advanced: TRUE, range: {forzi}, default: z] heuristics/lpface/subscipobjective = z # the minimum active search tree path length along which lower bound hasn't changed before heuristic becomes active # [type: int, advanced: TRUE, range: [0,65531], default: 5] heuristics/lpface/minpathlen = 5 # priority of heuristic <alns> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1100500] heuristics/alns/priority = -1100500 # frequency for calling primal heuristic <alns> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 20] heuristics/alns/freq = 20 # frequency offset for calling primal heuristic <alns> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/alns/freqofs = 0 # maximal depth level to call primal heuristic <alns> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/alns/maxdepth = -1 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/alns/rens/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/alns/rens/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/alns/rens/active = TRUE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/alns/rens/priority = 1 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/alns/rins/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/alns/rins/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/alns/rins/active = TRUE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/alns/rins/priority = 1 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/alns/mutation/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/alns/mutation/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/alns/mutation/active = TRUE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/alns/mutation/priority = 1 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/alns/localbranching/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/alns/localbranching/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/alns/localbranching/active = TRUE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/alns/localbranching/priority = 1 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/alns/crossover/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/alns/crossover/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/alns/crossover/active = TRUE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/alns/crossover/priority = 1 # the number of solutions that crossover should combine # [type: int, advanced: TRUE, range: [2,10], default: 2] heuristics/alns/crossover/nsols = 2 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/alns/proximity/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/alns/proximity/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/alns/proximity/active = TRUE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/alns/proximity/priority = 1 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/alns/zeroobjective/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/alns/zeroobjective/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/alns/zeroobjective/active = TRUE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/alns/zeroobjective/priority = 1 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/alns/dins/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/alns/dins/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/alns/dins/active = TRUE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/alns/dins/priority = 1 # number of pool solutions where binary solution values must agree # [type: int, advanced: TRUE, range: [1,100], default: 5] heuristics/alns/dins/npoolsols = 5 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/alns/trustregion/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/alns/trustregion/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/alns/trustregion/active = FALSE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/alns/trustregion/priority = 1 # the penalty for each change in the binary variables from the candidate solution # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 100] heuristics/alns/trustregion/violpenalty = 100 # show statistics on neighborhoods? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/alns/shownbstats = FALSE # maximum number of nodes to regard in the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 5000] heuristics/alns/maxnodes = 5000 # offset added to the nodes budget # [type: longint, advanced: FALSE, range: [0,9223372036854775807], default: 500] heuristics/alns/nodesofs = 500 # minimum number of nodes required to start a sub-SCIP # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 50] heuristics/alns/minnodes = 50 # number of nodes since last incumbent solution that the heuristic should wait # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 25] heuristics/alns/waitingnodes = 25 # fraction of nodes compared to the main SCIP for budget computation # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/alns/nodesquot = 0.1 # lower bound fraction of nodes compared to the main SCIP for budget computation # [type: real, advanced: FALSE, range: [0,1], default: 0] heuristics/alns/nodesquotmin = 0 # initial factor by which ALNS should at least improve the incumbent # [type: real, advanced: TRUE, range: [0,1], default: 0.01] heuristics/alns/startminimprove = 0.01 # lower threshold for the minimal improvement over the incumbent # [type: real, advanced: TRUE, range: [0,1], default: 0.01] heuristics/alns/minimprovelow = 0.01 # upper bound for the minimal improvement over the incumbent # [type: real, advanced: TRUE, range: [0,1], default: 0.01] heuristics/alns/minimprovehigh = 0.01 # limit on the number of improving solutions in a sub-SCIP call # [type: int, advanced: FALSE, range: [-1,2147483647], default: 3] heuristics/alns/nsolslim = 3 # the bandit algorithm: (u)pper confidence bounds, (e)xp.3, epsilon (g)reedy, exp.3-(i)x # [type: char, advanced: TRUE, range: {uegi}, default: i] heuristics/alns/banditalgo = i # weight between uniform (gamma ~ 1) and weight driven (gamma ~ 0) probability distribution for exp3 # [type: real, advanced: TRUE, range: [0,1], default: 0.07041455] heuristics/alns/gamma = 0.07041455 # reward offset between 0 and 1 at every observation for Exp.3 # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/alns/beta = 0 # parameter to increase the confidence width in UCB # [type: real, advanced: TRUE, range: [0,100], default: 0.0016] heuristics/alns/alpha = 0.0016 # distances from fixed variables be used for variable prioritization # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/alns/usedistances = TRUE # should reduced cost scores be used for variable prioritization? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/alns/useredcost = TRUE # should the ALNS heuristic do more fixings by itself based on variable prioritization until the target fixing rate is reached? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/alns/domorefixings = TRUE # should the heuristic adjust the target fixing rate based on the success? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/alns/adjustfixingrate = TRUE # should the heuristic activate other sub-SCIP heuristics during its search? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/alns/usesubscipheurs = FALSE # reward control to increase the weight of the simple solution indicator and decrease the weight of the closed gap reward # [type: real, advanced: TRUE, range: [0,1], default: 0.8] heuristics/alns/rewardcontrol = 0.8 # factor by which target node number is eventually increased # [type: real, advanced: TRUE, range: [1,100000], default: 1.05] heuristics/alns/targetnodefactor = 1.05 # initial random seed for bandit algorithms and random decisions by neighborhoods # [type: int, advanced: FALSE, range: [0,2147483647], default: 113] heuristics/alns/seed = 113 # number of allowed executions of the heuristic on the same incumbent solution (-1: no limit, 0: number of active neighborhoods) # [type: int, advanced: TRUE, range: [-1,100], default: -1] heuristics/alns/maxcallssamesol = -1 # should the factor by which the minimum improvement is bound be dynamically updated? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/alns/adjustminimprove = FALSE # should the target nodes be dynamically adjusted? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/alns/adjusttargetnodes = TRUE # increase exploration in epsilon-greedy bandit algorithm # [type: real, advanced: TRUE, range: [0,1], default: 0.4685844] heuristics/alns/eps = 0.4685844 # the reward baseline to separate successful and failed calls # [type: real, advanced: TRUE, range: [0,0.99], default: 0.5] heuristics/alns/rewardbaseline = 0.5 # should the bandit algorithms be reset when a new problem is read? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/alns/resetweights = TRUE # file name to store all rewards and the selection of the bandit # [type: string, advanced: TRUE, default: "-"] heuristics/alns/rewardfilename = "-" # should random seeds of sub-SCIPs be altered to increase diversification? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/alns/subsciprandseeds = FALSE # should the reward be scaled by the effort? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/alns/scalebyeffort = TRUE # should cutting planes be copied to the sub-SCIP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/alns/copycuts = FALSE # tolerance by which the fixing rate may be missed without generic fixing # [type: real, advanced: TRUE, range: [0,1], default: 0.1] heuristics/alns/fixtol = 0.1 # tolerance by which the fixing rate may be exceeded without generic unfixing # [type: real, advanced: TRUE, range: [0,1], default: 0.1] heuristics/alns/unfixtol = 0.1 # should local reduced costs be used for generic (un)fixing? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/alns/uselocalredcost = FALSE # should pseudo cost scores be used for variable priorization? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/alns/usepscost = TRUE # should the heuristic be executed multiple times during the root node? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/alns/initduringroot = FALSE # is statistics table <neighborhood> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/neighborhood/active = TRUE # priority of heuristic <nlpdiving> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1003010] heuristics/nlpdiving/priority = -1003010 # frequency for calling primal heuristic <nlpdiving> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] heuristics/nlpdiving/freq = 10 # frequency offset for calling primal heuristic <nlpdiving> # [type: int, advanced: FALSE, range: [0,1073741822], default: 3] heuristics/nlpdiving/freqofs = 3 # maximal depth level to call primal heuristic <nlpdiving> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/nlpdiving/maxdepth = -1 # minimal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/nlpdiving/minreldepth = 0 # maximal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/nlpdiving/maxreldepth = 1 # minimial absolute number of allowed NLP iterations # [type: int, advanced: FALSE, range: [0,2147483647], default: 200] heuristics/nlpdiving/maxnlpiterabs = 200 # additional allowed number of NLP iterations relative to successfully found solutions # [type: int, advanced: FALSE, range: [0,2147483647], default: 10] heuristics/nlpdiving/maxnlpiterrel = 10 # maximal quotient (curlowerbound - lowerbound)/(cutoffbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.8] heuristics/nlpdiving/maxdiveubquot = 0.8 # maximal quotient (curlowerbound - lowerbound)/(avglowerbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/nlpdiving/maxdiveavgquot = 0 # maximal UBQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.1] heuristics/nlpdiving/maxdiveubquotnosol = 0.1 # maximal AVGQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/nlpdiving/maxdiveavgquotnosol = 0 # maximal number of NLPs with feasible solution to solve during one dive # [type: int, advanced: FALSE, range: [1,2147483647], default: 10] heuristics/nlpdiving/maxfeasnlps = 10 # use one level of backtracking if infeasibility is encountered? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/nlpdiving/backtrack = TRUE # should the LP relaxation be solved before the NLP relaxation? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/nlpdiving/lp = FALSE # prefer variables that are also fractional in LP solution? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/nlpdiving/preferlpfracs = FALSE # heuristic will not run if less then this percentage of calls succeeded (0.0: no limit) # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/nlpdiving/minsuccquot = 0.1 # percentage of fractional variables that should be fixed before the next NLP solve # [type: real, advanced: FALSE, range: [0,1], default: 0.2] heuristics/nlpdiving/fixquot = 0.2 # should variables in a minimal cover be preferred? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/nlpdiving/prefercover = TRUE # should a sub-MIP be solved if all cover variables are fixed? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/nlpdiving/solvesubmip = FALSE # should the NLP solver stop early if it converges slow? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/nlpdiving/nlpfastfail = TRUE # which point should be used as starting point for the NLP solver? ('n'one, last 'f'easible, from dive's'tart) # [type: char, advanced: TRUE, range: {fns}, default: s] heuristics/nlpdiving/nlpstart = s # which variable selection should be used? ('f'ractionality, 'c'oefficient, 'p'seudocost, 'g'uided, 'd'ouble, 'v'eclen) # [type: char, advanced: FALSE, range: {fcpgdv}, default: d] heuristics/nlpdiving/varselrule = d # priority of heuristic <mutation> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1103010] heuristics/mutation/priority = -1103010 # frequency for calling primal heuristic <mutation> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] heuristics/mutation/freq = -1 # frequency offset for calling primal heuristic <mutation> # [type: int, advanced: FALSE, range: [0,1073741822], default: 8] heuristics/mutation/freqofs = 8 # maximal depth level to call primal heuristic <mutation> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/mutation/maxdepth = -1 # number of nodes added to the contingent of the total nodes # [type: int, advanced: FALSE, range: [0,2147483647], default: 500] heuristics/mutation/nodesofs = 500 # maximum number of nodes to regard in the subproblem # [type: int, advanced: TRUE, range: [0,2147483647], default: 5000] heuristics/mutation/maxnodes = 5000 # minimum number of nodes required to start the subproblem # [type: int, advanced: TRUE, range: [0,2147483647], default: 500] heuristics/mutation/minnodes = 500 # number of nodes without incumbent change that heuristic should wait # [type: int, advanced: TRUE, range: [0,2147483647], default: 200] heuristics/mutation/nwaitingnodes = 200 # contingent of sub problem nodes in relation to the number of nodes of the original problem # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/mutation/nodesquot = 0.1 # percentage of integer variables that have to be fixed # [type: real, advanced: FALSE, range: [1e-06,0.999999], default: 0.8] heuristics/mutation/minfixingrate = 0.8 # factor by which mutation should at least improve the incumbent # [type: real, advanced: TRUE, range: [0,1], default: 0.01] heuristics/mutation/minimprove = 0.01 # should subproblem be created out of the rows in the LP rows? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/mutation/uselprows = FALSE # if uselprows == FALSE, should all active cuts from cutpool be copied to constraints in subproblem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/mutation/copycuts = TRUE # limit on number of improving incumbent solutions in sub-CIP # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] heuristics/mutation/bestsollimit = -1 # should uct node selection be used at the beginning of the search? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/mutation/useuct = FALSE # priority of heuristic <multistart> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -2100000] heuristics/multistart/priority = -2100000 # frequency for calling primal heuristic <multistart> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] heuristics/multistart/freq = 0 # frequency offset for calling primal heuristic <multistart> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/multistart/freqofs = 0 # maximal depth level to call primal heuristic <multistart> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/multistart/maxdepth = -1 # number of random points generated per execution call # [type: int, advanced: FALSE, range: [0,2147483647], default: 100] heuristics/multistart/nrndpoints = 100 # maximum variable domain size for unbounded variables # [type: real, advanced: FALSE, range: [0,1e+20], default: 20000] heuristics/multistart/maxboundsize = 20000 # number of iterations to reduce the maximum violation of a point # [type: int, advanced: FALSE, range: [0,2147483647], default: 300] heuristics/multistart/maxiter = 300 # minimum required improving factor to proceed in improvement of a single point # [type: real, advanced: FALSE, range: [-1e+20,1e+20], default: 0.05] heuristics/multistart/minimprfac = 0.05 # number of iteration when checking the minimum improvement # [type: int, advanced: FALSE, range: [1,2147483647], default: 10] heuristics/multistart/minimpriter = 10 # maximum distance between two points in the same cluster # [type: real, advanced: FALSE, range: [0,1e+20], default: 0.15] heuristics/multistart/maxreldist = 0.15 # limit for gradient computations for all improvePoint() calls (0 for no limit) # [type: real, advanced: FALSE, range: [0,1e+20], default: 5000000] heuristics/multistart/gradlimit = 5000000 # maximum number of considered clusters per heuristic call # [type: int, advanced: FALSE, range: [0,2147483647], default: 3] heuristics/multistart/maxncluster = 3 # should the heuristic run only on continuous problems? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/multistart/onlynlps = TRUE # priority of heuristic <mpec> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -2050000] heuristics/mpec/priority = -2050000 # frequency for calling primal heuristic <mpec> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 50] heuristics/mpec/freq = 50 # frequency offset for calling primal heuristic <mpec> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/mpec/freqofs = 0 # maximal depth level to call primal heuristic <mpec> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/mpec/maxdepth = -1 # initial regularization right-hand side value # [type: real, advanced: FALSE, range: [0,0.25], default: 0.125] heuristics/mpec/inittheta = 0.125 # regularization update factor # [type: real, advanced: FALSE, range: [0,1], default: 0.5] heuristics/mpec/sigma = 0.5 # maximum number of NLP iterations per solve # [type: real, advanced: FALSE, range: [0,1], default: 0.001] heuristics/mpec/subnlptrigger = 0.001 # maximum cost available for solving NLPs per call of the heuristic # [type: real, advanced: FALSE, range: [0,1e+20], default: 100000000] heuristics/mpec/maxnlpcost = 100000000 # factor by which heuristic should at least improve the incumbent # [type: real, advanced: FALSE, range: [0,1], default: 0.01] heuristics/mpec/minimprove = 0.01 # minimum amount of gap left in order to call the heuristic # [type: real, advanced: FALSE, range: [0,1e+20], default: 0.05] heuristics/mpec/mingapleft = 0.05 # maximum number of iterations of the MPEC loop # [type: int, advanced: FALSE, range: [0,2147483647], default: 100] heuristics/mpec/maxiter = 100 # maximum number of NLP iterations per solve # [type: int, advanced: FALSE, range: [0,2147483647], default: 500] heuristics/mpec/maxnlpiter = 500 # maximum number of consecutive calls for which the heuristic did not find an improving solution # [type: int, advanced: FALSE, range: [0,2147483647], default: 10] heuristics/mpec/maxnunsucc = 10 # priority of heuristic <objpscostdiving> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1004000] heuristics/objpscostdiving/priority = -1004000 # frequency for calling primal heuristic <objpscostdiving> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 20] heuristics/objpscostdiving/freq = 20 # frequency offset for calling primal heuristic <objpscostdiving> # [type: int, advanced: FALSE, range: [0,1073741822], default: 4] heuristics/objpscostdiving/freqofs = 4 # maximal depth level to call primal heuristic <objpscostdiving> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/objpscostdiving/maxdepth = -1 # minimal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/objpscostdiving/minreldepth = 0 # maximal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/objpscostdiving/maxreldepth = 1 # maximal fraction of diving LP iterations compared to total iteration number # [type: real, advanced: FALSE, range: [0,1], default: 0.01] heuristics/objpscostdiving/maxlpiterquot = 0.01 # additional number of allowed LP iterations # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] heuristics/objpscostdiving/maxlpiterofs = 1000 # total number of feasible solutions found up to which heuristic is called (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] heuristics/objpscostdiving/maxsols = -1 # maximal diving depth: number of binary/integer variables times depthfac # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.5] heuristics/objpscostdiving/depthfac = 0.5 # maximal diving depth factor if no feasible solution was found yet # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 2] heuristics/objpscostdiving/depthfacnosol = 2 # priority of heuristic <octane> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1008000] heuristics/octane/priority = -1008000 # frequency for calling primal heuristic <octane> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] heuristics/octane/freq = -1 # frequency offset for calling primal heuristic <octane> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/octane/freqofs = 0 # maximal depth level to call primal heuristic <octane> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/octane/maxdepth = -1 # number of 0-1-points to be tested as possible solutions by OCTANE # [type: int, advanced: TRUE, range: [1,2147483647], default: 100] heuristics/octane/fmax = 100 # number of 0-1-points to be tested at first whether they violate a common row # [type: int, advanced: TRUE, range: [1,2147483647], default: 10] heuristics/octane/ffirst = 10 # execute OCTANE only in the space of fractional variables (TRUE) or in the full space? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/octane/usefracspace = TRUE # should the inner normal of the objective be used as one ray direction? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/octane/useobjray = TRUE # should the average of the basic cone be used as one ray direction? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/octane/useavgray = TRUE # should the difference between the root solution and the current LP solution be used as one ray direction? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/octane/usediffray = FALSE # should the weighted average of the basic cone be used as one ray direction? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/octane/useavgwgtray = TRUE # should the weighted average of the nonbasic cone be used as one ray direction? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/octane/useavgnbray = TRUE # priority of heuristic <ofins> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 60000] heuristics/ofins/priority = 60000 # frequency for calling primal heuristic <ofins> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] heuristics/ofins/freq = 0 # frequency offset for calling primal heuristic <ofins> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/ofins/freqofs = 0 # maximal depth level to call primal heuristic <ofins> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 0] heuristics/ofins/maxdepth = 0 # maximum number of nodes to regard in the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 5000] heuristics/ofins/maxnodes = 5000 # minimum number of nodes required to start the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 50] heuristics/ofins/minnodes = 50 # maximal rate of changed coefficients # [type: real, advanced: FALSE, range: [0,1], default: 0.5] heuristics/ofins/maxchangerate = 0.5 # maximal rate of change per coefficient to get fixed # [type: real, advanced: FALSE, range: [0,1], default: 0.04] heuristics/ofins/maxchange = 0.04 # should all active cuts from cutpool be copied to constraints in subproblem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/ofins/copycuts = TRUE # should all subproblem solutions be added to the original SCIP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/ofins/addallsols = FALSE # number of nodes added to the contingent of the total nodes # [type: longint, advanced: FALSE, range: [0,9223372036854775807], default: 500] heuristics/ofins/nodesofs = 500 # contingent of sub problem nodes in relation to the number of nodes of the original problem # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/ofins/nodesquot = 0.1 # factor by which RENS should at least improve the incumbent # [type: real, advanced: TRUE, range: [0,1], default: 0.01] heuristics/ofins/minimprove = 0.01 # factor by which the limit on the number of LP depends on the node limit # [type: real, advanced: TRUE, range: [1,1.79769313486232e+308], default: 2] heuristics/ofins/lplimfac = 2 # priority of heuristic <oneopt> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -20000] heuristics/oneopt/priority = -20000 # frequency for calling primal heuristic <oneopt> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] heuristics/oneopt/freq = 1 # frequency offset for calling primal heuristic <oneopt> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/oneopt/freqofs = 0 # maximal depth level to call primal heuristic <oneopt> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/oneopt/maxdepth = -1 # should the objective be weighted with the potential shifting value when sorting the shifting candidates? