 |
Google OR-Tools v9.14
a fast and portable software suite for combinatorial optimization
|
|
Google OR-Tools v9.14
a fast and portable software suite for combinatorial optimization
|
Contains the definitions for all the sat algorithm parameters and their default values. NEXT TAG: 325
Protobuf type operations_research.sat.SatParameters
Definition at line 15702 of file SatParameters.java.
Static Public Member Functions | |
| static final com.google.protobuf.Descriptors.Descriptor | getDescriptor () |
Protected Member Functions | |
| com.google.protobuf.GeneratedMessage.FieldAccessorTable | internalGetFieldAccessorTable () |
| Builder com.google.ortools.sat.SatParameters.Builder.addAllExtraSubsolvers | ( | java.lang.Iterable< java.lang.String > | values | ) |
A convenient way to add more workers types. These will be added at the beginning of the list.
repeated string extra_subsolvers = 219;
| values | The extraSubsolvers to add. |
Definition at line 26480 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addAllFilterSubsolvers | ( | java.lang.Iterable< java.lang.String > | values | ) |
repeated string filter_subsolvers = 293;
| values | The filterSubsolvers to add. |
Definition at line 26818 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addAllIgnoreSubsolvers | ( | java.lang.Iterable< java.lang.String > | values | ) |
Rather than fully specifying subsolvers, it is often convenient to just remove the ones that are not useful on a given problem or only keep specific ones for testing. Each string is interpreted as a "glob", so we support '*' and '?'. The way this work is that we will only accept a name that match a filter pattern (if non-empty) and do not match an ignore pattern. Note also that these fields work on LNS or LS names even if these are currently not specified via the subsolvers field.
repeated string ignore_subsolvers = 209;
| values | The ignoreSubsolvers to add. |
Definition at line 26684 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addAllRestartAlgorithms | ( | java.lang.Iterable<? extends com.google.ortools.sat.SatParameters.RestartAlgorithm > | values | ) |
The restart strategies will change each time the strategy_counter is increased. The current strategy will simply be the one at index strategy_counter modulo the number of strategy. Note that if this list includes a NO_RESTART, nothing will change when it is reached because the strategy_counter will only increment after a restart. The idea of switching of search strategy tailored for SAT/UNSAT comes from Chanseok Oh with his COMiniSatPS solver, see http://cs.nyu.edu/~chanseok/. But more generally, it seems REALLY beneficial to try different strategy.
repeated .operations_research.sat.SatParameters.RestartAlgorithm restart_algorithms = 61;
| values | The restartAlgorithms to add. |
Definition at line 21508 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addAllSubsolverParams | ( | java.lang.Iterable<? extends com.google.ortools.sat.SatParameters > | values | ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Definition at line 27070 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addAllSubsolvers | ( | java.lang.Iterable< java.lang.String > | values | ) |
In multi-thread, the solver can be mainly seen as a portfolio of solvers with different parameters. This field indicates the names of the parameters that are used in multithread. This only applies to "full" subsolvers. See cp_model_search.cc to see a list of the names and the default value (if left empty) that looks like: - default_lp (linearization_level:1) - fixed (only if fixed search specified or scheduling) - no_lp (linearization_level:0) - max_lp (linearization_level:2) - pseudo_costs (only if objective, change search heuristic) - reduced_costs (only if objective, change search heuristic) - quick_restart (kind of probing) - quick_restart_no_lp (kind of probing with linearization_level:0) - lb_tree_search (to improve lower bound, MIP like tree search) - probing (continuous probing and shaving) Also, note that some set of parameters will be ignored if they do not make sense. For instance if there is no objective, pseudo_cost or reduced_cost search will be ignored. Core based search will only work if the objective has many terms. If there is no fixed strategy fixed will be ignored. And so on. The order is important, as only the first num_full_subsolvers will be scheduled. You can see in the log which one are selected for a given run.
repeated string subsolvers = 207;
| values | The subsolvers to add. |
Definition at line 26279 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addExtraSubsolvers | ( | java.lang.String | value | ) |
A convenient way to add more workers types. These will be added at the beginning of the list.