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/oneopt/weightedobj = TRUE # should the heuristic be called before and during the root node? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/oneopt/duringroot = TRUE # should the construction of the LP be forced even if LP solving is deactivated? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/oneopt/forcelpconstruction = FALSE # should the heuristic be called before presolving? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/oneopt/beforepresol = FALSE # should the heuristic continue to run as long as improvements are found? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/oneopt/useloop = TRUE # priority of heuristic <padm> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 70000] heuristics/padm/priority = 70000 # frequency for calling primal heuristic <padm> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] heuristics/padm/freq = 0 # frequency offset for calling primal heuristic <padm> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/padm/freqofs = 0 # maximal depth level to call primal heuristic <padm> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/padm/maxdepth = -1 # maximum number of nodes to regard in all subproblems # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 5000] heuristics/padm/maxnodes = 5000 # minimum number of nodes to regard in one subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 50] heuristics/padm/minnodes = 50 # factor to control nodelimits of subproblems # [type: real, advanced: TRUE, range: [0,0.99], default: 0.8] heuristics/padm/nodefac = 0.8 # maximal number of ADM iterations in each penalty loop # [type: int, advanced: TRUE, range: [1,100], default: 4] heuristics/padm/admiterations = 4 # maximal number of penalty iterations # [type: int, advanced: TRUE, range: [1,100000], default: 100] heuristics/padm/penaltyiterations = 100 # mipgap at start # [type: real, advanced: TRUE, range: [0,16], default: 2] heuristics/padm/gap = 2 # should the problem get reoptimized with the original objective function? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/padm/reoptimize = TRUE # enable sigmoid rescaling of penalty parameters # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/padm/scaling = TRUE # should linking constraints be assigned? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/padm/assignlinking = TRUE # should the original problem be used? This is only for testing and not recommended! # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/padm/original = FALSE # should the heuristic run before or after the processing of the node? (0: before, 1: after, 2: both) # [type: int, advanced: FALSE, range: [0,2], default: 0] heuristics/padm/timing = 0 # priority of heuristic <proximity> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -2000000] heuristics/proximity/priority = -2000000 # frequency for calling primal heuristic <proximity> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] heuristics/proximity/freq = -1 # frequency offset for calling primal heuristic <proximity> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/proximity/freqofs = 0 # maximal depth level to call primal heuristic <proximity> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/proximity/maxdepth = -1 # should subproblem be constructed based on LP row information? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/proximity/uselprows = FALSE # should the heuristic immediately run again on its newly found solution? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/proximity/restart = TRUE # should the heuristic solve a final LP in case of continuous objective variables? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/proximity/usefinallp = FALSE # maximum number of nodes to regard in the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 10000] heuristics/proximity/maxnodes = 10000 # number of nodes added to the contingent of the total nodes # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 50] heuristics/proximity/nodesofs = 50 # minimum number of nodes required to start the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 1] heuristics/proximity/minnodes = 1 # maximum number of LP iterations to be performed in the subproblem # [type: longint, advanced: TRUE, range: [-1,9223372036854775807], default: 100000] heuristics/proximity/maxlpiters = 100000 # minimum number of LP iterations performed in subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 200] heuristics/proximity/minlpiters = 200 # waiting nodes since last incumbent before heuristic is executed # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 100] heuristics/proximity/waitingnodes = 100 # factor by which proximity should at least improve the incumbent # [type: real, advanced: TRUE, range: [0,1], default: 0.02] heuristics/proximity/minimprove = 0.02 # sub-MIP node limit w.r.t number of original nodes # [type: real, advanced: TRUE, range: [0,1e+20], default: 0.1] heuristics/proximity/nodesquot = 0.1 # threshold for percentage of binary variables required to start # [type: real, advanced: TRUE, range: [0,1], default: 0.1] heuristics/proximity/binvarquot = 0.1 # quotient of sub-MIP LP iterations with respect to LP iterations so far # [type: real, advanced: TRUE, range: [0,1], default: 0.2] heuristics/proximity/lpitersquot = 0.2 # minimum primal-dual gap for which the heuristic is executed # [type: real, advanced: TRUE, range: [0,1e+20], default: 0.01] heuristics/proximity/mingap = 0.01 # should uct node selection be used at the beginning of the search? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/proximity/useuct = FALSE # priority of heuristic <pscostdiving> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1002000] heuristics/pscostdiving/priority = -1002000 # frequency for calling primal heuristic <pscostdiving> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] heuristics/pscostdiving/freq = 10 # frequency offset for calling primal heuristic <pscostdiving> # [type: int, advanced: FALSE, range: [0,1073741822], default: 2] heuristics/pscostdiving/freqofs = 2 # maximal depth level to call primal heuristic <pscostdiving> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/pscostdiving/maxdepth = -1 # minimal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/pscostdiving/minreldepth = 0 # maximal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/pscostdiving/maxreldepth = 1 # maximal fraction of diving LP iterations compared to node LP iterations # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.05] heuristics/pscostdiving/maxlpiterquot = 0.05 # additional number of allowed LP iterations # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] heuristics/pscostdiving/maxlpiterofs = 1000 # maximal quotient (curlowerbound - lowerbound)/(cutoffbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.8] heuristics/pscostdiving/maxdiveubquot = 0.8 # maximal quotient (curlowerbound - lowerbound)/(avglowerbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/pscostdiving/maxdiveavgquot = 0 # maximal UBQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.1] heuristics/pscostdiving/maxdiveubquotnosol = 0.1 # maximal AVGQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/pscostdiving/maxdiveavgquotnosol = 0 # use one level of backtracking if infeasibility is encountered? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/pscostdiving/backtrack = TRUE # percentage of immediate domain changes during probing to trigger LP resolve # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.15] heuristics/pscostdiving/lpresolvedomchgquot = 0.15 # LP solve frequency for diving heuristics (0: only after enough domain changes have been found) # [type: int, advanced: FALSE, range: [0,2147483647], default: 0] heuristics/pscostdiving/lpsolvefreq = 0 # should only LP branching candidates be considered instead of the slower but more general constraint handler diving variable selection? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/pscostdiving/onlylpbranchcands = TRUE # priority of heuristic <randrounding> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -200] heuristics/randrounding/priority = -200 # frequency for calling primal heuristic <randrounding> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 20] heuristics/randrounding/freq = 20 # frequency offset for calling primal heuristic <randrounding> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/randrounding/freqofs = 0 # maximal depth level to call primal heuristic <randrounding> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/randrounding/maxdepth = -1 # should the heuristic only be called once per node? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/randrounding/oncepernode = FALSE # should the heuristic apply the variable lock strategy of simple rounding, if possible? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/randrounding/usesimplerounding = FALSE # should the probing part of the heuristic be applied exclusively at the root node? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/randrounding/propagateonlyroot = TRUE # limit of rounds for each propagation call # [type: int, advanced: TRUE, range: [-1,2147483647], default: 1] heuristics/randrounding/maxproprounds = 1 # priority of heuristic <rens> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1100000] heuristics/rens/priority = -1100000 # frequency for calling primal heuristic <rens> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] heuristics/rens/freq = 0 # frequency offset for calling primal heuristic <rens> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/rens/freqofs = 0 # maximal depth level to call primal heuristic <rens> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/rens/maxdepth = -1 # minimum percentage of integer variables that have to be fixable # [type: real, advanced: FALSE, range: [0,1], default: 0.5] heuristics/rens/minfixingrate = 0.5 # maximum number of nodes to regard in the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 5000] heuristics/rens/maxnodes = 5000 # number of nodes added to the contingent of the total nodes # [type: longint, advanced: FALSE, range: [0,9223372036854775807], default: 500] heuristics/rens/nodesofs = 500 # minimum number of nodes required to start the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 50] heuristics/rens/minnodes = 50 # contingent of sub problem nodes in relation to the number of nodes of the original problem # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/rens/nodesquot = 0.1 # factor by which RENS should at least improve the incumbent # [type: real, advanced: TRUE, range: [0,1], default: 0.01] heuristics/rens/minimprove = 0.01 # factor by which the limit on the number of LP depends on the node limit # [type: real, advanced: TRUE, range: [1,1.79769313486232e+308], default: 2] heuristics/rens/lplimfac = 2 # solution that is used for fixing values ('l'p relaxation, 'n'lp relaxation) # [type: char, advanced: FALSE, range: {nl}, default: l] heuristics/rens/startsol = l # should general integers get binary bounds [floor(.),ceil(.)] ? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/rens/binarybounds = TRUE # should subproblem be created out of the rows in the LP rows? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/rens/uselprows = FALSE # if uselprows == FALSE, should all active cuts from cutpool be copied to constraints in subproblem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/rens/copycuts = TRUE # should the RENS sub-CIP get its own full time limit? This is only for testing and not recommended! # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/rens/extratime = FALSE # should all subproblem solutions be added to the original SCIP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/rens/addallsols = FALSE # should the RENS sub-CIP be solved with cuts, conflicts, strong branching,... This is only for testing and not recommended! # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/rens/fullscale = FALSE # limit on number of improving incumbent solutions in sub-CIP # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] heuristics/rens/bestsollimit = -1 # should uct node selection be used at the beginning of the search? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/rens/useuct = FALSE # priority of heuristic <reoptsols> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 40000] heuristics/reoptsols/priority = 40000 # frequency for calling primal heuristic <reoptsols> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] heuristics/reoptsols/freq = 0 # frequency offset for calling primal heuristic <reoptsols> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/reoptsols/freqofs = 0 # maximal depth level to call primal heuristic <reoptsols> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 0] heuristics/reoptsols/maxdepth = 0 # maximal number solutions which should be checked. (-1: all) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 1000] heuristics/reoptsols/maxsols = 1000 # check solutions of the last k runs. (-1: all) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] heuristics/reoptsols/maxruns = -1 # priority of heuristic <repair> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -20] heuristics/repair/priority = -20 # frequency for calling primal heuristic <repair> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] heuristics/repair/freq = -1 # frequency offset for calling primal heuristic <repair> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/repair/freqofs = 0 # maximal depth level to call primal heuristic <repair> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/repair/maxdepth = -1 # file name of a solution to be used as infeasible starting point, [-] if not available # [type: string, advanced: FALSE, default: "-"] heuristics/repair/filename = "-" # True : fractional variables which are not fractional in the given solution are rounded, FALSE : solving process of this heuristic is stopped. # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/repair/roundit = TRUE # should a scaled objective function for original variables be used in repair subproblem? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/repair/useobjfactor = FALSE # should variable fixings be used in repair subproblem? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/repair/usevarfix = TRUE # should slack variables be used in repair subproblem? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/repair/useslackvars = FALSE # factor for the potential of var fixings # [type: real, advanced: TRUE, range: [0,100], default: 2] heuristics/repair/alpha = 2 # number of nodes added to the contingent of the total nodes # [type: int, advanced: FALSE, range: [0,2147483647], default: 500] heuristics/repair/nodesofs = 500 # maximum number of nodes to regard in the subproblem # [type: int, advanced: TRUE, range: [0,2147483647], default: 5000] heuristics/repair/maxnodes = 5000 # minimum number of nodes required to start the subproblem # [type: int, advanced: TRUE, range: [0,2147483647], default: 50] heuristics/repair/minnodes = 50 # contingent of sub problem nodes in relation to the number of nodes of the original problem # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/repair/nodesquot = 0.1 # minimum percentage of integer variables that have to be fixed # [type: real, advanced: FALSE, range: [0,1], default: 0.3] heuristics/repair/minfixingrate = 0.3 # priority of heuristic <rins> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1101000] heuristics/rins/priority = -1101000 # frequency for calling primal heuristic <rins> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 25] heuristics/rins/freq = 25 # frequency offset for calling primal heuristic <rins> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/rins/freqofs = 0 # maximal depth level to call primal heuristic <rins> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/rins/maxdepth = -1 # number of nodes added to the contingent of the total nodes # [type: int, advanced: FALSE, range: [0,2147483647], default: 500] heuristics/rins/nodesofs = 500 # maximum number of nodes to regard in the subproblem # [type: int, advanced: TRUE, range: [0,2147483647], default: 5000] heuristics/rins/maxnodes = 5000 # minimum number of nodes required to start the subproblem # [type: int, advanced: TRUE, range: [0,2147483647], default: 50] heuristics/rins/minnodes = 50 # contingent of sub problem nodes in relation to the number of nodes of the original problem # [type: real, advanced: FALSE, range: [0,1], default: 0.3] heuristics/rins/nodesquot = 0.3 # number of nodes without incumbent change that heuristic should wait # [type: int, advanced: TRUE, range: [0,2147483647], default: 200] heuristics/rins/nwaitingnodes = 200 # factor by which rins should at least improve the incumbent # [type: real, advanced: TRUE, range: [0,1], default: 0.01] heuristics/rins/minimprove = 0.01 # minimum percentage of integer variables that have to be fixed # [type: real, advanced: FALSE, range: [0,1], default: 0.3] heuristics/rins/minfixingrate = 0.3 # factor by which the limit on the number of LP depends on the node limit # [type: real, advanced: TRUE, range: [1,1.79769313486232e+308], default: 2] heuristics/rins/lplimfac = 2 # should subproblem be created out of the rows in the LP rows? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/rins/uselprows = FALSE # if uselprows == FALSE, should all active cuts from cutpool be copied to constraints in subproblem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/rins/copycuts = TRUE # should uct node selection be used at the beginning of the search? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/rins/useuct = FALSE # priority of heuristic <rootsoldiving> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1005000] heuristics/rootsoldiving/priority = -1005000 # frequency for calling primal heuristic <rootsoldiving> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 20] heuristics/rootsoldiving/freq = 20 # frequency offset for calling primal heuristic <rootsoldiving> # [type: int, advanced: FALSE, range: [0,1073741822], default: 5] heuristics/rootsoldiving/freqofs = 5 # maximal depth level to call primal heuristic <rootsoldiving> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/rootsoldiving/maxdepth = -1 # minimal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/rootsoldiving/minreldepth = 0 # maximal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/rootsoldiving/maxreldepth = 1 # maximal fraction of diving LP iterations compared to node LP iterations # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.01] heuristics/rootsoldiving/maxlpiterquot = 0.01 # additional number of allowed LP iterations # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] heuristics/rootsoldiving/maxlpiterofs = 1000 # total number of feasible solutions found up to which heuristic is called (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] heuristics/rootsoldiving/maxsols = -1 # maximal diving depth: number of binary/integer variables times depthfac # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.5] heuristics/rootsoldiving/depthfac = 0.5 # maximal diving depth factor if no feasible solution was found yet # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 2] heuristics/rootsoldiving/depthfacnosol = 2 # soft rounding factor to fade out objective coefficients # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/rootsoldiving/alpha = 0.9 # priority of heuristic <rounding> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1000] heuristics/rounding/priority = -1000 # frequency for calling primal heuristic <rounding> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] heuristics/rounding/freq = 1 # frequency offset for calling primal heuristic <rounding> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/rounding/freqofs = 0 # maximal depth level to call primal heuristic <rounding> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/rounding/maxdepth = -1 # number of calls per found solution that are considered as standard success, a higher factor causes the heuristic to be called more often # [type: int, advanced: TRUE, range: [-1,2147483647], default: 100] heuristics/rounding/successfactor = 100 # should the heuristic only be called once per node? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/rounding/oncepernode = FALSE # priority of heuristic <scheduler> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -30000] heuristics/scheduler/priority = -30000 # frequency for calling primal heuristic <scheduler> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] heuristics/scheduler/freq = -1 # frequency offset for calling primal heuristic <scheduler> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/scheduler/freqofs = 0 # maximal depth level to call primal heuristic <scheduler> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/scheduler/maxdepth = -1 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/scheduler/rens/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/scheduler/rens/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/scheduler/rens/active = TRUE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/scheduler/rens/priority = 1 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/scheduler/rins/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/scheduler/rins/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/scheduler/rins/active = TRUE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/scheduler/rins/priority = 1 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/scheduler/mutation/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/scheduler/mutation/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/scheduler/mutation/active = TRUE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/scheduler/mutation/priority = 1 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/scheduler/localbranching/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/scheduler/localbranching/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/scheduler/localbranching/active = TRUE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/scheduler/localbranching/priority = 1 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/scheduler/crossover/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/scheduler/crossover/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/scheduler/crossover/active = TRUE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/scheduler/crossover/priority = 1 # the number of solutions that crossover should combine # [type: int, advanced: TRUE, range: [2,10], default: 2] heuristics/scheduler/crossover/nsols = 2 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/scheduler/proximity/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/scheduler/proximity/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/scheduler/proximity/active = TRUE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/scheduler/proximity/priority = 1 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/scheduler/zeroobjective/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/scheduler/zeroobjective/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/scheduler/zeroobjective/active = TRUE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/scheduler/zeroobjective/priority = 1 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/scheduler/dins/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/scheduler/dins/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/scheduler/dins/active = TRUE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/scheduler/dins/priority = 1 # number of pool solutions where binary solution values must agree # [type: int, advanced: TRUE, range: [1,100], default: 5] heuristics/scheduler/dins/npoolsols = 5 # minimum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/scheduler/trustregion/minfixingrate = 0.3 # maximum fixing rate for this neighborhood # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/scheduler/trustregion/maxfixingrate = 0.9 # is this neighborhood active? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/scheduler/trustregion/active = FALSE # positive call priority to initialize bandit algorithms # [type: real, advanced: TRUE, range: [0.01,1], default: 1] heuristics/scheduler/trustregion/priority = 1 # the penalty for each change in the binary variables from the candidate solution # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 100] heuristics/scheduler/trustregion/violpenalty = 100 # maximum number of nodes to regard in the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 500] heuristics/scheduler/maxnodes = 500 # offset added to the nodes budget # [type: longint, advanced: FALSE, range: [0,9223372036854775807], default: 500] heuristics/scheduler/nodesofs = 500 # minimum number of nodes required to start a sub-SCIP # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 50] heuristics/scheduler/minnodes = 50 # number of nodes since last incumbent solution that the heuristic should wait # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 0] heuristics/scheduler/waitingnodes = 0 # initial node limit for LNS heuristics # [type: int, advanced: TRUE, range: [0,2147483647], default: 50] heuristics/scheduler/initlnsnodelimit = 50 # initial node limit for diving heuristics # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 500] heuristics/scheduler/initdivingnodelimit = 500 # fraction of nodes compared to the main SCIP for budget computation # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/scheduler/nodesquot = 0.1 # lower bound fraction of nodes compared to the main SCIP for budget computation # [type: real, advanced: FALSE, range: [0,1], default: 0] heuristics/scheduler/nodesquotmin = 0 # limit on the number of improving solutions in a sub-SCIP call # [type: int, advanced: FALSE, range: [-1,2147483647], default: 3] heuristics/scheduler/nsolslim = 3 # the bandit algorithm: (u)pper confidence bounds, (e)xp.3, epsilon (g)reedy, exp.3-(i)x # [type: char, advanced: TRUE, range: {uegi}, default: i] heuristics/scheduler/banditalgo = i # weight between uniform (gamma ~ 1) and weight driven (gamma ~ 0) probability distribution for exp3 # [type: real, advanced: TRUE, range: [0,1], default: 0.07041455] heuristics/scheduler/gamma = 0.07041455 # reward offset between 0 and 1 at every observation for Exp.3 # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/scheduler/beta = 0 # parameter to increase the confidence width in UCB # [type: real, advanced: TRUE, range: [0,100], default: 0.0016] heuristics/scheduler/alpha = 0.0016 # distances from fixed variables be used for variable prioritization # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/scheduler/usedistances = TRUE # should reduced cost scores be used for variable prioritization? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/scheduler/useredcost = TRUE # should pseudo cost scores be used for variable priorization? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/scheduler/usepscost = TRUE # should local reduced costs be used for generic (un)fixing? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/scheduler/uselocalredcost = FALSE # should the heuristic activate other sub-SCIP heuristics during its search? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/scheduler/usesubscipheurs = FALSE # factor by which target node number is eventually increased # [type: real, advanced: TRUE, range: [1,100000], default: 1.05] heuristics/scheduler/targetnodefactor = 1.05 # initial random seed for bandit algorithms and random decisions by neighborhoods # [type: int, advanced: FALSE, range: [0,2147483647], default: 113] heuristics/scheduler/seed = 113 # number of allowed executions of the heuristic on the same incumbent solution (-1: no limit, 0: number of active neighborhoods) # [type: int, advanced: TRUE, range: [-1,100], default: -1] heuristics/scheduler/maxcallssamesol = -1 # increase exploration in epsilon-greedy bandit algorithm # [type: real, advanced: TRUE, range: [0,1], default: 0.4685844] heuristics/scheduler/eps = 0.4685844 # TRUE if modified version of the epsilon-greedy bandit algorithm should be used # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/scheduler/epsgreedy_usemod = TRUE # weight by how much finding a new incumbent is rewarded in reward function # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/scheduler/solrewardweight = 0.3 # weight by how much effort is rewarded in reward function # [type: real, advanced: TRUE, range: [0,1], default: 0.2] heuristics/scheduler/effortrewardweight = 0.2 # weight by how much quality of a new incumbent is rewarded in reward function # [type: real, advanced: TRUE, range: [0,1], default: 0.3] heuristics/scheduler/qualrewardweight = 0.3 # weight by how much number of conflicts found by diving is rewarded in reward function # [type: real, advanced: TRUE, range: [0,1], default: 0.2] heuristics/scheduler/conflictrewardweight = 0.2 # should the bandit algorithms be reset when a new problem is read? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/scheduler/resetweights = FALSE # should random seeds of sub-SCIPs be altered to increase diversification? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/scheduler/subsciprandseeds = FALSE # should cutting planes be copied to the sub-SCIP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/scheduler/copycuts = FALSE # tolerance by which the fixing rate may be missed without generic fixing # [type: real, advanced: TRUE, range: [0,1], default: 0.1] heuristics/scheduler/fixtol = 0.1 # tolerance by which the fixing rate may be exceeded without generic unfixing # [type: real, advanced: TRUE, range: [0,1], default: 0.1] heuristics/scheduler/unfixtol = 0.1 # time limit for a single heuristic run # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 60] heuristics/scheduler/heurtimelimit = 60 # should the heuristic be executed multiple times during the root node? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/scheduler/initduringroot = FALSE # should the default priorities be used at the root node? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/scheduler/defaultroot = TRUE # number of heuristics picked by the scheduler in one call (-1: number of controlled heuristics, 0: until new incumbent is found) # [type: int, advanced: TRUE, range: [-1,100], default: 5] heuristics/scheduler/nselections = 5 # is statistics table <scheduler> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/scheduler/active = TRUE # priority of heuristic <shiftandpropagate> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 1000] heuristics/shiftandpropagate/priority = 1000 # frequency for calling primal heuristic <shiftandpropagate> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] heuristics/shiftandpropagate/freq = 0 # frequency offset for calling primal heuristic <shiftandpropagate> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/shiftandpropagate/freqofs = 0 # maximal depth level to call primal heuristic <shiftandpropagate> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/shiftandpropagate/maxdepth = -1 # The number of propagation rounds used for each propagation # [type: int, advanced: TRUE, range: [-1,1000], default: 10] heuristics/shiftandpropagate/nproprounds = 10 # Should continuous variables be relaxed? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/shiftandpropagate/relax = TRUE # Should domains be reduced by probing? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/shiftandpropagate/probing = TRUE # Should heuristic only be executed if no primal solution was found, yet? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/shiftandpropagate/onlywithoutsol = TRUE # The number of cutoffs before heuristic stops # [type: int, advanced: TRUE, range: [-1,1000000], default: 15] heuristics/shiftandpropagate/cutoffbreaker = 15 # the key for variable sorting: (n)orms down, norms (u)p, (v)iolations down, viola(t)ions up, or (r)andom # [type: char, advanced: TRUE, range: {nrtuv}, default: v] heuristics/shiftandpropagate/sortkey = v # Should variables be sorted for the heuristic? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/shiftandpropagate/sortvars = TRUE # should variable statistics be collected during probing? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/shiftandpropagate/collectstats = TRUE # Should the heuristic stop calculating optimal shift values when no more rows are violated? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/shiftandpropagate/stopafterfeasible = TRUE # Should binary variables be shifted first? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/shiftandpropagate/preferbinaries = TRUE # should variables with a zero shifting value be delayed instead of being fixed? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/shiftandpropagate/nozerofixing = FALSE # should binary variables with no locks in one direction be fixed to that direction? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/shiftandpropagate/fixbinlocks = TRUE # should binary variables with no locks be preferred in the ordering? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/shiftandpropagate/binlocksfirst = FALSE # should coefficients and left/right hand sides be normalized by max row coeff? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/shiftandpropagate/normalize = TRUE # should row weight be increased every time the row is violated? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/shiftandpropagate/updateweights = FALSE # should implicit integer variables be treated as continuous variables? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/shiftandpropagate/impliscontinuous = TRUE # should the heuristic choose the best candidate in every round? (set to FALSE for static order)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/shiftandpropagate/selectbest = FALSE # maximum percentage of allowed cutoffs before stopping the heuristic # [type: real, advanced: TRUE, range: [0,2], default: 0] heuristics/shiftandpropagate/maxcutoffquot = 0 # minimum fixing rate over all variables (including continuous) to solve LP # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/shiftandpropagate/minfixingratelp = 0 # priority of heuristic <shifting> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -5000] heuristics/shifting/priority = -5000 # frequency for calling primal heuristic <shifting> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] heuristics/shifting/freq = 10 # frequency offset for calling primal heuristic <shifting> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/shifting/freqofs = 0 # maximal depth level to call primal heuristic <shifting> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/shifting/maxdepth = -1 # priority of heuristic <simplerounding> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -30] heuristics/simplerounding/priority = -30 # frequency for calling primal heuristic <simplerounding> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] heuristics/simplerounding/freq = 1 # frequency offset for calling primal heuristic <simplerounding> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/simplerounding/freqofs = 0 # maximal depth level to call primal heuristic <simplerounding> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/simplerounding/maxdepth = -1 # should the heuristic only be called once per node? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/simplerounding/oncepernode = FALSE # priority of heuristic <subnlp> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -2000010] heuristics/subnlp/priority = -2000010 # frequency for calling primal heuristic <subnlp> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] heuristics/subnlp/freq = 1 # frequency offset for calling primal heuristic <subnlp> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/subnlp/freqofs = 0 # maximal depth level to call primal heuristic <subnlp> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/subnlp/maxdepth = -1 # verbosity level of NLP solver # [type: int, advanced: FALSE, range: [0,65535], default: 0] heuristics/subnlp/nlpverblevel = 0 # number of nodes added to the current number of nodes when computing itercontingent (higher value runs heuristic more often in early search) # [type: int, advanced: FALSE, range: [0,2147483647], default: 1600] heuristics/subnlp/nodesoffset = 1600 # factor on number of nodes in SCIP (plus nodesoffset) to compute itercontingent (higher value runs heuristics more frequently) # [type: real, advanced: FALSE, range: [0,1e+20], default: 0.3] heuristics/subnlp/nodesfactor = 0.3 # exponent for power of success rate to be multiplied with itercontingent (lower value decreases impact of success rate) # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 1] heuristics/subnlp/successrateexp = 1 # number of iterations used for initial NLP solves # [type: int, advanced: FALSE, range: [0,2147483647], default: 300] heuristics/subnlp/iterinit = 300 # number of successful NLP solves until switching to iterlimit guess and using success rate # [type: int, advanced: FALSE, range: [0,2147483647], default: 2] heuristics/subnlp/ninitsolves = 2 # minimal number of iterations for NLP solves # [type: int, advanced: FALSE, range: [0,2147483647], default: 20] heuristics/subnlp/itermin = 20 # absolute optimality tolerance to use for NLP solves # [type: real, advanced: TRUE, range: [0,1], default: 1e-07] heuristics/subnlp/opttol = 1e-07 # factor on SCIP feasibility tolerance for NLP solves if resolving when NLP solution not feasible in CIP # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/subnlp/feastolfactor = 0.1 # limit on number of presolve rounds in sub-SCIP (-1 for unlimited, 0 for no presolve) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] heuristics/subnlp/maxpresolverounds = -1 # presolve emphasis in sub-SCIP (0: default, 1: aggressive, 2: fast, 3: off) # [type: int, advanced: FALSE, range: [0,3], default: 2] heuristics/subnlp/presolveemphasis = 2 # whether to set cutoff in sub-SCIP to current primal bound # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/subnlp/setcutoff = TRUE # whether to add constraints that forbid specific fixings that turned out to be infeasible # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/subnlp/forbidfixings = FALSE # whether to keep SCIP copy or to create new copy each time heuristic is applied # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/subnlp/keepcopy = TRUE # percentage of NLP solves with infeasible status required to tell NLP solver to expect an infeasible NLP # [type: real, advanced: FALSE, range: [0,1], default: 0] heuristics/subnlp/expectinfeas = 0 # priority of heuristic <trivial> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 10000] heuristics/trivial/priority = 10000 # frequency for calling primal heuristic <trivial> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] heuristics/trivial/freq = 0 # frequency offset for calling primal heuristic <trivial> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/trivial/freqofs = 0 # maximal depth level to call primal heuristic <trivial> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/trivial/maxdepth = -1 # priority of heuristic <trivialnegation> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 39990] heuristics/trivialnegation/priority = 39990 # frequency for calling primal heuristic <trivialnegation> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] heuristics/trivialnegation/freq = 0 # frequency offset for calling primal heuristic <trivialnegation> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/trivialnegation/freqofs = 0 # maximal depth level to call primal heuristic <trivialnegation> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: 0] heuristics/trivialnegation/maxdepth = 0 # priority of heuristic <trustregion> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1102010] heuristics/trustregion/priority = -1102010 # frequency for calling primal heuristic <trustregion> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] heuristics/trustregion/freq = -1 # frequency offset for calling primal heuristic <trustregion> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/trustregion/freqofs = 0 # maximal depth level to call primal heuristic <trustregion> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/trustregion/maxdepth = -1 # number of nodes added to the contingent of the total nodes # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] heuristics/trustregion/nodesofs = 1000 # the number of binary variables necessary to run the heuristic # [type: int, advanced: FALSE, range: [1,2147483647], default: 10] heuristics/trustregion/minbinvars = 10 # contingent of sub problem nodes in relation to the number of nodes of the original problem # [type: real, advanced: FALSE, range: [0,1], default: 0.05] heuristics/trustregion/nodesquot = 0.05 # factor by which the limit on the number of LP depends on the node limit # [type: real, advanced: TRUE, range: [1,1.79769313486232e+308], default: 1.5] heuristics/trustregion/lplimfac = 1.5 # minimum number of nodes required to start the subproblem # [type: int, advanced: TRUE, range: [0,2147483647], default: 100] heuristics/trustregion/minnodes = 100 # maximum number of nodes to regard in the subproblem # [type: int, advanced: TRUE, range: [0,2147483647], default: 10000] heuristics/trustregion/maxnodes = 10000 # number of nodes without incumbent change that heuristic should wait # [type: int, advanced: TRUE, range: [0,2147483647], default: 1] heuristics/trustregion/nwaitingnodes = 1 # should subproblem be created out of the rows in the LP rows? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/trustregion/uselprows = FALSE # if uselprows == FALSE, should all active cuts from cutpool be copied to constraints in subproblem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/trustregion/copycuts = TRUE # limit on number of improving incumbent solutions in sub-CIP # [type: int, advanced: FALSE, range: [-1,2147483647], default: 3] heuristics/trustregion/bestsollimit = 3 # the penalty for each change in the binary variables from the candidate solution # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 100] heuristics/trustregion/violpenalty = 100 # the minimum absolute improvement in the objective function value # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.01] heuristics/trustregion/objminimprove = 0.01 # priority of heuristic <trysol> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -3000010] heuristics/trysol/priority = -3000010 # frequency for calling primal heuristic <trysol> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] heuristics/trysol/freq = 1 # frequency offset for calling primal heuristic <trysol> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/trysol/freqofs = 0 # maximal depth level to call primal heuristic <trysol> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/trysol/maxdepth = -1 # priority of heuristic <twoopt> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -20100] heuristics/twoopt/priority = -20100 # frequency for calling primal heuristic <twoopt> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] heuristics/twoopt/freq = -1 # frequency offset for calling primal heuristic <twoopt> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/twoopt/freqofs = 0 # maximal depth level to call primal heuristic <twoopt> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/twoopt/maxdepth = -1 # Should Integer-2-Optimization be applied or not? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/twoopt/intopt = FALSE # user parameter to determine number of nodes to wait after last best solution before calling heuristic # [type: int, advanced: TRUE, range: [0,10000], default: 0] heuristics/twoopt/waitingnodes = 0 # maximum number of slaves for one master variable # [type: int, advanced: TRUE, range: [-1,1000000], default: 199] heuristics/twoopt/maxnslaves = 199 # parameter to determine the percentage of rows two variables have to share before they are considered equal # [type: real, advanced: TRUE, range: [0,1], default: 0.5] heuristics/twoopt/matchingrate = 0.5 # priority of heuristic <undercover> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1110000] heuristics/undercover/priority = -1110000 # frequency for calling primal heuristic <undercover> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] heuristics/undercover/freq = 0 # frequency offset for calling primal heuristic <undercover> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/undercover/freqofs = 0 # maximal depth level to call primal heuristic <undercover> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/undercover/maxdepth = -1 # prioritized sequence of fixing values used ('l'p relaxation, 'n'lp relaxation, 'i'ncumbent solution) # [type: string, advanced: FALSE, default: "li"] heuristics/undercover/fixingalts = "li" # maximum number of nodes to regard in the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 500] heuristics/undercover/maxnodes = 500 # minimum number of nodes required to start the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 500] heuristics/undercover/minnodes = 500 # number of nodes added to the contingent of the total nodes # [type: longint, advanced: FALSE, range: [0,9223372036854775807], default: 500] heuristics/undercover/nodesofs = 500 # weight for conflict score in fixing order # [type: real, advanced: TRUE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 1000] heuristics/undercover/conflictweight = 1000 # weight for cutoff score in fixing order # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 1] heuristics/undercover/cutoffweight = 1 # weight for inference score in fixing order # [type: real, advanced: TRUE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 1] heuristics/undercover/inferenceweight = 1 # maximum coversize (as fraction of total number of variables) # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/undercover/maxcoversizevars = 1 # maximum coversize (as ratio to the percentage of non-affected constraints) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 1.79769313486232e+308] heuristics/undercover/maxcoversizeconss = 1.79769313486232e+308 # minimum percentage of nonlinear constraints in the original problem # [type: real, advanced: TRUE, range: [0,1], default: 0.15] heuristics/undercover/mincoveredrel = 0.15 # factor by which the heuristic should at least improve the incumbent # [type: real, advanced: TRUE, range: [-1,1], default: 0] heuristics/undercover/minimprove = 0 # contingent of sub problem nodes in relation to the number of nodes of the original problem # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/undercover/nodesquot = 0.1 # fraction of covering variables in the last cover which need to change their value when recovering # [type: real, advanced: TRUE, range: [0,1], default: 0.9] heuristics/undercover/recoverdiv = 0.9 # minimum number of nonlinear constraints in the original problem # [type: int, advanced: TRUE, range: [0,2147483647], default: 5] heuristics/undercover/mincoveredabs = 5 # maximum number of backtracks in fix-and-propagate # [type: int, advanced: TRUE, range: [0,2147483647], default: 6] heuristics/undercover/maxbacktracks = 6 # maximum number of recoverings # [type: int, advanced: TRUE, range: [0,2147483647], default: 0] heuristics/undercover/maxrecovers = 0 # maximum number of reorderings of the fixing order # [type: int, advanced: TRUE, range: [0,2147483647], default: 1] heuristics/undercover/maxreorders = 1 # objective function of the covering problem (influenced nonlinear 'c'onstraints/'t'erms, 'd'omain size, 'l'ocks, 'm'in of up/down locks, 'u'nit penalties) # [type: char, advanced: TRUE, range: {cdlmtu}, default: u] heuristics/undercover/coveringobj = u # order in which variables should be fixed (increasing 'C'onflict score, decreasing 'c'onflict score, increasing 'V'ariable index, decreasing 'v'ariable index # [type: char, advanced: TRUE, range: {CcVv}, default: v] heuristics/undercover/fixingorder = v # should the heuristic be called at root node before cut separation? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/undercover/beforecuts = TRUE # should integer variables in the cover be fixed first? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/undercover/fixintfirst = FALSE # shall LP values for integer vars be rounded according to locks? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/undercover/locksrounding = TRUE # should we only fix variables in order to obtain a convex problem? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/undercover/onlyconvexify = FALSE # should the NLP heuristic be called to polish a feasible solution? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/undercover/postnlp = TRUE # should and constraints be covered (or just copied)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/undercover/coverand = TRUE # should bounddisjunction constraints be covered (or just copied)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/undercover/coverbd = FALSE # should indicator constraints be covered (or just copied)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/undercover/coverind = FALSE # should nonlinear constraints be covered (or just copied)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/undercover/covernl = TRUE # should all active cuts from cutpool be copied to constraints in subproblem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/undercover/copycuts = TRUE # shall the cover be reused if a conflict was added after an infeasible subproblem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/undercover/reusecover = FALSE # priority of heuristic <vbounds> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 2500] heuristics/vbounds/priority = 2500 # frequency for calling primal heuristic <vbounds> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] heuristics/vbounds/freq = 0 # frequency offset for calling primal heuristic <vbounds> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/vbounds/freqofs = 0 # maximal depth level to call primal heuristic <vbounds> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/vbounds/maxdepth = -1 # minimum percentage of integer variables that have to be fixed # [type: real, advanced: FALSE, range: [0,1], default: 0.65] heuristics/vbounds/minintfixingrate = 0.65 # minimum percentage of variables that have to be fixed within sub-SCIP (integer and continuous) # [type: real, advanced: FALSE, range: [0,1], default: 0.65] heuristics/vbounds/minmipfixingrate = 0.65 # maximum number of nodes to regard in the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 5000] heuristics/vbounds/maxnodes = 5000 # number of nodes added to the contingent of the total nodes # [type: longint, advanced: FALSE, range: [0,9223372036854775807], default: 500] heuristics/vbounds/nodesofs = 500 # minimum number of nodes required to start the subproblem # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 500] heuristics/vbounds/minnodes = 500 # contingent of sub problem nodes in relation to the number of nodes of the original problem # [type: real, advanced: FALSE, range: [0,1], default: 0.1] heuristics/vbounds/nodesquot = 0.1 # factor by which vbounds heuristic should at least improve the incumbent # [type: real, advanced: TRUE, range: [0,1], default: 0.01] heuristics/vbounds/minimprove = 0.01 # maximum number of propagation rounds during probing (-1 infinity) # [type: int, advanced: TRUE, range: [-1,536870911], default: 2] heuristics/vbounds/maxproprounds = 2 # should all active cuts from cutpool be copied to constraints in subproblem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/vbounds/copycuts = TRUE # should more variables be fixed based on variable locks if the fixing rate was not reached? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] heuristics/vbounds/uselockfixings = FALSE # maximum number of backtracks during the fixing process # [type: int, advanced: TRUE, range: [-1,536870911], default: 10] heuristics/vbounds/maxbacktracks = 10 # which variants of the vbounds heuristic that try to stay feasible should be called? (0: off, 1: w/o looking at obj, 2: only fix to best bound, 4: only fix to worst bound # [type: int, advanced: TRUE, range: [0,7], default: 6] heuristics/vbounds/feasvariant = 6 # which tightening variants of the vbounds heuristic should be called? (0: off, 1: w/o looking at obj, 2: only fix to best bound, 4: only fix to worst bound # [type: int, advanced: TRUE, range: [0,7], default: 7] heuristics/vbounds/tightenvariant = 7 # priority of heuristic <veclendiving> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1003100] heuristics/veclendiving/priority = -1003100 # frequency for calling primal heuristic <veclendiving> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] heuristics/veclendiving/freq = 10 # frequency offset for calling primal heuristic <veclendiving> # [type: int, advanced: FALSE, range: [0,1073741822], default: 4] heuristics/veclendiving/freqofs = 4 # maximal depth level to call primal heuristic <veclendiving> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/veclendiving/maxdepth = -1 # minimal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 0] heuristics/veclendiving/minreldepth = 0 # maximal relative depth to start diving # [type: real, advanced: TRUE, range: [0,1], default: 1] heuristics/veclendiving/maxreldepth = 1 # maximal fraction of diving LP iterations compared to node LP iterations # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.05] heuristics/veclendiving/maxlpiterquot = 0.05 # additional number of allowed LP iterations # [type: int, advanced: FALSE, range: [0,2147483647], default: 1000] heuristics/veclendiving/maxlpiterofs = 1000 # maximal quotient (curlowerbound - lowerbound)/(cutoffbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.8] heuristics/veclendiving/maxdiveubquot = 0.8 # maximal quotient (curlowerbound - lowerbound)/(avglowerbound - lowerbound) where diving is performed (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/veclendiving/maxdiveavgquot = 0 # maximal UBQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1], default: 0.1] heuristics/veclendiving/maxdiveubquotnosol = 0.1 # maximal AVGQUOT when no solution was found yet (0.0: no limit) # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] heuristics/veclendiving/maxdiveavgquotnosol = 0 # use one level of backtracking if infeasibility is encountered? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] heuristics/veclendiving/backtrack = TRUE # percentage of immediate domain changes during probing to trigger LP resolve # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.15] heuristics/veclendiving/lpresolvedomchgquot = 0.15 # LP solve frequency for diving heuristics (0: only after enough domain changes have been found) # [type: int, advanced: FALSE, range: [0,2147483647], default: 0] heuristics/veclendiving/lpsolvefreq = 0 # should only LP branching candidates be considered instead of the slower but more general constraint handler diving variable selection? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] heuristics/veclendiving/onlylpbranchcands = FALSE # priority of heuristic <zirounding> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -500] heuristics/zirounding/priority = -500 # frequency for calling primal heuristic <zirounding> (-1: never, 0: only at depth freqofs) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] heuristics/zirounding/freq = 1 # frequency offset for calling primal heuristic <zirounding> # [type: int, advanced: FALSE, range: [0,1073741822], default: 0] heuristics/zirounding/freqofs = 0 # maximal depth level to call primal heuristic <zirounding> (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] heuristics/zirounding/maxdepth = -1 # determines maximum number of rounding loops # [type: int, advanced: TRUE, range: [-1,2147483647], default: 2] heuristics/zirounding/maxroundingloops = 2 # flag to determine if Zirounding is deactivated after a certain percentage of unsuccessful calls # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] heuristics/zirounding/stopziround = TRUE # if percentage of found solutions falls below this parameter, Zirounding will be deactivated # [type: real, advanced: TRUE, range: [0,1], default: 0.02] heuristics/zirounding/stoppercentage = 0.02 # determines the minimum number of calls before percentage-based deactivation of Zirounding is applied # [type: int, advanced: TRUE, range: [1,2147483647], default: 1000] heuristics/zirounding/minstopncalls = 1000 # priority of propagator <dualfix> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 8000000] propagating/dualfix/priority = 8000000 # frequency for calling propagator <dualfix> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] propagating/dualfix/freq = 0 # should propagator be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/dualfix/delay = FALSE # timing when propagator should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS)) # [type: int, advanced: TRUE, range: [1,15], default: 1] propagating/dualfix/timingmask = 1 # presolving priority of propagator <dualfix> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 8000000] propagating/dualfix/presolpriority = 8000000 # maximal number of presolving rounds the propagator participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] propagating/dualfix/maxprerounds = -1 # timing mask of the presolving method of propagator <dualfix> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [2,60], default: 4] propagating/dualfix/presoltiming = 4 # priority of propagator <genvbounds> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 3000000] propagating/genvbounds/priority = 3000000 # frequency for calling propagator <genvbounds> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] propagating/genvbounds/freq = 1 # should propagator be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/genvbounds/delay = FALSE # timing when propagator should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS)) # [type: int, advanced: TRUE, range: [1,15], default: 15] propagating/genvbounds/timingmask = 15 # presolving priority of propagator <genvbounds> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -2000000] propagating/genvbounds/presolpriority = -2000000 # maximal number of presolving rounds the propagator participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] propagating/genvbounds/maxprerounds = -1 # timing mask of the presolving method of propagator <genvbounds> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [2,60], default: 4] propagating/genvbounds/presoltiming = 4 # apply global propagation? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/genvbounds/global = TRUE # apply genvbounds in root node if no new incumbent was found? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/genvbounds/propinrootnode = TRUE # sort genvbounds and wait for bound change events? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/genvbounds/sort = TRUE # should genvbounds be transformed to (linear) constraints? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/genvbounds/propasconss = FALSE # priority of propagator <obbt> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1000000] propagating/obbt/priority = -1000000 # frequency for calling propagator <obbt> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] propagating/obbt/freq = 0 # should propagator be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/obbt/delay = TRUE # timing when propagator should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS)) # [type: int, advanced: TRUE, range: [1,15], default: 4] propagating/obbt/timingmask = 4 # presolving priority of propagator <obbt> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 0] propagating/obbt/presolpriority = 0 # maximal number of presolving rounds the propagator participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] propagating/obbt/maxprerounds = -1 # timing mask of the presolving method of propagator <obbt> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [2,60], default: 28] propagating/obbt/presoltiming = 28 # should obbt try to provide genvbounds if possible? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/obbt/creategenvbounds = TRUE # should coefficients in filtering be normalized w.r.t. the domains sizes? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/obbt/normalize = TRUE # try to filter bounds in so-called filter rounds by solving auxiliary LPs? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/obbt/applyfilterrounds = FALSE # try to filter bounds with the LP solution after each solve? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/obbt/applytrivialfilter = TRUE # should we try to generate genvbounds during trivial and aggressive filtering? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/obbt/genvbdsduringfilter = TRUE # try to create genvbounds during separation process? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/obbt/genvbdsduringsepa = TRUE # minimal number of filtered bounds to apply another filter round # [type: int, advanced: TRUE, range: [1,2147483647], default: 2] propagating/obbt/minfilter = 2 # multiple of root node LP iterations used as total LP iteration limit for obbt (<= 0: no limit ) # [type: real, advanced: FALSE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 10] propagating/obbt/itlimitfactor = 10 # multiple of OBBT LP limit used as total LP iteration limit for solving bilinear inequality LPs (< 0 for no limit) # [type: real, advanced: FALSE, range: [-1.79769313486232e+308,1.79769313486232e+308], default: 3] propagating/obbt/itlimitfactorbilin = 3 # minimum absolute value of nonconvex eigenvalues for a bilinear term # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.1] propagating/obbt/minnonconvexity = 0.1 # minimum LP iteration limit # [type: longint, advanced: FALSE, range: [0,9223372036854775807], default: 5000] propagating/obbt/minitlimit = 5000 # feasibility tolerance for reduced costs used in obbt; this value is used if SCIP's dual feastol is greater # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 1e-09] propagating/obbt/dualfeastol = 1e-09 # maximum condition limit used in LP solver (-1.0: no limit) # [type: real, advanced: FALSE, range: [-1,1.79769313486232e+308], default: -1] propagating/obbt/conditionlimit = -1 # minimal relative improve for strengthening bounds # [type: real, advanced: FALSE, range: [0,1], default: 0.001] propagating/obbt/boundstreps = 0.001 # threshold whether upper bounds of vars of indicator conss are considered or tightened # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 1000000] propagating/obbt/indicatorthreshold = 1000000 # only apply obbt on non-convex variables # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/obbt/onlynonconvexvars = TRUE # apply obbt on variables of indicator constraints? (independent of convexity) # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/obbt/indicators = FALSE # should integral bounds be tightened during the probing mode? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/obbt/tightintboundsprobing = TRUE # should continuous bounds be tightened during the probing mode? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/obbt/tightcontboundsprobing = FALSE # solve auxiliary LPs in order to find valid inequalities for bilinear terms? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/obbt/createbilinineqs = TRUE # create linear constraints from inequalities for bilinear terms? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/obbt/createlincons = FALSE # select the type of ordering algorithm which should be used (0: no special ordering, 1: greedy, 2: greedy reverse) # [type: int, advanced: TRUE, range: [0,2], default: 1] propagating/obbt/orderingalgo = 1 # should the obbt LP solution be separated? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/obbt/separatesol = FALSE # minimum number of iteration spend to separate an obbt LP solution # [type: int, advanced: TRUE, range: [0,2147483647], default: 0] propagating/obbt/sepaminiter = 0 # maximum number of iteration spend to separate an obbt LP solution # [type: int, advanced: TRUE, range: [0,2147483647], default: 10] propagating/obbt/sepamaxiter = 10 # trigger a propagation round after that many bound tightenings (0: no propagation) # [type: int, advanced: TRUE, range: [0,2147483647], default: 0] propagating/obbt/propagatefreq = 0 # priority of propagator <nlobbt> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1100000] propagating/nlobbt/priority = -1100000 # frequency for calling propagator <nlobbt> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] propagating/nlobbt/freq = -1 # should propagator be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/nlobbt/delay = TRUE # timing when propagator should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS)) # [type: int, advanced: TRUE, range: [1,15], default: 4] propagating/nlobbt/timingmask = 4 # presolving priority of propagator <nlobbt> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 0] propagating/nlobbt/presolpriority = 0 # maximal number of presolving rounds the propagator participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] propagating/nlobbt/maxprerounds = -1 # timing mask of the presolving method of propagator <nlobbt> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [2,60], default: 28] propagating/nlobbt/presoltiming = 28 # factor for NLP feasibility tolerance # [type: real, advanced: TRUE, range: [0,1], default: 0.01] propagating/nlobbt/feastolfac = 0.01 # factor for NLP relative objective tolerance # [type: real, advanced: TRUE, range: [0,1], default: 0.01] propagating/nlobbt/relobjtolfac = 0.01 # (#convex nlrows)/(#nonconvex nlrows) threshold to apply propagator # [type: real, advanced: TRUE, range: [0,1e+20], default: 0.2] propagating/nlobbt/minnonconvexfrac = 0.2 # minimum (#convex nlrows)/(#linear nlrows) threshold to apply propagator # [type: real, advanced: TRUE, range: [0,1e+20], default: 0.02] propagating/nlobbt/minlinearfrac = 0.02 # should non-initial LP rows be used? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] propagating/nlobbt/addlprows = TRUE # iteration limit of NLP solver; 0 for no limit # [type: int, advanced: TRUE, range: [0,2147483647], default: 500] propagating/nlobbt/nlpiterlimit = 500 # time limit of NLP solver; 0.0 for no limit # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] propagating/nlobbt/nlptimelimit = 0 # verbosity level of NLP solver # [type: int, advanced: TRUE, range: [0,5], default: 0] propagating/nlobbt/nlpverblevel = 0 # LP iteration limit for nlobbt will be this factor times total LP iterations in root node # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 2] propagating/nlobbt/itlimitfactor = 2 # priority of propagator <probing> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -100000] propagating/probing/priority = -100000 # frequency for calling propagator <probing> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] propagating/probing/freq = -1 # should propagator be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/probing/delay = TRUE # timing when propagator should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS)) # [type: int, advanced: TRUE, range: [1,15], default: 4] propagating/probing/timingmask = 4 # presolving priority of propagator <probing> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -100000] propagating/probing/presolpriority = -100000 # maximal number of presolving rounds the propagator participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] propagating/probing/maxprerounds = -1 # timing mask of the presolving method of propagator <probing> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [2,60], default: 16] propagating/probing/presoltiming = 16 # maximal number of runs, probing participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1] propagating/probing/maxruns = 1 # maximal number of propagation rounds in probing subproblems (-1: no limit, 0: auto) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] propagating/probing/proprounds = -1 # maximal number of fixings found, until probing is interrupted (0: don't iterrupt) # [type: int, advanced: TRUE, range: [0,2147483647], default: 25] propagating/probing/maxfixings = 25 # maximal number of successive probings without fixings, until probing is aborted (0: don't abort) # [type: int, advanced: TRUE, range: [0,2147483647], default: 1000] propagating/probing/maxuseless = 1000 # maximal number of successive probings without fixings, bound changes, and implications, until probing is aborted (0: don't abort) # [type: int, advanced: TRUE, range: [0,2147483647], default: 50] propagating/probing/maxtotaluseless = 50 # maximal number of probings without fixings, until probing is aborted (0: don't abort) # [type: int, advanced: TRUE, range: [0,2147483647], default: 0] propagating/probing/maxsumuseless = 0 # maximal depth until propagation is executed(-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] propagating/probing/maxdepth = -1 # priority of propagator <pseudoobj> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 3000000] propagating/pseudoobj/priority = 3000000 # frequency for calling propagator <pseudoobj> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] propagating/pseudoobj/freq = 1 # should propagator be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/pseudoobj/delay = FALSE # timing when propagator should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS)) # [type: int, advanced: TRUE, range: [1,15], default: 7] propagating/pseudoobj/timingmask = 7 # presolving priority of propagator <pseudoobj> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 6000000] propagating/pseudoobj/presolpriority = 6000000 # maximal number of presolving rounds the propagator participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] propagating/pseudoobj/maxprerounds = -1 # timing mask of the presolving method of propagator <pseudoobj> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [2,60], default: 4] propagating/pseudoobj/presoltiming = 4 # minimal number of successive non-binary variable propagations without a bound reduction before aborted # [type: int, advanced: TRUE, range: [0,2147483647], default: 100] propagating/pseudoobj/minuseless = 100 # maximal fraction of non-binary variables with non-zero objective without a bound reduction before aborted # [type: real, advanced: TRUE, range: [0,1], default: 0.1] propagating/pseudoobj/maxvarsfrac = 0.1 # whether to propagate all non-binary variables when we are propagating the root node # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/pseudoobj/propfullinroot = TRUE # propagate new cutoff bound directly globally # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/pseudoobj/propcutoffbound = TRUE # should the propagator be forced even if active pricer are present? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/pseudoobj/force = FALSE # number of variables added after the propagator is reinitialized? # [type: int, advanced: TRUE, range: [0,2147483647], default: 1000] propagating/pseudoobj/maxnewvars = 1000 # use implications to strengthen the propagation of binary variable (increasing the objective change)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/pseudoobj/propuseimplics = TRUE # use implications to strengthen the resolve propagation of binary variable (increasing the objective change)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/pseudoobj/respropuseimplics = TRUE # maximum number of binary variables the implications are used if turned on (-1: unlimited)? # [type: int, advanced: TRUE, range: [-1,2147483647], default: 50000] propagating/pseudoobj/maximplvars = 50000 # priority of propagator <redcost> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 1000000] propagating/redcost/priority = 1000000 # frequency for calling propagator <redcost> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] propagating/redcost/freq = 1 # should propagator be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/redcost/delay = FALSE # timing when propagator should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS)) # [type: int, advanced: TRUE, range: [1,15], default: 6] propagating/redcost/timingmask = 6 # presolving priority of propagator <redcost> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 0] propagating/redcost/presolpriority = 0 # maximal number of presolving rounds the propagator participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] propagating/redcost/maxprerounds = -1 # timing mask of the presolving method of propagator <redcost> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [2,60], default: 28] propagating/redcost/presoltiming = 28 # should reduced cost fixing be also applied to continuous variables? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] propagating/redcost/continuous = FALSE # should implications be used to strength the reduced cost for binary variables? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] propagating/redcost/useimplics = FALSE # should the propagator be forced even if active pricer are present? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/redcost/force = FALSE # priority of propagator <rootredcost> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 10000000] propagating/rootredcost/priority = 10000000 # frequency for calling propagator <rootredcost> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] propagating/rootredcost/freq = 1 # should propagator be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/rootredcost/delay = FALSE # timing when propagator should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS)) # [type: int, advanced: TRUE, range: [1,15], default: 5] propagating/rootredcost/timingmask = 5 # presolving priority of propagator <rootredcost> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 0] propagating/rootredcost/presolpriority = 0 # maximal number of presolving rounds the propagator participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] propagating/rootredcost/maxprerounds = -1 # timing mask of the presolving method of propagator <rootredcost> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [2,60], default: 28] propagating/rootredcost/presoltiming = 28 # should only binary variables be propagated? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/rootredcost/onlybinary = FALSE # should the propagator be forced even if active pricer are present? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/rootredcost/force = FALSE # priority of propagator <symmetry> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1000000] propagating/symmetry/priority = -1000000 # frequency for calling propagator <symmetry> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] propagating/symmetry/freq = 1 # should propagator be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/symmetry/delay = FALSE # timing when propagator should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS)) # [type: int, advanced: TRUE, range: [1,15], default: 1] propagating/symmetry/timingmask = 1 # presolving priority of propagator <symmetry> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -10000000] propagating/symmetry/presolpriority = -10000000 # maximal number of presolving rounds the propagator participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] propagating/symmetry/maxprerounds = -1 # timing mask of the presolving method of propagator <symmetry> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [2,60], default: 16] propagating/symmetry/presoltiming = 16 # is statistics table <symmetry> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/symmetry/active = TRUE # limit on the number of generators that should be produced within symmetry detection (0 = no limit) # [type: int, advanced: TRUE, range: [0,2147483647], default: 1500] propagating/symmetry/maxgenerators = 1500 # Should all symmetries be checked after computation? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/symmetry/checksymmetries = FALSE # Should the number of variables affected by some symmetry be displayed? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/symmetry/displaynorbitvars = FALSE # Double equations to positive/negative version? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/symmetry/doubleequations = FALSE # Should the symmetry breaking constraints be added to the LP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/symmetry/conssaddlp = TRUE # Add inequalities for symresacks for each generator? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/symmetry/addsymresacks = TRUE # Should we check whether the components of the symmetry group can be handled by double lex matrices? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/symmetry/detectdoublelex = TRUE # Should we check whether the components of the symmetry group can be handled by orbitopes? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/symmetry/detectorbitopes = TRUE # Should we try to detect symmetric subgroups of the symmetry group on binary variables? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/symmetry/detectsubgroups = TRUE # Should we add weak SBCs for enclosing orbit of symmetric subgroups? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/symmetry/addweaksbcs = TRUE # timing of adding constraints (0 = before presolving, 1 = during presolving, 2 = after presolving) [disabled parameter] # [type: int, advanced: TRUE, range: [0,2], default: 2] propagating/symmetry/addconsstiming = 2 # timing of symmetry computation (0 = before presolving, 1 = during presolving, 2 = at first call) [disabled parameter] # [type: int, advanced: TRUE, range: [0,2], default: 2] propagating/symmetry/ofsymcomptiming = 2 # run orbital fixing during presolving? (disabled) # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/symmetry/performpresolving = FALSE # recompute symmetries after a restart has occurred? (0 = never) # [type: int, advanced: TRUE, range: [0,0], default: 0] propagating/symmetry/recomputerestart = 0 # Should non-affected variables be removed from permutation to save memory? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/symmetry/compresssymmetries = TRUE # Compression is used if percentage of moved vars is at most the threshold. # [type: real, advanced: TRUE, range: [0,1], default: 0.5] propagating/symmetry/compressthreshold = 0.5 # Should the number of conss a variable is contained in be exploited in symmetry detection? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/symmetry/usecolumnsparsity = FALSE # maximum number of constraints up to which subgroup structures are detected # [type: int, advanced: TRUE, range: [0,2147483647], default: 500000] propagating/symmetry/maxnconsssubgroup = 500000 # whether dynamified symmetry handling constraint methods should be used # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/symmetry/usedynamicprop = TRUE # Should strong SBCs for enclosing orbit of symmetric subgroups be added if orbitopes are not used? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/symmetry/addstrongsbcs = FALSE # rule to select the orbit in Schreier Sims inequalities (variable in 0: minimum size orbit; 1: maximum size orbit; 2: orbit with most variables in conflict with leader) # [type: int, advanced: TRUE, range: [0,2], default: 1] propagating/symmetry/ssttiebreakrule = 1 # rule to select the leader in an orbit (0: first var; 1: last var; 2: var having most conflicting vars in orbit) # [type: int, advanced: TRUE, range: [0,2], default: 0] propagating/symmetry/sstleaderrule = 0 # bitset encoding which variable types can be leaders (1: binary; 2: integer; 4: impl. int; 8: continuous);if multiple types are allowed, take the one with most affected vars # [type: int, advanced: TRUE, range: [1,15], default: 14] propagating/symmetry/sstleadervartype = 14 # Should Schreier Sims constraints be added if we use a conflict based rule? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/symmetry/addconflictcuts = TRUE # Should Schreier Sims constraints be added? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/symmetry/sstaddcuts = TRUE # Should Schreier Sims constraints be added if a symmetry component contains variables of different types? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/symmetry/sstmixedcomponents = TRUE # Whether all non-binary variables shall be not affected by symmetries if OF is active? (disabled) # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/symmetry/symfixnonbinaryvars = FALSE # Is only symmetry on binary variables used? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/symmetry/enforcecomputesymmetry = FALSE # Shall orbitopes with less rows be preferred in detection? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/symmetry/preferlessrows = TRUE # Type of symmetries that shall be computed? # [type: int, advanced: TRUE, range: [0,1], default: 0] propagating/symmetry/symtype = 0 # timing of symmetry computation and handling (0 = before presolving, 1 = during presolving, 2 = after presolving) # [type: int, advanced: TRUE, range: [0,2], default: 2] propagating/symmetry/symtiming = 2 # The column ordering variant, respects enum SCIP_ColumnOrdering. # [type: int, advanced: TRUE, range: [0,4], default: 4] propagating/symmetry/orbitopalreduction/columnordering = 4 # priority of propagator <vbounds> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 3000000] propagating/vbounds/priority = 3000000 # frequency for calling propagator <vbounds> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 1] propagating/vbounds/freq = 1 # should propagator be delayed, if other propagators found reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/vbounds/delay = FALSE # timing when propagator should be called (1:BEFORELP, 2:DURINGLPLOOP, 4:AFTERLPLOOP, 15:ALWAYS)) # [type: int, advanced: TRUE, range: [1,15], default: 5] propagating/vbounds/timingmask = 5 # presolving priority of propagator <vbounds> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -90000] propagating/vbounds/presolpriority = -90000 # maximal number of presolving rounds the propagator participates in (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] propagating/vbounds/maxprerounds = -1 # timing mask of the presolving method of propagator <vbounds> (4:FAST, 8:MEDIUM, 16:EXHAUSTIVE, 32:FINAL) # [type: int, advanced: TRUE, range: [2,60], default: 24] propagating/vbounds/presoltiming = 24 # should bound widening be used to initialize conflict analysis? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] propagating/vbounds/usebdwidening = TRUE # should implications be propagated? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/vbounds/useimplics = FALSE # should cliques be propagated? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/vbounds/usecliques = FALSE # should vbounds be propagated? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] propagating/vbounds/usevbounds = TRUE # should the bounds be topologically sorted in advance? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] propagating/vbounds/dotoposort = TRUE # should cliques be regarded for the topological sort? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] propagating/vbounds/sortcliques = FALSE # should cycles in the variable bound graph be identified? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] propagating/vbounds/detectcycles = FALSE # minimum percentage of new cliques to trigger another clique table analysis # [type: real, advanced: FALSE, range: [0,1], default: 0.1] propagating/vbounds/minnewcliques = 0.1 # maximum number of cliques per variable to run clique table analysis in medium presolving # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 50] propagating/vbounds/maxcliquesmedium = 50 # maximum number of cliques per variable to run clique table analysis in exhaustive presolving # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 100] propagating/vbounds/maxcliquesexhaustive = 100 # priority of separator <cgmip> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1000] separating/cgmip/priority = -1000 # frequency for calling separator <cgmip> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] separating/cgmip/freq = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <cgmip> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 0] separating/cgmip/maxbounddist = 0 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/cgmip/delay = FALSE # base for exponential increase of frequency at which separator <cgmip> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/cgmip/expbackoff = 4 # maximal number of cgmip separation rounds per node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 5] separating/cgmip/maxrounds = 5 # maximal number of cgmip separation rounds in the root node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 50] separating/cgmip/maxroundsroot = 50 # maximal depth at which the separator is applied (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] separating/cgmip/maxdepth = -1 # Use decision tree to turn separation on/off? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/cgmip/decisiontree = FALSE # time limit for sub-MIP # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 1e+20] separating/cgmip/timelimit = 1e+20 # memory limit for sub-MIP # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 1e+20] separating/cgmip/memorylimit = 1e+20 # minimum number of nodes considered for sub-MIP (-1: unlimited) # [type: longint, advanced: FALSE, range: [-1,9223372036854775807], default: 500] separating/cgmip/minnodelimit = 500 # maximum number of nodes considered for sub-MIP (-1: unlimited) # [type: longint, advanced: FALSE, range: [-1,9223372036854775807], default: 5000] separating/cgmip/maxnodelimit = 5000 # bounds on the values of the coefficients in the CG-cut # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 1000] separating/cgmip/cutcoefbnd = 1000 # Use only active rows to generate cuts? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/cgmip/onlyactiverows = FALSE # maximal age of rows to consider if onlyactiverows is false # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] separating/cgmip/maxrowage = -1 # Separate only rank 1 inequalities w.r.t. CG-MIP separator? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/cgmip/onlyrankone = FALSE # Generate cuts for problems with only integer variables? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/cgmip/onlyintvars = FALSE # Convert some integral variables to be continuous to reduce the size of the sub-MIP? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/cgmip/contconvert = FALSE # fraction of integral variables converted to be continuous (if contconvert) # [type: real, advanced: FALSE, range: [0,1], default: 0.1] separating/cgmip/contconvfrac = 0.1 # minimum number of integral variables before some are converted to be continuous # [type: int, advanced: FALSE, range: [-1,2147483647], default: 100] separating/cgmip/contconvmin = 100 # Convert some integral variables attaining fractional values to have integral value? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/cgmip/intconvert = FALSE # fraction of frac. integral variables converted to have integral value (if intconvert) # [type: real, advanced: FALSE, range: [0,1], default: 0.1] separating/cgmip/intconvfrac = 0.1 # minimum number of integral variables before some are converted to have integral value # [type: int, advanced: FALSE, range: [-1,2147483647], default: 100] separating/cgmip/intconvmin = 100 # Skip the upper bounds on the multipliers in the sub-MIP? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/cgmip/skipmultbounds = TRUE # Should the objective of the sub-MIP minimize the l1-norm of the multipliers? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/cgmip/objlone = FALSE # weight used for the row combination coefficient in the sub-MIP objective # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.001] separating/cgmip/objweight = 0.001 # Weight each row by its size? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/cgmip/objweightsize = TRUE # should generated cuts be removed from the LP if they are no longer tight? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/cgmip/dynamiccuts = TRUE # use CMIR-generator (otherwise add cut directly)? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/cgmip/usecmir = TRUE # use strong CG-function to strengthen cut? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/cgmip/usestrongcg = FALSE # tell CMIR-generator which bounds to used in rounding? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/cgmip/cmirownbounds = FALSE # use cutpool to store CG-cuts even if the are not efficient? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/cgmip/usecutpool = TRUE # only separate cuts that are tight for the best feasible solution? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/cgmip/primalseparation = TRUE # terminate separation if a violated (but possibly sub-optimal) cut has been found? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/cgmip/earlyterm = TRUE # add constraint to subscip that only allows violated cuts (otherwise add obj. limit)? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/cgmip/addviolationcons = FALSE # add constraint handler to filter out violated cuts? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/cgmip/addviolconshdlr = FALSE # should the violation constraint handler use the norm of a cut to check for feasibility? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/cgmip/conshdlrusenorm = TRUE # Use upper bound on objective function (via primal solution)? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/cgmip/useobjub = FALSE # Use lower bound on objective function (via primal solution)? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/cgmip/useobjlb = FALSE # Should the settings for the sub-MIP be optimized for speed? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/cgmip/subscipfast = TRUE # Should information about the sub-MIP and cuts be displayed? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/cgmip/output = FALSE # Try to generate primal solutions from Gomory cuts? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/cgmip/genprimalsols = FALSE # priority of separator <clique> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -5000] separating/clique/priority = -5000 # frequency for calling separator <clique> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] separating/clique/freq = 0 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <clique> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 0] separating/clique/maxbounddist = 0 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/clique/delay = FALSE # base for exponential increase of frequency at which separator <clique> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/clique/expbackoff = 4 # factor for scaling weights # [type: real, advanced: TRUE, range: [1,1.79769313486232e+308], default: 1000] separating/clique/scaleval = 1000 # maximal number of nodes in branch and bound tree (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 10000] separating/clique/maxtreenodes = 10000 # frequency for premature backtracking up to tree level 1 (0: no backtracking) # [type: int, advanced: TRUE, range: [0,2147483647], default: 1000] separating/clique/backtrackfreq = 1000 # maximal number of clique cuts separated per separation round (-1: no limit) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 10] separating/clique/maxsepacuts = 10 # maximal number of zero-valued variables extending the clique (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 1000] separating/clique/maxzeroextensions = 1000 # maximal memory size of dense clique table (in kb) # [type: real, advanced: TRUE, range: [0,2097151.99902344], default: 20000] separating/clique/cliquetablemem = 20000 # minimal density of cliques to use a dense clique table # [type: real, advanced: TRUE, range: [0,1], default: 0] separating/clique/cliquedensity = 0 # priority of separator <closecuts> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 1000000] separating/closecuts/priority = 1000000 # frequency for calling separator <closecuts> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] separating/closecuts/freq = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <closecuts> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/closecuts/maxbounddist = 1 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/closecuts/delay = FALSE # base for exponential increase of frequency at which separator <closecuts> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/closecuts/expbackoff = 4 # generate close cuts w.r.t. relative interior point (best solution otherwise)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/closecuts/separelint = TRUE # convex combination value for close cuts # [type: real, advanced: TRUE, range: [0,1], default: 0.3] separating/closecuts/sepacombvalue = 0.3 # threshold on number of generated cuts below which the ordinary separation is started # [type: int, advanced: TRUE, range: [-1,2147483647], default: 50] separating/closecuts/closethres = 50 # include an objective cutoff when computing the relative interior? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/closecuts/inclobjcutoff = FALSE # recompute relative interior point in each separation call? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/closecuts/recomputerelint = FALSE # turn off separation in current node after unsuccessful calls (-1 never turn off) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 0] separating/closecuts/maxunsuccessful = 0 # factor for maximal LP iterations in relative interior computation compared to node LP iterations (negative for no limit) # [type: real, advanced: TRUE, range: [-1,1.79769313486232e+308], default: 10] separating/closecuts/maxlpiterfactor = 10 # priority of separator <flowcover> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -100000] separating/flowcover/priority = -100000 # frequency for calling separator <flowcover> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] separating/flowcover/freq = 10 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <flowcover> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 0] separating/flowcover/maxbounddist = 0 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/flowcover/delay = FALSE # base for exponential increase of frequency at which separator <flowcover> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/flowcover/expbackoff = 4 # priority of separator <cmir> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -100000] separating/cmir/priority = -100000 # frequency for calling separator <cmir> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] separating/cmir/freq = 10 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <cmir> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 0] separating/cmir/maxbounddist = 0 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/cmir/delay = FALSE # base for exponential increase of frequency at which separator <cmir> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/cmir/expbackoff = 4 # priority of separator <knapsackcover> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -100000] separating/knapsackcover/priority = -100000 # frequency for calling separator <knapsackcover> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] separating/knapsackcover/freq = 10 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <knapsackcover> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 0] separating/knapsackcover/maxbounddist = 0 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/knapsackcover/delay = FALSE # base for exponential increase of frequency at which separator <knapsackcover> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/knapsackcover/expbackoff = 4 # priority of separator <aggregation> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -3000] separating/aggregation/priority = -3000 # frequency for calling separator <aggregation> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] separating/aggregation/freq = 10 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <aggregation> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/aggregation/maxbounddist = 1 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/aggregation/delay = FALSE # base for exponential increase of frequency at which separator <aggregation> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/aggregation/expbackoff = 4 # maximal number of cmir separation rounds per node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] separating/aggregation/maxrounds = -1 # maximal number of cmir separation rounds in the root node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] separating/aggregation/maxroundsroot = -1 # maximal number of rows to start aggregation with per separation round (-1: unlimited) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 200] separating/aggregation/maxtries = 200 # maximal number of rows to start aggregation with per separation round in the root node (-1: unlimited) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] separating/aggregation/maxtriesroot = -1 # maximal number of consecutive unsuccessful aggregation tries (-1: unlimited) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 20] separating/aggregation/maxfails = 20 # maximal number of consecutive unsuccessful aggregation tries in the root node (-1: unlimited) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 100] separating/aggregation/maxfailsroot = 100 # maximal number of aggregations for each row per separation round # [type: int, advanced: TRUE, range: [0,2147483647], default: 3] separating/aggregation/maxaggrs = 3 # maximal number of aggregations for each row per separation round in the root node # [type: int, advanced: TRUE, range: [0,2147483647], default: 6] separating/aggregation/maxaggrsroot = 6 # maximal number of cmir cuts separated per separation round # [type: int, advanced: FALSE, range: [0,2147483647], default: 100] separating/aggregation/maxsepacuts = 100 # maximal number of cmir cuts separated per separation round in the root node # [type: int, advanced: FALSE, range: [0,2147483647], default: 500] separating/aggregation/maxsepacutsroot = 500 # maximal slack of rows to be used in aggregation # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] separating/aggregation/maxslack = 0 # maximal slack of rows to be used in aggregation in the root node # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.1] separating/aggregation/maxslackroot = 0.1 # weight of row density in the aggregation scoring of the rows # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.0001] separating/aggregation/densityscore = 0.0001 # weight of slack in the aggregation scoring of the rows # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.001] separating/aggregation/slackscore = 0.001 # maximal density of aggregated row # [type: real, advanced: TRUE, range: [0,1], default: 0.2] separating/aggregation/maxaggdensity = 0.2 # maximal density of row to be used in aggregation # [type: real, advanced: TRUE, range: [0,1], default: 0.05] separating/aggregation/maxrowdensity = 0.05 # additional number of variables allowed in row on top of density # [type: int, advanced: TRUE, range: [0,2147483647], default: 100] separating/aggregation/densityoffset = 100 # maximal row aggregation factor # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 10000] separating/aggregation/maxrowfac = 10000 # maximal number of different deltas to try (-1: unlimited) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] separating/aggregation/maxtestdelta = -1 # tolerance for bound distances used to select continuous variable in current aggregated constraint to be eliminated # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.01] separating/aggregation/aggrtol = 0.01 # should negative values also be tested in scaling? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/aggregation/trynegscaling = TRUE # should an additional variable be complemented if f0 = 0? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/aggregation/fixintegralrhs = TRUE # should generated cuts be removed from the LP if they are no longer tight? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/aggregation/dynamiccuts = TRUE # priority of separator <convexproj> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 0] separating/convexproj/priority = 0 # frequency for calling separator <convexproj> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] separating/convexproj/freq = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <convexproj> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/convexproj/maxbounddist = 1 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/convexproj/delay = TRUE # base for exponential increase of frequency at which separator <convexproj> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/convexproj/expbackoff = 4 # maximal depth at which the separator is applied (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] separating/convexproj/maxdepth = -1 # iteration limit of NLP solver; 0 for no limit # [type: int, advanced: TRUE, range: [0,2147483647], default: 250] separating/convexproj/nlpiterlimit = 250 # priority of separator <disjunctive> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 10] separating/disjunctive/priority = 10 # frequency for calling separator <disjunctive> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] separating/disjunctive/freq = 0 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <disjunctive> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 0] separating/disjunctive/maxbounddist = 0 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/disjunctive/delay = TRUE # base for exponential increase of frequency at which separator <disjunctive> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/disjunctive/expbackoff = 4 # strengthen cut if integer variables are present. # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/disjunctive/strengthen = TRUE # node depth of separating bipartite disjunctive cuts (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] separating/disjunctive/maxdepth = -1 # maximal number of separation rounds per iteration in a branching node (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 25] separating/disjunctive/maxrounds = 25 # maximal number of separation rounds in the root node (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 100] separating/disjunctive/maxroundsroot = 100 # maximal number of cuts investigated per iteration in a branching node # [type: int, advanced: TRUE, range: [0,2147483647], default: 50] separating/disjunctive/maxinvcuts = 50 # maximal number of cuts investigated per iteration in the root node # [type: int, advanced: TRUE, range: [0,2147483647], default: 250] separating/disjunctive/maxinvcutsroot = 250 # delay separation if number of conflict graph edges is larger than predefined value (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 100000] separating/disjunctive/maxconfsdelay = 100000 # maximal rank of a disj. cut that could not be scaled to integral coefficients (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 20] separating/disjunctive/maxrank = 20 # maximal rank of a disj. cut that could be scaled to integral coefficients (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] separating/disjunctive/maxrankintegral = -1 # maximal valid range max(|weights|)/min(|weights|) of row weights # [type: real, advanced: TRUE, range: [1,1.79769313486232e+308], default: 1000] separating/disjunctive/maxweightrange = 1000 # priority of separator <eccuts> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -13000] separating/eccuts/priority = -13000 # frequency for calling separator <eccuts> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] separating/eccuts/freq = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <eccuts> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/eccuts/maxbounddist = 1 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/eccuts/delay = FALSE # base for exponential increase of frequency at which separator <eccuts> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/eccuts/expbackoff = 4 # should generated cuts be removed from the LP if they are no longer tight? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/eccuts/dynamiccuts = TRUE # maximal number of eccuts separation rounds per node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 10] separating/eccuts/maxrounds = 10 # maximal number of eccuts separation rounds in the root node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 250] separating/eccuts/maxroundsroot = 250 # maximal depth at which the separator is applied (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] separating/eccuts/maxdepth = -1 # maximal number of edge-concave cuts separated per separation round # [type: int, advanced: FALSE, range: [0,2147483647], default: 10] separating/eccuts/maxsepacuts = 10 # maximal number of edge-concave cuts separated per separation round in the root node # [type: int, advanced: FALSE, range: [0,2147483647], default: 50] separating/eccuts/maxsepacutsroot = 50 # maximal coef. range of a cut (max coef. divided by min coef.) in order to be added to LP relaxation # [type: real, advanced: FALSE, range: [0,1e+20], default: 10000000] separating/eccuts/cutmaxrange = 10000000 # minimal violation of an edge-concave cut to be separated # [type: real, advanced: FALSE, range: [0,0.5], default: 0.3] separating/eccuts/minviolation = 0.3 # search for edge-concave aggregations of at least this size # [type: int, advanced: TRUE, range: [3,5], default: 3] separating/eccuts/minaggrsize = 3 # search for edge-concave aggregations of at most this size # [type: int, advanced: TRUE, range: [3,5], default: 4] separating/eccuts/maxaggrsize = 4 # maximum number of bilinear terms allowed to be in a quadratic constraint # [type: int, advanced: TRUE, range: [0,2147483647], default: 500] separating/eccuts/maxbilinterms = 500 # maximum number of unsuccessful rounds in the edge-concave aggregation search # [type: int, advanced: TRUE, range: [0,2147483647], default: 5] separating/eccuts/maxstallrounds = 5 # priority of separator <gauge> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 0] separating/gauge/priority = 0 # frequency for calling separator <gauge> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] separating/gauge/freq = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <gauge> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/gauge/maxbounddist = 1 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/gauge/delay = FALSE # base for exponential increase of frequency at which separator <gauge> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/gauge/expbackoff = 4 # iteration limit of NLP solver; 0 for no limit # [type: int, advanced: TRUE, range: [0,2147483647], default: 1000] separating/gauge/nlpiterlimit = 1000 # priority of separator <gomory> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1000] separating/gomory/priority = -1000 # frequency for calling separator <gomory> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] separating/gomory/freq = 10 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <gomory> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/gomory/maxbounddist = 1 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/gomory/delay = FALSE # base for exponential increase of frequency at which separator <gomory> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/gomory/expbackoff = 4 # priority of separator <strongcg> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -100000] separating/strongcg/priority = -100000 # frequency for calling separator <strongcg> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] separating/strongcg/freq = 10 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <strongcg> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 0] separating/strongcg/maxbounddist = 0 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/strongcg/delay = FALSE # base for exponential increase of frequency at which separator <strongcg> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/strongcg/expbackoff = 4 # priority of separator <gomorymi> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -100000] separating/gomorymi/priority = -100000 # frequency for calling separator <gomorymi> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] separating/gomorymi/freq = 10 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <gomorymi> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 0] separating/gomorymi/maxbounddist = 0 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/gomorymi/delay = FALSE # base for exponential increase of frequency at which separator <gomorymi> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/gomorymi/expbackoff = 4 # maximal number of gomory separation rounds per node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 5] separating/gomory/maxrounds = 5 # maximal number of gomory separation rounds in the root node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 10] separating/gomory/maxroundsroot = 10 # maximal number of gomory cuts separated per separation round # [type: int, advanced: FALSE, range: [0,2147483647], default: 50] separating/gomory/maxsepacuts = 50 # maximal number of gomory cuts separated per separation round in the root node # [type: int, advanced: FALSE, range: [0,2147483647], default: 200] separating/gomory/maxsepacutsroot = 200 # maximal rank of a gomory cut that could not be scaled to integral coefficients (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] separating/gomory/maxrank = -1 # maximal rank of a gomory cut that could be scaled to integral coefficients (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] separating/gomory/maxrankintegral = -1 # minimal integrality violation of a basis variable in order to try Gomory cut # [type: real, advanced: FALSE, range: [0.0001,0.5], default: 0.01] separating/gomory/away = 0.01 # should generated cuts be removed from the LP if they are no longer tight? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/gomory/dynamiccuts = TRUE # try to scale cuts to integral coefficients # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/gomory/makeintegral = FALSE # if conversion to integral coefficients failed still consider the cut # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/gomory/forcecuts = TRUE # separate rows with integral slack # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/gomory/separaterows = TRUE # should cuts be added to the delayed cut pool? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/gomory/delayedcuts = FALSE # choose side types of row (lhs/rhs) based on basis information? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/gomory/sidetypebasis = TRUE # try to generate strengthened Chvatal-Gomory cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/gomory/trystrongcg = TRUE # Should both Gomory and strong CG cuts be generated (otherwise take best)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/gomory/genbothgomscg = TRUE # priority of separator <impliedbounds> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -50] separating/impliedbounds/priority = -50 # frequency for calling separator <impliedbounds> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] separating/impliedbounds/freq = 10 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <impliedbounds> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/impliedbounds/maxbounddist = 1 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/impliedbounds/delay = FALSE # base for exponential increase of frequency at which separator <impliedbounds> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/impliedbounds/expbackoff = 4 # should violated inequalities for cliques with 2 variables be separated? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/impliedbounds/usetwosizecliques = TRUE # priority of separator <interminor> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 0] separating/interminor/priority = 0 # frequency for calling separator <interminor> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] separating/interminor/freq = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <interminor> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/interminor/maxbounddist = 1 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/interminor/delay = FALSE # base for exponential increase of frequency at which separator <interminor> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/interminor/expbackoff = 4 # whether to use strengthened intersection cuts to separate minors # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/interminor/usestrengthening = FALSE # whether to also enforce nonegativity bounds of principle minors # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/interminor/usebounds = FALSE # minimum required violation of a cut # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.0001] separating/interminor/mincutviol = 0.0001 # maximal number of separation rounds per node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 10] separating/interminor/maxrounds = 10 # maximal number of separation rounds in the root node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] separating/interminor/maxroundsroot = -1 # priority of separator <intobj> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -100] separating/intobj/priority = -100 # frequency for calling separator <intobj> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] separating/intobj/freq = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <intobj> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 0] separating/intobj/maxbounddist = 0 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/intobj/delay = FALSE # base for exponential increase of frequency at which separator <intobj> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/intobj/expbackoff = 4 # priority of separator <lagromory> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -8000] separating/lagromory/priority = -8000 # frequency for calling separator <lagromory> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] separating/lagromory/freq = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <lagromory> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/lagromory/maxbounddist = 1 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/lagromory/delay = FALSE # base for exponential increase of frequency at which separator <lagromory> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/lagromory/expbackoff = 4 # minimal integrality violation of a basis variable to try separation # [type: real, advanced: FALSE, range: [0,1], default: 0.01] separating/lagromory/away = 0.01 # factor w.r.t. root node LP iterations for maximal separating LP iterations in the root node (negative for no limit) # [type: real, advanced: TRUE, range: [-1,1.79769313486232e+308], default: -1] separating/lagromory/rootlpiterlimitfactor = -1 # factor w.r.t. root node LP iterations for maximal separating LP iterations in the tree (negative for no limit) # [type: real, advanced: TRUE, range: [-1,1.79769313486232e+308], default: -1] separating/lagromory/totallpiterlimitfactor = -1 # factor w.r.t. root node LP iterations for maximal separating LP iterations per separation round (negative for no limit) # [type: real, advanced: TRUE, range: [-1,1.79769313486232e+308], default: -1] separating/lagromory/perroundlpiterlimitfactor = -1 # factor w.r.t. number of integer columns for number of cuts separated per separation round in root node # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 1] separating/lagromory/perroundcutsfactorroot = 1 # factor w.r.t. number of integer columns for number of cuts separated per separation round at a non-root node # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.5] separating/lagromory/perroundcutsfactor = 0.5 # factor w.r.t. number of integer columns for total number of cuts separated # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 50] separating/lagromory/totalcutsfactor = 50 # initial value of the mu parameter (factor for step length) # [type: real, advanced: TRUE, range: [0,100], default: 0.01] separating/lagromory/muparaminit = 0.01 # lower bound of the mu parameter (factor for step length) # [type: real, advanced: TRUE, range: [0,1], default: 0] separating/lagromory/muparamlb = 0 # upper bound of the mu parameter (factor for step length) # [type: real, advanced: TRUE, range: [1,10], default: 2] separating/lagromory/muparamub = 2 # factor of mu while backtracking the mu parameter (factor for step length) # [type: real, advanced: TRUE, range: [0,1], default: 0.5] separating/lagromory/mubacktrackfactor = 0.5 # factor of mu parameter (factor for step length) for larger increment # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 10] separating/lagromory/muslab1factor = 10 # factor of mu parameter (factor for step length) for smaller increment # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 2] separating/lagromory/muslab2factor = 2 # factor of mu parameter (factor for step length) for reduction # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.5] separating/lagromory/muslab3factor = 0.5 # factor of delta deciding larger increment of mu parameter (factor for step length) # [type: real, advanced: TRUE, range: [0,1], default: 0.001] separating/lagromory/deltaslab1ub = 0.001 # factor of delta deciding smaller increment of mu parameter (factor for step length) # [type: real, advanced: TRUE, range: [0,1], default: 0.01] separating/lagromory/deltaslab2ub = 0.01 # factor for positive upper bound used as an estimate for the optimal Lagrangian dual value # [type: real, advanced: TRUE, range: [1,100], default: 2] separating/lagromory/ubparamposfactor = 2 # factor for negative upper bound used as an estimate for the optimal Lagrangian dual value # [type: real, advanced: TRUE, range: [0,1], default: 0.5] separating/lagromory/ubparamnegfactor = 0.5 # factor w.r.t. root node LP iterations for iteration limit of each separating LP (negative for no limit) # [type: real, advanced: TRUE, range: [-1,1.79769313486232e+308], default: 0.2] separating/lagromory/perrootlpiterfactor = 0.2 # factor w.r.t. non-root node LP iterations for iteration limit of each separating LP (negative for no limit) # [type: real, advanced: TRUE, range: [-1,1.79769313486232e+308], default: 0.1] separating/lagromory/perlpiterfactor = 0.1 # fraction of generated cuts per explored basis to accept from separator # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/lagromory/cutsfilterfactor = 1 # initial radius of the ball used in stabilization of Lagrangian multipliers # [type: real, advanced: TRUE, range: [0,1], default: 0.5] separating/lagromory/radiusinit = 0.5 # maximum radius of the ball used in stabilization of Lagrangian multipliers # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 20] separating/lagromory/radiusmax = 20 # minimum radius of the ball used in stabilization of Lagrangian multipliers # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 1e-06] separating/lagromory/radiusmin = 1e-06 # a constant for stablity center based stabilization of Lagrangian multipliers # [type: real, advanced: TRUE, range: [2,1.79769313486232e+308], default: 2] separating/lagromory/constant = 2 # multiplier to evaluate cut violation score used for updating ball radius # [type: real, advanced: TRUE, range: [0,1], default: 0.98] separating/lagromory/radiusupdateweight = 0.98 # minimum dual degeneracy rate for separator execution # [type: real, advanced: FALSE, range: [0,1], default: 0.5] separating/lagromory/dualdegeneracyratethreshold = 0.5 # minimum variable-constraint ratio on optimal face for separator execution # [type: real, advanced: FALSE, range: [1,1.79769313486232e+308], default: 1] separating/lagromory/varconsratiothreshold = 1 # is the mu parameter (factor for step length) constant? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/lagromory/muparamconst = TRUE # separate rows with integral slack? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/lagromory/separaterows = TRUE # sort fractional integer columnsbased on fractionality? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/lagromory/sortcutoffsol = TRUE # choose side types of row (lhs/rhs) based on basis information? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/lagromory/sidetypebasis = TRUE # should generated cuts be removed from LP if they are no longer tight? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/lagromory/dynamiccuts = TRUE # try to scale all cuts to integral coefficients? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/lagromory/makeintegral = FALSE # force cuts to be added to the LP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/lagromory/forcecuts = FALSE # should cuts be added to the delayed cut pool # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/lagromory/delayedcuts = FALSE # should locally valid cuts be generated? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/lagromory/allowlocal = FALSE # aggregate all generated cuts using the Lagrangian multipliers? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/lagromory/aggregatecuts = TRUE # maximal number of separation rounds per node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1] separating/lagromory/maxrounds = 1 # maximal number of separation rounds in the root node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1] separating/lagromory/maxroundsroot = 1 # maximal number of separating LP iterations per separation round (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 50000] separating/lagromory/perroundnmaxlpiters = 50000 # maximal number of cuts separated per Lagromory LP in the non-root node # [type: int, advanced: FALSE, range: [0,2147483647], default: 10] separating/lagromory/nmaxcutsperlp = 10 # maximal number of cuts separated per Lagromory LP in the root node # [type: int, advanced: FALSE, range: [0,2147483647], default: 50] separating/lagromory/nmaxcutsperlproot = 50 # maximal number of main loop iterations of the relax-and-cut algorithm # [type: int, advanced: TRUE, range: [0,2147483647], default: 4] separating/lagromory/nmaxmainiters = 4 # maximal number of subgradient loop iterations of the relax-and-cut algorithm # [type: int, advanced: TRUE, range: [0,2147483647], default: 6] separating/lagromory/nmaxsubgradientiters = 6 # frequency of subgradient iterations for generating cuts # [type: int, advanced: TRUE, range: [0,2147483647], default: 1] separating/lagromory/cutgenfreq = 1 # frequency of subgradient iterations for adding cuts to objective function # [type: int, advanced: TRUE, range: [0,2147483647], default: 1] separating/lagromory/cutaddfreq = 1 # maximal number of iterations for rolling average of Lagrangian value # [type: int, advanced: TRUE, range: [0,2147483647], default: 2] separating/lagromory/nmaxlagrangianvalsforavg = 2 # consecutive number of iterations used to determine if mu needs to be backtracked # [type: int, advanced: TRUE, range: [0,2147483647], default: 10] separating/lagromory/nmaxconsecitersformuupdate = 10 # the ball into which the Lagrangian multipliers are projected for stabilization (0: no projection, 1: L1-norm ball projection, 2: L2-norm ball projection, 3: L_inf-norm ball projection) # [type: int, advanced: TRUE, range: [0,3], default: 2] separating/lagromory/projectiontype = 2 # type of stability center for taking weighted average of Lagrangian multipliers for stabilization (0: no weighted stabilization, 1: best Lagrangian multipliers) # [type: int, advanced: TRUE, range: [0,1], default: 1] separating/lagromory/stabilitycentertype = 1 # priority of the optimal face for separator execution (0: low priority, 1: medium priority, 2: high priority) # [type: int, advanced: TRUE, range: [0,2], default: 2] separating/lagromory/optimalfacepriority = 2 # minimum restart round for separator execution (0: from beginning of the instance solving, >= n with n >= 1: from restart round n) # [type: int, advanced: TRUE, range: [0,2147483647], default: 1] separating/lagromory/minrestart = 1 # priority of separator <mcf> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -10000] separating/mcf/priority = -10000 # frequency for calling separator <mcf> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] separating/mcf/freq = 0 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <mcf> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 0] separating/mcf/maxbounddist = 0 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/mcf/delay = FALSE # base for exponential increase of frequency at which separator <mcf> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/mcf/expbackoff = 4 # number of clusters to generate in the shrunken network -- default separation # [type: int, advanced: TRUE, range: [2,32], default: 5] separating/mcf/nclusters = 5 # maximal valid range max(|weights|)/min(|weights|) of row weights # [type: real, advanced: TRUE, range: [1,1.79769313486232e+308], default: 1000000] separating/mcf/maxweightrange = 1000000 # maximal number of different deltas to try (-1: unlimited) -- default separation # [type: int, advanced: TRUE, range: [-1,2147483647], default: 20] separating/mcf/maxtestdelta = 20 # should negative values also be tested in scaling? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/mcf/trynegscaling = FALSE # should an additional variable be complemented if f0 = 0? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/mcf/fixintegralrhs = TRUE # should generated cuts be removed from the LP if they are no longer tight? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/mcf/dynamiccuts = TRUE # model type of network (0: auto, 1:directed, 2:undirected) # [type: int, advanced: TRUE, range: [0,2], default: 0] separating/mcf/modeltype = 0 # maximal number of mcf cuts separated per separation round # [type: int, advanced: FALSE, range: [-1,2147483647], default: 100] separating/mcf/maxsepacuts = 100 # maximal number of mcf cuts separated per separation round in the root node -- default separation # [type: int, advanced: FALSE, range: [-1,2147483647], default: 200] separating/mcf/maxsepacutsroot = 200 # maximum inconsistency ratio for separation at all # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.02] separating/mcf/maxinconsistencyratio = 0.02 # maximum inconsistency ratio of arcs not to be deleted # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.5] separating/mcf/maxarcinconsistencyratio = 0.5 # should we separate only if the cuts shores are connected? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/mcf/checkcutshoreconnectivity = TRUE # should we separate inequalities based on single-node cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/mcf/separatesinglenodecuts = TRUE # should we separate flowcutset inequalities on the network cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/mcf/separateflowcutset = TRUE # should we separate knapsack cover inequalities on the network cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/mcf/separateknapsack = TRUE # priority of separator <minor> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 0] separating/minor/priority = 0 # frequency for calling separator <minor> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] separating/minor/freq = 10 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <minor> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/minor/maxbounddist = 1 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/minor/delay = FALSE # base for exponential increase of frequency at which separator <minor> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/minor/expbackoff = 4 # constant for the maximum number of minors, i.e., max(const, fac * # quadratic terms) # [type: int, advanced: FALSE, range: [0,2147483647], default: 3000] separating/minor/maxminorsconst = 3000 # factor for the maximum number of minors, i.e., max(const, fac * # quadratic terms) # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 10] separating/minor/maxminorsfac = 10 # minimum required violation of a cut # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 0.0001] separating/minor/mincutviol = 0.0001 # maximal number of separation rounds per node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 10] separating/minor/maxrounds = 10 # maximal number of separation rounds in the root node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] separating/minor/maxroundsroot = -1 # whether to ignore circle packing constraints during minor detection # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/minor/ignorepackingconss = TRUE # priority of separator <mixing> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -50] separating/mixing/priority = -50 # frequency for calling separator <mixing> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] separating/mixing/freq = 10 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <mixing> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/mixing/maxbounddist = 1 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/mixing/delay = FALSE # base for exponential increase of frequency at which separator <mixing> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/mixing/expbackoff = 4 # Should local bounds be used? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/mixing/uselocalbounds = FALSE # Should general integer variables be used to generate cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/mixing/iscutsonints = FALSE # maximal number of mixing separation rounds per node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] separating/mixing/maxrounds = -1 # maximal number of mixing separation rounds in the root node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] separating/mixing/maxroundsroot = -1 # maximal number of consecutive unsuccessful iterations # [type: int, advanced: FALSE, range: [-1,2147483647], default: 10] separating/mixing/maxnunsuccessful = 10 # priority of separator <oddcycle> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -15000] separating/oddcycle/priority = -15000 # frequency for calling separator <oddcycle> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] separating/oddcycle/freq = -1 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <oddcycle> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/oddcycle/maxbounddist = 1 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/oddcycle/delay = FALSE # base for exponential increase of frequency at which separator <oddcycle> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/oddcycle/expbackoff = 4 # Should the search method by Groetschel, Lovasz, Schrijver be used? Otherwise use levelgraph method by Hoffman, Padberg. # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/oddcycle/usegls = TRUE # Should odd cycle cuts be lifted? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/oddcycle/liftoddcycles = FALSE # maximal number of oddcycle cuts separated per separation round # [type: int, advanced: FALSE, range: [0,2147483647], default: 5000] separating/oddcycle/maxsepacuts = 5000 # maximal number of oddcycle cuts separated per separation round in the root node # [type: int, advanced: FALSE, range: [0,2147483647], default: 5000] separating/oddcycle/maxsepacutsroot = 5000 # maximal number of oddcycle separation rounds per node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 10] separating/oddcycle/maxrounds = 10 # maximal number of oddcycle separation rounds in the root node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 10] separating/oddcycle/maxroundsroot = 10 # factor for scaling of the arc-weights # [type: int, advanced: TRUE, range: [1,2147483647], default: 1000] separating/oddcycle/scalingfactor = 1000 # add links between a variable and its negated # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/oddcycle/addselfarcs = TRUE # try to repair violated cycles with double appearance of a variable # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/oddcycle/repaircycles = TRUE # separate triangles found as 3-cycles or repaired larger cycles # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/oddcycle/includetriangles = TRUE # Even if a variable is already covered by a cut, still try it as start node for a cycle search? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/oddcycle/multiplecuts = FALSE # Even if a variable is already covered by a cut, still allow another cut to cover it too? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/oddcycle/allowmultiplecuts = TRUE # Choose lifting candidate by coef*lpvalue or only by coef? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/oddcycle/lpliftcoef = FALSE # Calculate lifting coefficient of every candidate in every step (or only if its chosen)? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/oddcycle/recalcliftcoef = TRUE # use sorted variable array (unsorted(0), maxlp(1), minlp(2), maxfrac(3), minfrac(4)) # [type: int, advanced: TRUE, range: [0,4], default: 3] separating/oddcycle/sortswitch = 3 # sort level of the root neighbors by fractionality (maxfrac) # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/oddcycle/sortrootneighbors = TRUE # percentage of variables to try the chosen method on [0-100] # [type: int, advanced: TRUE, range: [0,100], default: 0] separating/oddcycle/percenttestvars = 0 # offset of variables to try the chosen method on (additional to the percentage of testvars) # [type: int, advanced: TRUE, range: [0,2147483647], default: 100] separating/oddcycle/offsettestvars = 100 # percentage of nodes allowed in the same level of the level graph [0-100] # [type: int, advanced: TRUE, range: [0,100], default: 100] separating/oddcycle/maxpernodeslevel = 100 # offset of nodes allowed in the same level of the level graph (additional to the percentage of levelnodes) # [type: int, advanced: TRUE, range: [0,2147483647], default: 10] separating/oddcycle/offsetnodeslevel = 10 # maximal number of levels in level graph # [type: int, advanced: TRUE, range: [0,2147483647], default: 20] separating/oddcycle/maxnlevels = 20 # maximal number of oddcycle cuts generated per chosen variable as root of the level graph # [type: int, advanced: TRUE, range: [0,2147483647], default: 1] separating/oddcycle/maxcutsroot = 1 # maximal number of oddcycle cuts generated in every level of the level graph # [type: int, advanced: TRUE, range: [0,2147483647], default: 50] separating/oddcycle/maxcutslevel = 50 # minimal weight on an edge (in level graph or bipartite graph) # [type: int, advanced: TRUE, range: [0,2147483647], default: 0] separating/oddcycle/maxreference = 0 # number of unsuccessful calls at current node # [type: int, advanced: TRUE, range: [0,2147483647], default: 3] separating/oddcycle/maxunsucessfull = 3 # maximal number of other cuts s.t. separation is applied (-1 for direct call) # [type: int, advanced: TRUE, range: [-1,2147483647], default: -1] separating/oddcycle/cutthreshold = -1 # priority of separator <rapidlearning> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -1200000] separating/rapidlearning/priority = -1200000 # frequency for calling separator <rapidlearning> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 5] separating/rapidlearning/freq = 5 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <rapidlearning> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/rapidlearning/maxbounddist = 1 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/rapidlearning/delay = FALSE # base for exponential increase of frequency at which separator <rapidlearning> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/rapidlearning/expbackoff = 4 # should the found conflicts be applied in the original SCIP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/rapidlearning/applyconflicts = TRUE # should the found global bound deductions be applied in the original SCIP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/rapidlearning/applybdchgs = TRUE # should the inference values be used as initialization in the original SCIP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/rapidlearning/applyinfervals = TRUE # should the inference values only be used when rapidlearning found other reductions? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/rapidlearning/reducedinfer = FALSE # should the incumbent solution be copied to the original SCIP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/rapidlearning/applyprimalsol = TRUE # should a solved status be copied to the original SCIP? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/rapidlearning/applysolved = TRUE # should local LP degeneracy be checked? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/rapidlearning/checkdegeneracy = TRUE # should the progress on the dual bound be checked? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/rapidlearning/checkdualbound = FALSE # should the ratio of leaves proven to be infeasible and exceeding the cutoff bound be checked? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/rapidlearning/checkleaves = FALSE # check whether rapid learning should be executed # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/rapidlearning/checkexec = TRUE # should the (local) objective function be checked? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/rapidlearning/checkobj = FALSE # should the number of solutions found so far be checked? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/rapidlearning/checknsols = TRUE # should rapid learning be applied when there are continuous variables? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/rapidlearning/contvars = FALSE # maximal portion of continuous variables to apply rapid learning # [type: real, advanced: TRUE, range: [0,1], default: 0.3] separating/rapidlearning/contvarsquot = 0.3 # maximal fraction of LP iterations compared to node LP iterations # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.2] separating/rapidlearning/lpiterquot = 0.2 # minimal degeneracy threshold to allow local rapid learning # [type: real, advanced: TRUE, range: [0,1], default: 0.7] separating/rapidlearning/mindegeneracy = 0.7 # minimal threshold of inf/obj leaves to allow local rapid learning # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 10] separating/rapidlearning/mininflpratio = 10 # minimal ratio of unfixed variables in relation to basis size to allow local rapid learning # [type: real, advanced: TRUE, range: [1,1.79769313486232e+308], default: 2] separating/rapidlearning/minvarconsratio = 2 # maximum problem size (variables) for which rapid learning will be called # [type: int, advanced: TRUE, range: [0,2147483647], default: 10000] separating/rapidlearning/maxnvars = 10000 # maximum problem size (constraints) for which rapid learning will be called # [type: int, advanced: TRUE, range: [0,2147483647], default: 10000] separating/rapidlearning/maxnconss = 10000 # maximum number of overall calls # [type: int, advanced: TRUE, range: [0,2147483647], default: 100] separating/rapidlearning/maxcalls = 100 # maximum number of nodes considered in rapid learning run # [type: int, advanced: TRUE, range: [0,2147483647], default: 5000] separating/rapidlearning/maxnodes = 5000 # minimum number of nodes considered in rapid learning run # [type: int, advanced: TRUE, range: [0,2147483647], default: 500] separating/rapidlearning/minnodes = 500 # number of nodes that should be processed before rapid learning is executed locally based on the progress of the dualbound # [type: longint, advanced: TRUE, range: [0,9223372036854775807], default: 100] separating/rapidlearning/nwaitingnodes = 100 # should all active cuts from cutpool be copied to constraints in subproblem? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] separating/rapidlearning/copycuts = TRUE # priority of separator <rlt> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 10] separating/rlt/priority = 10 # frequency for calling separator <rlt> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 0] separating/rlt/freq = 0 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <rlt> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/rlt/maxbounddist = 1 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/rlt/delay = FALSE # base for exponential increase of frequency at which separator <rlt> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/rlt/expbackoff = 4 # maximal number of rlt-cuts that are added per round (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: -1] separating/rlt/maxncuts = -1 # maximal number of unknown bilinear terms a row is still used with (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 0] separating/rlt/maxunknownterms = 0 # maximal number of variables used to compute rlt cuts (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 100] separating/rlt/maxusedvars = 100 # maximal number of separation rounds per node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1] separating/rlt/maxrounds = 1 # maximal number of separation rounds in the root node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 10] separating/rlt/maxroundsroot = 10 # if set to true, only equality rows are used for rlt cuts # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/rlt/onlyeqrows = FALSE # if set to true, only continuous rows are used for rlt cuts # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/rlt/onlycontrows = FALSE # if set to true, only original rows and variables are used # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/rlt/onlyoriginal = TRUE # if set to true, rlt is also used in sub-scips # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/rlt/useinsubscip = FALSE # if set to true, projected rows are checked first # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/rlt/useprojection = FALSE # if set to true, hidden products are detected and separated by McCormick cuts # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/rlt/detecthidden = FALSE # whether RLT cuts (TRUE) or only McCormick inequalities (FALSE) should be added for hidden products # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] separating/rlt/hiddenrlt = FALSE # if set to true, globally valid RLT cuts are added to the global cut pool # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/rlt/addtopool = TRUE # threshold for score of cut relative to best score to be considered good, so that less strict filtering is applied # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/rlt/goodscore = 1 # threshold for score of cut relative to best score to be discarded # [type: real, advanced: TRUE, range: [0,1], default: 0.5] separating/rlt/badscore = 0.5 # weight of objective parallelism in cut score calculation # [type: real, advanced: TRUE, range: [0,1], default: 0] separating/rlt/objparalweight = 0 # weight of efficacy in cut score calculation # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/rlt/efficacyweight = 1 # weight of directed cutoff distance in cut score calculation # [type: real, advanced: TRUE, range: [0,1], default: 0] separating/rlt/dircutoffdistweight = 0 # maximum parallelism for good cuts # [type: real, advanced: TRUE, range: [0,1], default: 0.1] separating/rlt/goodmaxparall = 0.1 # maximum parallelism for non-good cuts # [type: real, advanced: TRUE, range: [0,1], default: 0.1] separating/rlt/maxparall = 0.1 # priority of separator <zerohalf> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: -6000] separating/zerohalf/priority = -6000 # frequency for calling separator <zerohalf> (-1: never, 0: only in root node) # [type: int, advanced: FALSE, range: [-1,1073741822], default: 10] separating/zerohalf/freq = 10 # maximal relative distance from current node's dual bound to primal bound compared to best node's dual bound for applying separator <zerohalf> (0.0: only on current best node, 1.0: on all nodes) # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/zerohalf/maxbounddist = 1 # should separator be delayed, if other separators found cuts? # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] separating/zerohalf/delay = FALSE # base for exponential increase of frequency at which separator <zerohalf> is called (1: call at each multiple of frequency) # [type: int, advanced: TRUE, range: [1,100], default: 4] separating/zerohalf/expbackoff = 4 # maximal number of zerohalf separation rounds per node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 5] separating/zerohalf/maxrounds = 5 # maximal number of zerohalf separation rounds in the root node (-1: unlimited) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 20] separating/zerohalf/maxroundsroot = 20 # maximal number of zerohalf cuts separated per separation round # [type: int, advanced: FALSE, range: [0,2147483647], default: 20] separating/zerohalf/maxsepacuts = 20 # initial seed used for random tie-breaking in cut selection # [type: int, advanced: FALSE, range: [0,2147483647], default: 24301] separating/zerohalf/initseed = 24301 # maximal number of zerohalf cuts separated per separation round in the root node # [type: int, advanced: FALSE, range: [0,2147483647], default: 100] separating/zerohalf/maxsepacutsroot = 100 # maximal number of zerohalf cuts considered per separation round # [type: int, advanced: FALSE, range: [0,2147483647], default: 2000] separating/zerohalf/maxcutcands = 2000 # maximal slack of rows to be used in aggregation # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] separating/zerohalf/maxslack = 0 # maximal slack of rows to be used in aggregation in the root node # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0] separating/zerohalf/maxslackroot = 0 # threshold for score of cut relative to best score to be considered good, so that less strict filtering is applied # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/zerohalf/goodscore = 1 # threshold for score of cut relative to best score to be discarded # [type: real, advanced: TRUE, range: [0,1], default: 0.5] separating/zerohalf/badscore = 0.5 # weight of objective parallelism in cut score calculation # [type: real, advanced: TRUE, range: [0,1], default: 0] separating/zerohalf/objparalweight = 0 # weight of efficacy in cut score calculation # [type: real, advanced: TRUE, range: [0,1], default: 1] separating/zerohalf/efficacyweight = 1 # weight of directed cutoff distance in cut score calculation # [type: real, advanced: TRUE, range: [0,1], default: 0] separating/zerohalf/dircutoffdistweight = 0 # maximum parallelism for good cuts # [type: real, advanced: TRUE, range: [0,1], default: 0.1] separating/zerohalf/goodmaxparall = 0.1 # maximum parallelism for non-good cuts # [type: real, advanced: TRUE, range: [0,1], default: 0.1] separating/zerohalf/maxparall = 0.1 # minimal violation to generate zerohalfcut for # [type: real, advanced: TRUE, range: [0,1.79769313486232e+308], default: 0.1] separating/zerohalf/minviol = 0.1 # should generated cuts be removed from the LP if they are no longer tight? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] separating/zerohalf/dynamiccuts = TRUE # maximal density of row to be used in aggregation # [type: real, advanced: TRUE, range: [0,1], default: 0.05] separating/zerohalf/maxrowdensity = 0.05 # additional number of variables allowed in row on top of density # [type: int, advanced: TRUE, range: [0,2147483647], default: 100] separating/zerohalf/densityoffset = 100 # display activation status of display column <solfound> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/solfound/active = 1 # display activation status of display column <concsolfound> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/concsolfound/active = 1 # display activation status of display column <time> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/time/active = 1 # display activation status of display column <nnodes> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/nnodes/active = 1 # display activation status of display column <nodesleft> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/nodesleft/active = 1 # display activation status of display column <nobjleaves> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/nobjleaves/active = 1 # display activation status of display column <ninfeasleaves> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/ninfeasleaves/active = 1 # display activation status of display column <lpiterations> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/lpiterations/active = 1 # display activation status of display column <lpavgiterations> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/lpavgiterations/active = 1 # display activation status of display column <lpcond> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/lpcond/active = 1 # display activation status of display column <memused> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/memused/active = 1 # display activation status of display column <concmemused> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/concmemused/active = 1 # display activation status of display column <memtotal> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/memtotal/active = 1 # display activation status of display column <depth> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/depth/active = 1 # display activation status of display column <maxdepth> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/maxdepth/active = 1 # display activation status of display column <plungedepth> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/plungedepth/active = 1 # display activation status of display column <nfrac> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/nfrac/active = 1 # display activation status of display column <nexternbranchcands> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/nexternbranchcands/active = 1 # display activation status of display column <vars> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/vars/active = 1 # display activation status of display column <conss> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/conss/active = 1 # display activation status of display column <curconss> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/curconss/active = 1 # display activation status of display column <curcols> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/curcols/active = 1 # display activation status of display column <currows> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/currows/active = 1 # display activation status of display column <cuts> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/cuts/active = 1 # display activation status of display column <separounds> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/separounds/active = 1 # display activation status of display column <poolsize> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/poolsize/active = 1 # display activation status of display column <conflicts> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/conflicts/active = 1 # display activation status of display column <strongbranchs> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/strongbranchs/active = 1 # display activation status of display column <pseudoobj> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/pseudoobj/active = 1 # display activation status of display column <lpobj> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/lpobj/active = 1 # display activation status of display column <curdualbound> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/curdualbound/active = 1 # display activation status of display column <estimate> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/estimate/active = 1 # display activation status of display column <avgdualbound> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/avgdualbound/active = 1 # display activation status of display column <dualbound> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/dualbound/active = 1 # display activation status of display column <primalbound> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/primalbound/active = 1 # display activation status of display column <concdualbound> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/concdualbound/active = 1 # display activation status of display column <concprimalbound> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/concprimalbound/active = 1 # display activation status of display column <cutoffbound> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/cutoffbound/active = 1 # display activation status of display column <gap> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/gap/active = 1 # display activation status of display column <concgap> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/concgap/active = 1 # display activation status of display column <primalgap> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 0] display/primalgap/active = 0 # display activation status of display column <nsols> (0: off, 1: auto, 2:on) # [type: int, advanced: FALSE, range: [0,2], default: 1] display/nsols/active = 1 # is statistics table <status> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/status/active = TRUE # is statistics table <timing> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/timing/active = TRUE # is statistics table <origprob> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/origprob/active = TRUE # is statistics table <presolvedprob> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/presolvedprob/active = TRUE # is statistics table <presolver> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/presolver/active = TRUE # is statistics table <constraint> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/constraint/active = TRUE # is statistics table <constiming> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/constiming/active = TRUE # is statistics table <propagator> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/propagator/active = TRUE # is statistics table <conflict> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/conflict/active = TRUE # is statistics table <separator> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/separator/active = TRUE # is statistics table <cutsel> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/cutsel/active = TRUE # is statistics table <pricer> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/pricer/active = TRUE # is statistics table <branchrules> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/branchrules/active = TRUE # is statistics table <heuristics> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/heuristics/active = TRUE # is statistics table <compression> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/compression/active = TRUE # is statistics table <benders> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/benders/active = TRUE # is statistics table <exprhdlr> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/exprhdlr/active = TRUE # is statistics table <lp> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/lp/active = TRUE # is statistics table <nlp> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/nlp/active = TRUE # is statistics table <nlpi> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/nlpi/active = TRUE # is statistics table <relaxator> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/relaxator/active = TRUE # is statistics table <tree> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/tree/active = TRUE # is statistics table <root> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/root/active = TRUE # is statistics table <solution> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/solution/active = TRUE # is statistics table <concurrentsolver> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] table/concurrentsolver/active = TRUE # soft time limit which should be applied after first solution was found (-1.0: disabled) # [type: real, advanced: FALSE, range: [-1,1.79769313486232e+308], default: -1] limits/softtime = -1 # the preferred number concurrent solvers of type <scip> with respect to the number of threads # [type: real, advanced: FALSE, range: [0,1], default: 1] concurrent/scip/prefprio = 1 # the preferred number concurrent solvers of type <scip-default> with respect to the number of threads # [type: real, advanced: FALSE, range: [0,1], default: 0] concurrent/scip-default/prefprio = 0 # the preferred number concurrent solvers of type <scip-cpsolver> with respect to the number of threads # [type: real, advanced: FALSE, range: [0,1], default: 0] concurrent/scip-cpsolver/prefprio = 0 # the preferred number concurrent solvers of type <scip-easycip> with respect to the number of threads # [type: real, advanced: FALSE, range: [0,1], default: 0] concurrent/scip-easycip/prefprio = 0 # the preferred number concurrent solvers of type <scip-feas> with respect to the number of threads # [type: real, advanced: FALSE, range: [0,1], default: 0] concurrent/scip-feas/prefprio = 0 # the preferred number concurrent solvers of type <scip-hardlp> with respect to the number of threads # [type: real, advanced: FALSE, range: [0,1], default: 0] concurrent/scip-hardlp/prefprio = 0 # the preferred number concurrent solvers of type <scip-opti> with respect to the number of threads # [type: real, advanced: FALSE, range: [0,1], default: 0] concurrent/scip-opti/prefprio = 0 # the preferred number concurrent solvers of type <scip-counter> with respect to the number of threads # [type: real, advanced: FALSE, range: [0,1], default: 0] concurrent/scip-counter/prefprio = 0 # priority of Benders' decomposition <default> # [type: int, advanced: FALSE, range: [-536870912,536870911], default: 0] benders/default/priority = 0 # should Benders' cuts be generated for LP solutions? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] benders/default/cutlp = TRUE # should Benders' cuts be generated for pseudo solutions? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] benders/default/cutpseudo = TRUE # should Benders' cuts be generated for relaxation solutions? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] benders/default/cutrelax = TRUE # should Benders' cuts from LNS heuristics be transferred to the main SCIP instance? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] benders/default/transfercuts = FALSE # should Benders' decomposition be used in LNS heurisics? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] benders/default/lnscheck = TRUE # maximum depth at which the LNS check is performed (-1: no limit) # [type: int, advanced: TRUE, range: [-1,1073741822], default: -1] benders/default/lnsmaxdepth = -1 # the maximum number of Benders' decomposition calls in LNS heuristics (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 10] benders/default/lnsmaxcalls = 10 # the maximum number of root node Benders' decomposition calls in LNS heuristics (-1: no limit) # [type: int, advanced: TRUE, range: [-1,2147483647], default: 0] benders/default/lnsmaxcallsroot = 0 # should the transferred cuts be added as constraints? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] benders/default/cutsasconss = TRUE # fraction of subproblems that are solved in each iteration # [type: real, advanced: FALSE, range: [0,1], default: 1] benders/default/subprobfrac = 1 # should the auxiliary variable bound be updated by solving the subproblem? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] benders/default/updateauxvarbound = FALSE # if the subproblem objective is integer, then define the auxiliary variables as implicit integers? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] benders/default/auxvarsimplint = FALSE # should Benders' cuts be generated while checking solutions? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] benders/default/cutcheck = TRUE # the convex combination multiplier for the cut strengthening # [type: real, advanced: FALSE, range: [0,1], default: 0.5] benders/default/cutstrengthenmult = 0.5 # the maximum number of cut strengthening without improvement # [type: int, advanced: TRUE, range: [0,2147483647], default: 5] benders/default/noimprovelimit = 5 # the constant use to perturb the cut strengthening core point # [type: real, advanced: FALSE, range: [0,1], default: 1e-06] benders/default/corepointperturb = 1e-06 # should the core point cut strengthening be employed (only applied to fractional solutions or continuous subproblems)? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] benders/default/cutstrengthenenabled = FALSE # where should the strengthening interior point be sourced from ('l'p relaxation, 'f'irst solution, 'i'ncumbent solution, 'r'elative interior point, vector of 'o'nes, vector of 'z'eros) # [type: char, advanced: FALSE, range: {lfiroz}, default: r] benders/default/cutstrengthenintpoint = r # the number of threads to use when solving the subproblems # [type: int, advanced: TRUE, range: [1,2147483647], default: 1] benders/default/numthreads = 1 # should a feasibility phase be executed during the root node, i.e. adding slack variables to constraints to ensure feasibility # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] benders/default/execfeasphase = FALSE # the initial objective coefficient of the slack variables in the subproblem # [type: real, advanced: FALSE, range: [0,1e+20], default: 1000000] benders/default/slackvarcoef = 1000000 # the maximal objective coefficient of the slack variables in the subproblem # [type: real, advanced: FALSE, range: [0,1e+20], default: 1000000000] benders/default/maxslackvarcoef = 1000000000 # should the constraints of the subproblems be checked for convexity? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] benders/default/checkconsconvexity = TRUE # iteration limit for NLP solver # [type: int, advanced: FALSE, range: [0,2147483647], default: 10000] benders/default/nlpiterlimit = 10000 # priority of Benders' cut <feas> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 10000] benders/default/benderscut/feas/priority = 10000 # is this Benders' decomposition cut method used to generate cuts? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] benders/default/benderscut/feas/enabled = TRUE # priority of Benders' cut <feasalt> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 10001] benders/default/benderscut/feasalt/priority = 10001 # is this Benders' decomposition cut method used to generate cuts? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] benders/default/benderscut/feasalt/enabled = TRUE # priority of Benders' cut <integer> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 0] benders/default/benderscut/integer/priority = 0 # is this Benders' decomposition cut method used to generate cuts? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] benders/default/benderscut/integer/enabled = TRUE # the constant term of the integer Benders' cuts. # [type: real, advanced: FALSE, range: [-1e+20,1e+20], default: -10000] benders/default/benderscut/integer/cutsconstant = -10000 # should cuts be generated and added to the cutpool instead of global constraints directly added to the problem. # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] benders/default/benderscut/integer/addcuts = FALSE # priority of Benders' cut <nogood> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 500] benders/default/benderscut/nogood/priority = 500 # is this Benders' decomposition cut method used to generate cuts? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] benders/default/benderscut/nogood/enabled = TRUE # should cuts be generated and added to the cutpool instead of global constraints directly added to the problem. # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] benders/default/benderscut/nogood/addcuts = FALSE # priority of Benders' cut <optimality> # [type: int, advanced: TRUE, range: [-536870912,536870911], default: 5000] benders/default/benderscut/optimality/priority = 5000 # is this Benders' decomposition cut method used to generate cuts? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] benders/default/benderscut/optimality/enabled = TRUE # should cuts be generated and added to the cutpool instead of global constraints directly added to the problem. # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] benders/default/benderscut/optimality/addcuts = FALSE # should the mixed integer rounding procedure be applied to cuts # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] benders/default/benderscut/optimality/mir = TRUE # priority of cut selection rule <ensemble> # [type: int, advanced: FALSE, range: [-536870912,1073741823], default: 7000] cutselection/ensemble/priority = 7000 # weight of normed-efficacy in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 0.75] cutselection/ensemble/efficacyweight = 0.75 # weight of normed-directed cutoff distance in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 0] cutselection/ensemble/dircutoffdistweight = 0 # weight of objective parallelism in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 0.25] cutselection/ensemble/objparalweight = 0.25 # weight of integral support in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 0.45] cutselection/ensemble/intsupportweight = 0.45 # weight of normed-expected obj improvement in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 0.1] cutselection/ensemble/expimprovweight = 0.1 # minimum score s.t. a cut can be added # [type: real, advanced: FALSE, range: [-1e+98,1e+98], default: 0] cutselection/ensemble/minscore = 0 # weight of normed-pseudo-costs in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 0.75] cutselection/ensemble/pscostweight = 0.75 # weight of normed-num-locks in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 0.25] cutselection/ensemble/locksweight = 0.25 # weight of maximum sparsity reward in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 0.5] cutselection/ensemble/maxsparsitybonus = 0.5 # weight of good numerics bonus (ratio of coefficients) in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 0] cutselection/ensemble/goodnumericsbonus = 0 # max sparsity value for which a bonus is applied in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 0.2] cutselection/ensemble/endsparsitybonus = 0.2 # threshold for when two cuts are considered parallel to each other # [type: real, advanced: FALSE, range: [0,1], default: 0.95] cutselection/ensemble/maxparal = 0.95 # penalty for weaker of two parallel cuts if penalising parallel cuts # [type: real, advanced: TRUE, range: [0,1e+98], default: 0.25] cutselection/ensemble/paralpenalty = 0.25 # max allowed cut density if filtering dense cuts # [type: real, advanced: TRUE, range: [0,1], default: 0.425] cutselection/ensemble/maxcutdensity = 0.425 # max non-zeros per round applied cuts (root). multiple num LP cols. # [type: real, advanced: FALSE, range: [0,1e+98], default: 4.5] cutselection/ensemble/maxnonzerorootround = 4.5 # max non-zeros per round applied cuts (tree). multiple num LP cols. # [type: real, advanced: FALSE, range: [0,1e+98], default: 9.5] cutselection/ensemble/maxnonzerotreeround = 9.5 # should cuts be filtered so no two parallel cuts are added # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] cutselection/ensemble/filterparalcuts = FALSE # should two parallel cuts be penalised instead of outright filtered # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] cutselection/ensemble/penaliseparalcuts = TRUE # should cuts over a given density threshold be filtered # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] cutselection/ensemble/filterdensecuts = TRUE # should the number of locks be penalised instead of rewarded # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] cutselection/ensemble/penaliselocks = TRUE # should objective parallelism be penalised instead of rewarded # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] cutselection/ensemble/penaliseobjparal = TRUE # max coefficient ratio for which numeric bonus is applied. # [type: int, advanced: TRUE, range: [1,1000000], default: 10000] cutselection/ensemble/maxcoefratiobonus = 10000 # max number of cuts such that cut selector is applied. # [type: int, advanced: TRUE, range: [1,1000000], default: 200] cutselection/ensemble/maxcuts = 200 # max number of variables such that cut selector is applied. # [type: int, advanced: TRUE, range: [1,1000000], default: 50000] cutselection/ensemble/maxnumvars = 50000 # priority of cut selection rule <hybrid> # [type: int, advanced: FALSE, range: [-536870912,1073741823], default: 8000] cutselection/hybrid/priority = 8000 # weight of efficacy in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 1] cutselection/hybrid/efficacyweight = 1 # weight of directed cutoff distance in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 0] cutselection/hybrid/dircutoffdistweight = 0 # weight of objective parallelism in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 0.1] cutselection/hybrid/objparalweight = 0.1 # weight of integral support in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 0.1] cutselection/hybrid/intsupportweight = 0.1 # minimal orthogonality for a cut to enter the LP # [type: real, advanced: FALSE, range: [0,1], default: 0.9] cutselection/hybrid/minortho = 0.9 # minimal orthogonality for a cut to enter the LP in the root node # [type: real, advanced: FALSE, range: [0,1], default: 0.9] cutselection/hybrid/minorthoroot = 0.9 # priority of cut selection rule <dynamic> # [type: int, advanced: FALSE, range: [-536870912,1073741823], default: 7000] cutselection/dynamic/priority = 7000 # weight of efficacy in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 1] cutselection/dynamic/efficacyweight = 1 # weight of directed cutoff distance in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 0] cutselection/dynamic/dircutoffdistweight = 0 # weight of objective parallelism in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 0] cutselection/dynamic/objparalweight = 0 # weight of integral support in cut score calculation # [type: real, advanced: FALSE, range: [0,1e+98], default: 0] cutselection/dynamic/intsupportweight = 0 # minimal efficacy gain for a cut to enter the LP # [type: real, advanced: FALSE, range: [0,1], default: 0.01] cutselection/dynamic/mingain = 0.01 # filtering strategy during cut selection # [type: char, advanced: FALSE, range: {df}, default: d] cutselection/dynamic/filtermode = d # minimal orthogonality for a cut to enter the LP # [type: real, advanced: FALSE, range: [0,1], default: 0.9] cutselection/dynamic/minortho = 0.9 # maximum depth at which this cutselector is employed # [type: int, advanced: FALSE, range: [-1,1073741822], default: -1] cutselection/dynamic/maxdepth = -1 # minimal distance from zero to enforce for child in bound tightening # [type: real, advanced: FALSE, range: [0,1], default: 1e-09] expr/log/minzerodistance = 1e-09 # minimal distance from zero to enforce for child in bound tightening # [type: real, advanced: FALSE, range: [0,1], default: 1e-09] expr/pow/minzerodistance = 1e-09 # maximal exponent when to expand power of sum in simplify # [type: int, advanced: FALSE, range: [1,2147483647], default: 2] expr/pow/expandmaxexponent = 2 # whether a fractional exponent is distributed onto factors on power of product # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] expr/pow/distribfracexponent = FALSE # whether to expand products of a sum and several factors in simplify # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] expr/prod/expandalways = FALSE # should this nonlinear handler be used # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/default/enabled = TRUE # should this nonlinear handler be used # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/convex/enabled = TRUE # whether to run convexity detection when the root of an expression is a non-quadratic sum # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] nlhdlr/convex/detectsum = FALSE # whether to create extended formulations instead of looking for maximal convex expressions # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/convex/extendedform = TRUE # maximal relative perturbation of non-differentiable reference point # [type: real, advanced: FALSE, range: [0,1], default: 0.01] nlhdlr/convex/maxperturb = 0.01 # whether to use convexity check on quadratics # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/convex/cvxquadratic = TRUE # whether to use convexity check on signomials # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/convex/cvxsignomial = TRUE # whether to use convexity check on product composition f(h)*h # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/convex/cvxprodcomp = TRUE # whether to also handle trivial convex expressions # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] nlhdlr/convex/handletrivial = FALSE # should this nonlinear handler be used # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/concave/enabled = TRUE # whether to run convexity detection when the root of an expression is a sum # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] nlhdlr/concave/detectsum = FALSE # whether to use convexity check on quadratics # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] nlhdlr/concave/cvxquadratic = FALSE # whether to use convexity check on signomials # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/concave/cvxsignomial = TRUE # whether to use convexity check on product composition f(h)*h # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/concave/cvxprodcomp = TRUE # whether to also handle trivial convex expressions # [type: bool, advanced: TRUE, range: {TRUE,FALSE}, default: FALSE] nlhdlr/concave/handletrivial = FALSE # should this nonlinear handler be used # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/bilinear/enabled = TRUE # whether to use the interval evaluation callback of the nlhdlr # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/bilinear/useinteval = TRUE # whether to use the reverse propagation callback of the nlhdlr # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/bilinear/usereverseprop = TRUE # maximum number of separation rounds in the root node # [type: int, advanced: FALSE, range: [0,2147483647], default: 10] nlhdlr/bilinear/maxseparoundsroot = 10 # maximum number of separation rounds in a local node # [type: int, advanced: FALSE, range: [0,2147483647], default: 1] nlhdlr/bilinear/maxseparounds = 1 # maximum depth to apply separation # [type: int, advanced: FALSE, range: [0,2147483647], default: 2147483647] nlhdlr/bilinear/maxsepadepth = 2147483647 # is statistics table <nlhdlr_bilinear> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] table/nlhdlr_bilinear/active = FALSE # should this nonlinear handler be used # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/perspective/enabled = TRUE # maximal number of propagation rounds in probing # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1] nlhdlr/perspective/maxproprounds = 1 # minimal relative reduction in a variable's domain for applying probing # [type: real, advanced: FALSE, range: [0,1], default: 0.1] nlhdlr/perspective/mindomreduction = 0.1 # minimal violation w.r.t. auxiliary variables for applying probing # [type: real, advanced: FALSE, range: [0,1.79769313486232e+308], default: 1e-05] nlhdlr/perspective/minviolprobing = 1e-05 # whether to do probing only in separation # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/perspective/probingonlyinsepa = TRUE # probing frequency (-1 - no probing, 0 - root node only) # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1] nlhdlr/perspective/probingfreq = 1 # whether perspective cuts are added only for convex expressions # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] nlhdlr/perspective/convexonly = FALSE # whether variable semicontinuity is used to tighten variable bounds # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/perspective/tightenbounds = TRUE # whether to adjust the reference point # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/perspective/adjrefpoint = TRUE # should this nonlinear handler be used # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/quadratic/enabled = TRUE # whether to use intersection cuts for quadratic constraints to separate # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] nlhdlr/quadratic/useintersectioncuts = FALSE # whether the strengthening should be used # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] nlhdlr/quadratic/usestrengthening = FALSE # whether monoidal strengthening should be used # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/quadratic/usemonoidal = TRUE # whether the minimal representation of the S-free set should be used (instead of the gauge) # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/quadratic/useminrep = TRUE # use bounds of variables in quadratic as rays for intersection cuts # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] nlhdlr/quadratic/useboundsasrays = FALSE # limit for number of cuts generated consecutively # [type: int, advanced: FALSE, range: [0,2147483647], default: 2] nlhdlr/quadratic/ncutslimit = 2 # limit for number of cuts generated at root node # [type: int, advanced: FALSE, range: [0,2147483647], default: 20] nlhdlr/quadratic/ncutslimitroot = 20 # maximal rank a slackvar can have # [type: int, advanced: FALSE, range: [0,2147483647], default: 2147483647] nlhdlr/quadratic/maxrank = 2147483647 # minimal cut violation the generated cuts must fulfill to be added to the LP # [type: real, advanced: FALSE, range: [0,1e+20], default: 0.0001] nlhdlr/quadratic/mincutviolation = 0.0001 # minimal violation the constraint must fulfill such that a cut is generated # [type: real, advanced: FALSE, range: [0,1e+20], default: 0.0001] nlhdlr/quadratic/minviolation = 0.0001 # determines at which nodes cut is used (if it's -1, it's used only at the root node, if it's n >= 0, it's used at every multiple of n # [type: int, advanced: FALSE, range: [-1,2147483647], default: 1] nlhdlr/quadratic/atwhichnodes = 1 # limit for number of rays we do the strengthening for # [type: int, advanced: FALSE, range: [0,2147483647], default: 2147483647] nlhdlr/quadratic/nstrengthlimit = 2147483647 # should we try to sparisfy the intersection cut? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] nlhdlr/quadratic/sparsifycuts = FALSE # should cut be generated even with bad numerics when restricting to ray? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/quadratic/ignorebadrayrestriction = TRUE # should cut be added even when range / efficacy is large? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/quadratic/ignorenhighre = TRUE # for monoidal strengthening, should we track more statistics (more expensive)? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] nlhdlr/quadratic/trackmore = FALSE # is statistics table <nlhdlr_quadratic> active # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: FALSE] table/nlhdlr_quadratic/active = FALSE # should this nonlinear handler be used # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/quotient/enabled = TRUE # should this nonlinear handler be used # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/signomial/enabled = TRUE # maximum number of variables when underestimating a concave power function # [type: int, advanced: TRUE, range: [2,14], default: 14] nlhdlr/signomial/maxnundervars = 14 # minimum scale factor when scaling a cut # [type: real, advanced: TRUE, range: [1e-06,1000000], default: 1e-05] nlhdlr/signomial/mincutscale = 1e-05 # should this nonlinear handler be used # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/soc/enabled = TRUE # Minimum efficacy which a cut needs in order to be added. # [type: real, advanced: FALSE, range: [0,1e+20], default: 1e-05] nlhdlr/soc/mincutefficacy = 1e-05 # Should Eigenvalue computations be done to detect complex cases in quadratic constraints? # [type: bool, advanced: FALSE, range: {TRUE,FALSE}, default: TRUE] nlhdlr/soc/compeigenvalues = TRUE # priority of NLPI <ipopt> # [type: int, advanced: FALSE, range: [-536870912,536870911], default: 1000] nlpi/ipopt/priority = 1000 # name of Ipopt options file # [type: string, advanced: FALSE, default: ""] nlpi/ipopt/optfile = "" # amount (relative and absolute) by which starting point is moved away from bounds in warmstarts # [type: real, advanced: FALSE, range: [0,1], default: 1e-09] nlpi/ipopt/warm_start_push = 1e-09 # Linear solver used for step computations. Valid values if not empty: ma27 ma57 ma77 ma86 ma97 pardiso mumps custom # [type: string, advanced: FALSE, default: ""] nlpi/ipopt/linear_solver = "" # Name of library containing HSL routines for load at runtime # [type: string, advanced: FALSE, default: ""] nlpi/ipopt/hsllib = "" # Name of library containing Pardiso routines (from pardiso-project.org) for load at runtime # [type: string, advanced: FALSE, default: ""] nlpi/ipopt/pardisolib = "" # Method for scaling the linear system. Valid values if not empty: none mc19 slack-based # [type: string, advanced: FALSE, default: ""] nlpi/ipopt/linear_system_scaling = "" # Select the technique used for scaling the NLP. Valid values if not empty: none user-scaling gradient-based equilibration-based # [type: string, advanced: FALSE, default: ""] nlpi/ipopt/nlp_scaling_method = "" # Update strategy for barrier parameter. Valid values if not empty: monotone adaptive # [type: string, advanced: FALSE, default: ""] nlpi/ipopt/mu_strategy = "" # Indicates what Hessian information is to be used. Valid values if not empty: exact limited-memory # [type: string, advanced: FALSE, default: ""] nlpi/ipopt/hessian_approximation = "" # Output verbosity level. -1 to use NLPI or Ipopt default. # [type: int, advanced: FALSE, range: [-1,12], default: -1] nlpi/ipopt/print_level = -1