repeated string extra_subsolvers = 219;
| value | The extraSubsolvers to add. |
Definition at line 26461 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addExtraSubsolversBytes | ( | com.google.protobuf.ByteString | value | ) |
A convenient way to add more workers types. These will be added at the beginning of the list.
repeated string extra_subsolvers = 219;
| value | The bytes of the extraSubsolvers to add. |
Definition at line 26515 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addFilterSubsolvers | ( | java.lang.String | value | ) |
repeated string filter_subsolvers = 293;
| value | The filterSubsolvers to add. |
Definition at line 26804 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addFilterSubsolversBytes | ( | com.google.protobuf.ByteString | value | ) |
repeated string filter_subsolvers = 293;
| value | The bytes of the filterSubsolvers to add. |
Definition at line 26843 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addIgnoreSubsolvers | ( | java.lang.String | value | ) |
Rather than fully specifying subsolvers, it is often convenient to just remove the ones that are not useful on a given problem or only keep specific ones for testing. Each string is interpreted as a "glob", so we support '*' and '?'. The way this work is that we will only accept a name that match a filter pattern (if non-empty) and do not match an ignore pattern. Note also that these fields work on LNS or LS names even if these are currently not specified via the subsolvers field.
repeated string ignore_subsolvers = 209;
| value | The ignoreSubsolvers to add. |
Definition at line 26658 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addIgnoreSubsolversBytes | ( | com.google.protobuf.ByteString | value | ) |
Rather than fully specifying subsolvers, it is often convenient to just remove the ones that are not useful on a given problem or only keep specific ones for testing. Each string is interpreted as a "glob", so we support '*' and '?'. The way this work is that we will only accept a name that match a filter pattern (if non-empty) and do not match an ignore pattern. Note also that these fields work on LNS or LS names even if these are currently not specified via the subsolvers field.
repeated string ignore_subsolvers = 209;
| value | The bytes of the ignoreSubsolvers to add. |
Definition at line 26733 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addRestartAlgorithms | ( | com.google.ortools.sat.SatParameters.RestartAlgorithm | value | ) |
The restart strategies will change each time the strategy_counter is increased. The current strategy will simply be the one at index strategy_counter modulo the number of strategy. Note that if this list includes a NO_RESTART, nothing will change when it is reached because the strategy_counter will only increment after a restart. The idea of switching of search strategy tailored for SAT/UNSAT comes from Chanseok Oh with his COMiniSatPS solver, see http://cs.nyu.edu/~chanseok/. But more generally, it seems REALLY beneficial to try different strategy.
repeated .operations_research.sat.SatParameters.RestartAlgorithm restart_algorithms = 61;
| value | The restartAlgorithms to add. |
Definition at line 21484 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addSubsolverParams | ( | com.google.ortools.sat.SatParameters | value | ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Definition at line 26977 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addSubsolverParams | ( | com.google.ortools.sat.SatParameters.Builder | builderForValue | ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Definition at line 27026 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addSubsolverParams | ( | int | index, |
| com.google.ortools.sat.SatParameters | value ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Definition at line 27001 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addSubsolverParams | ( | int | index, |
| com.google.ortools.sat.SatParameters.Builder | builderForValue ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Definition at line 27048 of file SatParameters.java.
| com.google.ortools.sat.SatParameters.Builder com.google.ortools.sat.SatParameters.Builder.addSubsolverParamsBuilder | ( | ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Definition at line 27187 of file SatParameters.java.
| com.google.ortools.sat.SatParameters.Builder com.google.ortools.sat.SatParameters.Builder.addSubsolverParamsBuilder | ( | int | index | ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Definition at line 27202 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addSubsolvers | ( | java.lang.String | value | ) |
In multi-thread, the solver can be mainly seen as a portfolio of solvers with different parameters. This field indicates the names of the parameters that are used in multithread. This only applies to "full" subsolvers. See cp_model_search.cc to see a list of the names and the default value (if left empty) that looks like: - default_lp (linearization_level:1) - fixed (only if fixed search specified or scheduling) - no_lp (linearization_level:0) - max_lp (linearization_level:2) - pseudo_costs (only if objective, change search heuristic) - reduced_costs (only if objective, change search heuristic) - quick_restart (kind of probing) - quick_restart_no_lp (kind of probing with linearization_level:0) - lb_tree_search (to improve lower bound, MIP like tree search) - probing (continuous probing and shaving) Also, note that some set of parameters will be ignored if they do not make sense. For instance if there is no objective, pseudo_cost or reduced_cost search will be ignored. Core based search will only work if the objective has many terms. If there is no fixed strategy fixed will be ignored. And so on. The order is important, as only the first num_full_subsolvers will be scheduled. You can see in the log which one are selected for a given run.
repeated string subsolvers = 207;
| value | The subsolvers to add. |
Definition at line 26237 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.addSubsolversBytes | ( | com.google.protobuf.ByteString | value | ) |
In multi-thread, the solver can be mainly seen as a portfolio of solvers with different parameters. This field indicates the names of the parameters that are used in multithread. This only applies to "full" subsolvers. See cp_model_search.cc to see a list of the names and the default value (if left empty) that looks like: - default_lp (linearization_level:1) - fixed (only if fixed search specified or scheduling) - no_lp (linearization_level:0) - max_lp (linearization_level:2) - pseudo_costs (only if objective, change search heuristic) - reduced_costs (only if objective, change search heuristic) - quick_restart (kind of probing) - quick_restart_no_lp (kind of probing with linearization_level:0) - lb_tree_search (to improve lower bound, MIP like tree search) - probing (continuous probing and shaving) Also, note that some set of parameters will be ignored if they do not make sense. For instance if there is no objective, pseudo_cost or reduced_cost search will be ignored. Core based search will only work if the objective has many terms. If there is no fixed strategy fixed will be ignored. And so on. The order is important, as only the first num_full_subsolvers will be scheduled. You can see in the log which one are selected for a given run.
repeated string subsolvers = 207;
| value | The bytes of the subsolvers to add. |
Definition at line 26360 of file SatParameters.java.
| com.google.ortools.sat.SatParameters com.google.ortools.sat.SatParameters.Builder.build | ( | ) |
Definition at line 16041 of file SatParameters.java.
| com.google.ortools.sat.SatParameters com.google.ortools.sat.SatParameters.Builder.buildPartial | ( | ) |
Definition at line 16050 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clear | ( | ) |
Definition at line 15730 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearAbsoluteGapLimit | ( | ) |
Stop the search when the gap between the best feasible objective (O) and our best objective bound (B) is smaller than a limit. The exact definition is: - Absolute: abs(O - B) - Relative: abs(O - B) / max(1, abs(O)). Important: The relative gap depends on the objective offset! If you artificially shift the objective, you will get widely different value of the relative gap. Note that if the gap is reached, the search status will be OPTIMAL. But one can check the best objective bound to see the actual gap. If the objective is integer, then any absolute gap < 1 will lead to a true optimal. If the objective is floating point, a gap of zero make little sense so is is why we use a non-zero default value. At the end of the search, we will display a warning if OPTIMAL is reported yet the gap is greater than this absolute gap.
optional double absolute_gap_limit = 159 [default = 0.0001];
Definition at line 22561 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearAddCgCuts | ( | ) |
Whether we generate and add Chvatal-Gomory cuts to the LP at root node. Note that for now, this is not heavily tuned.
optional bool add_cg_cuts = 117 [default = true];
Definition at line 35874 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearAddCliqueCuts | ( | ) |
Whether we generate clique cuts from the binary implication graph. Note that as the search goes on, this graph will contains new binary clauses learned by the SAT engine.
optional bool add_clique_cuts = 172 [default = true];
Definition at line 36058 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearAddLinMaxCuts | ( | ) |
For the lin max constraints, generates the cuts described in "Strong mixed-integer programming formulations for trained neural networks" by Ross Anderson et. (https://arxiv.org/pdf/1811.01988.pdf)
optional bool add_lin_max_cuts = 152 [default = true];
Definition at line 36246 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearAddLpConstraintsLazily | ( | ) |
If true, we start by an empty LP, and only add constraints not satisfied by the current LP solution batch by batch. A constraint that is only added like this is known as a "lazy" constraint in the literature, except that we currently consider all constraints as lazy here.
optional bool add_lp_constraints_lazily = 112 [default = true];
Definition at line 36398 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearAddMirCuts | ( | ) |
Whether we generate MIR cuts at root node. Note that for now, this is not heavily tuned.
optional bool add_mir_cuts = 120 [default = true];
Definition at line 35934 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearAddObjectiveCut | ( | ) |
When the LP objective is fractional, do we add the cut that forces the linear objective expression to be greater or equal to this fractional value rounded up? We can always do that since our objective is integer, and combined with MIR heuristic to reduce the coefficient of such cut, it can help.
optional bool add_objective_cut = 197 [default = false];
Definition at line 35814 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearAddRltCuts | ( | ) |
Whether we generate RLT cuts. This is still experimental but can help on binary problem with a lot of clauses of size 3.
optional bool add_rlt_cuts = 279 [default = true];
Definition at line 36118 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearAddZeroHalfCuts | ( | ) |
Whether we generate Zero-Half cuts at root node. Note that for now, this is not heavily tuned.
optional bool add_zero_half_cuts = 169 [default = true];
Definition at line 35994 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearAlsoBumpVariablesInConflictReasons | ( | ) |
When this is true, then the variables that appear in any of the reason of the variables in a conflict have their activity bumped. This is addition to the variables in the conflict, and the one that were used during conflict resolution.
optional bool also_bump_variables_in_conflict_reasons = 77 [default = false];
Definition at line 20432 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearAtMostOneMaxExpansionSize | ( | ) |
All at_most_one constraints with a size <= param will be replaced by a quadratic number of binary implications.
optional int32 at_most_one_max_expansion_size = 270 [default = 3];
Definition at line 34578 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearAutoDetectGreaterThanAtLeastOneOf | ( | ) |
If true, then the precedences propagator try to detect for each variable if it has a set of "optional incoming arc" for which at least one of them is present. This is usually useful to have but can be slow on model with a lot of precedence.
optional bool auto_detect_greater_than_at_least_one_of = 95 [default = true];
Definition at line 33374 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearBinaryMinimizationAlgorithm | ( | ) |
optional .operations_research.sat.SatParameters.BinaryMinizationAlgorithm binary_minimization_algorithm = 34 [default = BINARY_MINIMIZATION_FIRST];
Definition at line 20512 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearBinarySearchNumConflicts | ( | ) |
If non-negative, perform a binary search on the objective variable in order to find an [min, max] interval outside of which the solver proved unsat/sat under this amount of conflict. This can quickly reduce the objective domain on some problems.
optional int32 binary_search_num_conflicts = 99 [default = -1];
Definition at line 31486 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearBlockingRestartMultiplier | ( | ) |
optional double blocking_restart_multiplier = 66 [default = 1.4];
Definition at line 21973 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearBlockingRestartWindowSize | ( | ) |
optional int32 blocking_restart_window_size = 65 [default = 5000];
Definition at line 21933 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearBooleanEncodingLevel | ( | ) |
A non-negative level indicating how much we should try to fully encode Integer variables as Boolean.
optional int32 boolean_encoding_level = 107 [default = 1];
Definition at line 35478 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearCatchSigintSignal | ( | ) |
Indicates if the CP-SAT layer should catch Control-C (SIGINT) signals when calling solve. If set, catching the SIGINT signal will terminate the search gracefully, as if a time limit was reached.
optional bool catch_sigint_signal = 135 [default = true];
Definition at line 34642 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearClauseActivityDecay | ( | ) |
Clause activity parameters (same effect as the one on the variables).
optional double clause_activity_decay = 17 [default = 0.999];
Definition at line 21328 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearClauseCleanupLbdBound | ( | ) |
All the clauses with a LBD (literal blocks distance) lower or equal to this parameters will always be kept.
optional int32 clause_cleanup_lbd_bound = 59 [default = 5];
Definition at line 20860 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearClauseCleanupOrdering | ( | ) |
optional .operations_research.sat.SatParameters.ClauseOrdering clause_cleanup_ordering = 60 [default = CLAUSE_ACTIVITY];
Definition at line 20900 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearClauseCleanupPeriod | ( | ) |
Trigger a cleanup when this number of "deletable" clauses is learned.
optional int32 clause_cleanup_period = 11 [default = 10000];
Definition at line 20636 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearClauseCleanupProtection | ( | ) |
optional .operations_research.sat.SatParameters.ClauseProtection clause_cleanup_protection = 58 [default = PROTECTION_NONE];
Definition at line 20800 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearClauseCleanupRatio | ( | ) |
During a cleanup, if clause_cleanup_target is 0, we will delete the clause_cleanup_ratio of "deletable" clauses instead of aiming for a fixed target of clauses to keep.
optional double clause_cleanup_ratio = 190 [default = 0.5];
Definition at line 20760 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearClauseCleanupTarget | ( | ) |
During a cleanup, we will always keep that number of "deletable" clauses. Note that this doesn't include the "protected" clauses.
optional int32 clause_cleanup_target = 13 [default = 0];
Definition at line 20696 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearConvertIntervals | ( | ) |
Temporary flag util the feature is more mature. This convert intervals to the newer proto format that support affine start/var/end instead of just variables.
optional bool convert_intervals = 177 [default = true];
Definition at line 34950 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearCoreMinimizationLevel | ( | ) |
If positive, we spend some effort on each core: - At level 1, we use a simple heuristic to try to minimize an UNSAT core. - At level 2, we use propagation to minimize the core but also identify literal in at most one relationship in this core.
optional int32 core_minimization_level = 50 [default = 2];
Definition at line 28062 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearCountAssumptionLevelsInLbd | ( | ) |
Whether or not the assumption levels are taken into account during the LBD computation. According to the reference below, not counting them improves the solver in some situation. Note that this only impact solves under assumptions. Gilles Audemard, Jean-Marie Lagniez, Laurent Simon, "Improving Glucose for Incremental SAT Solving with Assumptions: Application to MUS Extraction" Theory and Applications of Satisfiability Testing - SAT 2013, Lecture Notes in Computer Science Volume 7962, 2013, pp 309-317.
optional bool count_assumption_levels_in_lbd = 49 [default = true];
Definition at line 23401 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearCoverOptimization | ( | ) |
If true, when the max-sat algo find a core, we compute the minimal number of literals in the core that needs to be true to have a feasible solution. This is also called core exhaustion in more recent max-SAT papers.
optional bool cover_optimization = 89 [default = true];
Definition at line 28186 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearCpModelPresolve | ( | ) |
Whether we presolve the cp_model before solving it.
optional bool cp_model_presolve = 86 [default = true];
Definition at line 24021 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearCpModelProbingLevel | ( | ) |
How much effort do we spend on probing. 0 disables it completely.
optional int32 cp_model_probing_level = 110 [default = 2];
Definition at line 24077 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearCpModelUseSatPresolve | ( | ) |
Whether we also use the sat presolve when cp_model_presolve is true.
optional bool cp_model_use_sat_presolve = 93 [default = true];
Definition at line 24133 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearCutActiveCountDecay | ( | ) |
optional double cut_active_count_decay = 156 [default = 0.8];
Definition at line 36762 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearCutCleanupTarget | ( | ) |
Target number of constraints to remove during cleanup.
optional int32 cut_cleanup_target = 157 [default = 1000];
Definition at line 36818 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearCutLevel | ( | ) |
Control the global cut effort. Zero will turn off all cut. For now we just have one level. Note also that most cuts are only used at linearization level >= 2.
optional int32 cut_level = 196 [default = 1];
Definition at line 35682 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearCutMaxActiveCountValue | ( | ) |
These parameters are similar to sat clause management activity parameters. They are effective only if the number of generated cuts exceed the storage limit. Default values are based on a few experiments on miplib instances.
optional double cut_max_active_count_value = 155 [default = 10000000000];
Definition at line 36722 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearDebugCrashIfPresolveBreaksHint | ( | ) |
Crash if presolve breaks a feasible hint.
optional bool debug_crash_if_presolve_breaks_hint = 306 [default = false];
Definition at line 27926 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearDebugCrashOnBadHint | ( | ) |
Crash if we do not manage to complete the hint into a full solution.
optional bool debug_crash_on_bad_hint = 195 [default = false];
Definition at line 27870 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearDebugMaxNumPresolveOperations | ( | ) |
If positive, try to stop just after that many presolve rules have been applied. This is mainly useful for debugging presolve.
optional int32 debug_max_num_presolve_operations = 151 [default = 0];
Definition at line 27814 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearDebugPostsolveWithFullSolver | ( | ) |
We have two different postsolve code. The default one should be better and it allows for a more powerful presolve, but it can be useful to postsolve using the full solver instead.
optional bool debug_postsolve_with_full_solver = 162 [default = false];
Definition at line 27754 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearDefaultRestartAlgorithms | ( | ) |
optional string default_restart_algorithms = 70 [default = "LUBY_RESTART,LBD_MOVING_AVERAGE_RESTART,DL_MOVING_AVERAGE_RESTART"];
Definition at line 21600 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearDetectLinearizedProduct | ( | ) |
Infer products of Boolean or of Boolean time IntegerVariable from the linear constrainst in the problem. This can be used in some cuts, altough for now we don't really exploit it.
optional bool detect_linearized_product = 277 [default = false];
Definition at line 37254 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearDetectTableWithCost | ( | ) |
If true, we detect variable that are unique to a table constraint and only there to encode a cost on each tuple. This is usually the case when a WCSP (weighted constraint program) is encoded into CP-SAT format. This can lead to a dramatic speed-up for such problems but is still experimental at this point.
optional bool detect_table_with_cost = 216 [default = false];
Definition at line 24277 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearDisableConstraintExpansion | ( | ) |
If true, it disable all constraint expansion. This should only be used to test the presolve of expanded constraints.
optional bool disable_constraint_expansion = 181 [default = false];
Definition at line 24825 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearDiversifyLnsParams | ( | ) |
If true, registers more lns subsolvers with different parameters.
optional bool diversify_lns_params = 137 [default = false];
Definition at line 34130 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearEncodeComplexLinearConstraintWithInteger | ( | ) |
Linear constraint with a complex right hand side (more than a single interval) need to be expanded, there is a couple of way to do that.
optional bool encode_complex_linear_constraint_with_integer = 223 [default = false];
Definition at line 24885 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearEncodeCumulativeAsReservoir | ( | ) |
Encore cumulative with fixed demands and capacity as a reservoir constraint. The only reason you might want to do that is to test the reservoir propagation code!
optional bool encode_cumulative_as_reservoir = 287 [default = false];
Definition at line 24693 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearEnumerateAllSolutions | ( | ) |
Whether we enumerate all solutions of a problem without objective. Note that setting this to true automatically disable some presolve reduction that can remove feasible solution. That is it has the same effect as setting keep_all_feasible_solutions_in_presolve. TODO(user): Do not do that and let the user choose what behavior is best by setting keep_all_feasible_solutions_in_presolve ?
optional bool enumerate_all_solutions = 87 [default = false];
Definition at line 32990 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearExpandAlldiffConstraints | ( | ) |
If true, expand all_different constraints that are not permutations. Permutations (#Variables = #Values) are always expanded.
optional bool expand_alldiff_constraints = 170 [default = false];
Definition at line 24405 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearExpandReservoirConstraints | ( | ) |
If true, expand the reservoir constraints by creating booleans for all possible precedences between event and encoding the constraint.
optional bool expand_reservoir_constraints = 182 [default = true];
Definition at line 24521 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearExpandReservoirUsingCircuit | ( | ) |
Mainly useful for testing. If this and expand_reservoir_constraints is true, we use a different encoding of the reservoir constraint using circuit instead of precedences. Note that this is usually slower, but can exercise different part of the solver. Note that contrary to the precedence encoding, this easily support variable demands. WARNING: with this encoding, the constraint takes a slightly different meaning. There must exist a permutation of the events occurring at the same time such that the level is within the reservoir after each of these events (in this permuted order). So we cannot have +100 and -100 at the same time if the level must be between 0 and 10 (as authorized by the reservoir constraint).
optional bool expand_reservoir_using_circuit = 288 [default = false];
Definition at line 24629 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearExploitAllLpSolution | ( | ) |
If true and the Lp relaxation of the problem has a solution, try to exploit it. This is same as above except in this case the lp solution might not be an integer solution.
optional bool exploit_all_lp_solution = 116 [default = true];
Definition at line 37010 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearExploitAllPrecedences | ( | ) |
optional bool exploit_all_precedences = 220 [default = false];
Definition at line 29162 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearExploitBestSolution | ( | ) |
When branching on a variable, follow the last best solution value.
optional bool exploit_best_solution = 130 [default = false];
Definition at line 37066 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearExploitIntegerLpSolution | ( | ) |
If true and the Lp relaxation of the problem has an integer optimal solution, try to exploit it. Note that since the LP relaxation may not contain all the constraints, such a solution is not necessarily a solution of the full problem.
optional bool exploit_integer_lp_solution = 94 [default = true];
Definition at line 36946 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearExploitObjective | ( | ) |
When branching an a variable that directly affect the objective, branch on the value that lead to the best objective first.
optional bool exploit_objective = 131 [default = true];
Definition at line 37190 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearExploitRelaxationSolution | ( | ) |
When branching on a variable, follow the last best relaxation solution value. We use the relaxation with the tightest bound on the objective as the best relaxation solution.
optional bool exploit_relaxation_solution = 161 [default = false];
Definition at line 37130 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearExtraSubsolvers | ( | ) |
A convenient way to add more workers types. These will be added at the beginning of the list.
repeated string extra_subsolvers = 219;
Definition at line 26498 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearFeasibilityJumpBatchDtime | ( | ) |
How much dtime for each LS batch.
optional double feasibility_jump_batch_dtime = 292 [default = 0.1];
Definition at line 31918 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearFeasibilityJumpDecay | ( | ) |
On each restart, we randomly choose if we use decay (with this parameter) or no decay.
optional double feasibility_jump_decay = 242 [default = 0.95];
Definition at line 31746 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearFeasibilityJumpEnableRestarts | ( | ) |
When stagnating, feasibility jump will either restart from a default solution (with some possible randomization), or randomly pertubate the current solution. This parameter selects the first option.
optional bool feasibility_jump_enable_restarts = 250 [default = true];
Definition at line 32102 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearFeasibilityJumpLinearizationLevel | ( | ) |
How much do we linearize the problem in the local search code.
optional int32 feasibility_jump_linearization_level = 257 [default = 2];
Definition at line 31802 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearFeasibilityJumpMaxExpandedConstraintSize | ( | ) |
Maximum size of no_overlap or no_overlap_2d constraint for a quadratic expansion. This might look a lot, but by expanding such constraint, we get a linear time evaluation per single variable moves instead of a slow O(n log n) one.
optional int32 feasibility_jump_max_expanded_constraint_size = 264 [default = 500];
Definition at line 32170 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearFeasibilityJumpRestartFactor | ( | ) |
This is a factor that directly influence the work before each restart. Increasing it leads to longer restart.
optional int32 feasibility_jump_restart_factor = 258 [default = 1];
Definition at line 31862 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearFeasibilityJumpVarPerburbationRangeRatio | ( | ) |
Max distance between the default value and the pertubated value relative to the range of the domain of the variable.
optional double feasibility_jump_var_perburbation_range_ratio = 248 [default = 0.2];
Definition at line 32038 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearFeasibilityJumpVarRandomizationProbability | ( | ) |
Probability for a variable to have a non default value upon restarts or perturbations.
optional double feasibility_jump_var_randomization_probability = 247 [default = 0.05];
Definition at line 31978 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearFillAdditionalSolutionsInResponse | ( | ) |
If true, the final response addition_solutions field will be filled with all solutions from our solutions pool. Note that if both this field and enumerate_all_solutions is true, we will copy to the pool all of the solution found. So if solution_pool_size is big enough, you can get all solutions this way instead of using the solution callback. Note that this only affect the "final" solution, not the one passed to the solution callbacks.
optional bool fill_additional_solutions_in_response = 194 [default = false];
Definition at line 33238 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearFillTightenedDomainsInResponse | ( | ) |
If true, add information about the derived variable domains to the CpSolverResponse. It is an option because it makes the response slighly bigger and there is a bit more work involved during the postsolve to construct it, but it should still have a low overhead. See the tightened_variables field in CpSolverResponse for more details.
optional bool fill_tightened_domains_in_response = 132 [default = false];
Definition at line 33146 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearFilterSatPostsolveClauses | ( | ) |
Internal parameter. During BVE, if we eliminate a variable x, by default we will push all clauses containing x and all clauses containing not(x) to the postsolve. However, it is possible to write the postsolve code so that only one such set is needed. The idea is that, if we push the set containing a literal l, is to set l to false except if it is needed to satisfy one of the clause in the set. This is always beneficial, but for historical reason, not all our postsolve algorithm support this.
optional bool filter_sat_postsolve_clauses = 324 [default = false];
Definition at line 23545 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearFilterSubsolvers | ( | ) |
repeated string filter_subsolvers = 293;
Definition at line 26831 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearFindBigLinearOverlap | ( | ) |
Try to find large "rectangle" in the linear constraint matrix with identical lines. If such rectangle is big enough, we can introduce a new integer variable corresponding to the common expression and greatly reduce the number of non-zero.
optional bool find_big_linear_overlap = 234 [default = true];
Definition at line 25429 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearFindMultipleCores | ( | ) |
Whether we try to find more independent cores for a given set of assumptions in the core based max-SAT algorithms.
optional bool find_multiple_cores = 84 [default = true];
Definition at line 28122 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearFixVariablesToTheirHintedValue | ( | ) |
If true, variables appearing in the solution hints will be fixed to their hinted value.
optional bool fix_variables_to_their_hinted_value = 192 [default = false];
Definition at line 30586 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearFpRounding | ( | ) |
optional .operations_research.sat.SatParameters.FPRoundingMethod fp_rounding = 165 [default = PROPAGATION_ASSISTED];
Definition at line 34074 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearGlucoseDecayIncrement | ( | ) |
optional double glucose_decay_increment = 23 [default = 0.01];
Definition at line 21232 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearGlucoseDecayIncrementPeriod | ( | ) |
optional int32 glucose_decay_increment_period = 24 [default = 5000];
Definition at line 21272 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearGlucoseMaxDecay | ( | ) |
The activity starts at 0.8 and increment by 0.01 every 5000 conflicts until 0.95. This "hack" seems to work well and comes from: Glucose 2.3 in the SAT 2013 Competition - SAT Competition 2013 http://edacc4.informatik.uni-ulm.de/SC13/solver-description-download/136
optional double glucose_max_decay = 22 [default = 0.95];
Definition at line 21192 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearHintConflictLimit | ( | ) |
Conflict limit used in the phase that exploit the solution hint.
optional int32 hint_conflict_limit = 153 [default = 10];
Definition at line 30458 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearIgnoreNames | ( | ) |
If true, we don't keep names in our internal copy of the user given model.
optional bool ignore_names = 202 [default = true];
Definition at line 25277 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearIgnoreSubsolvers | ( | ) |
Rather than fully specifying subsolvers, it is often convenient to just remove the ones that are not useful on a given problem or only keep specific ones for testing. Each string is interpreted as a "glob", so we support '*' and '?'. The way this work is that we will only accept a name that match a filter pattern (if non-empty) and do not match an ignore pattern. Note also that these fields work on LNS or LS names even if these are currently not specified via the subsolvers field.
repeated string ignore_subsolvers = 209;
Definition at line 26709 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearInferAllDiffs | ( | ) |
Run a max-clique code amongst all the x != y we can find and try to infer set of variables that are all different. This allows to close neos16.mps for instance. Note that we only run this code if there is no all_diff already in the model so that if a user want to add some all_diff, we assume it is well done and do not try to add more. This will also detect and add no_overlap constraints, if all the relations x != y have "offsets" between them. I.e. x > y + offset.
optional bool infer_all_diffs = 233 [default = true];
Definition at line 25361 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearInitialPolarity | ( | ) |
optional .operations_research.sat.SatParameters.Polarity initial_polarity = 2 [default = POLARITY_FALSE];
Definition at line 19868 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearInitialVariablesActivity | ( | ) |
The initial value of the variables activity. A non-zero value only make sense when use_erwa_heuristic is true. Experiments with a value of 1e-2 together with the ERWA heuristic showed slighthly better result than simply using zero. The idea is that when the "learning rate" of a variable becomes lower than this value, then we prefer to branch on never explored before variables. This is not in the ERWA paper.
optional double initial_variables_activity = 76 [default = 0];
Definition at line 20364 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearInprocessingDtimeRatio | ( | ) |
Proportion of deterministic time we should spend on inprocessing. At each "restart", if the proportion is below this ratio, we will do some inprocessing, otherwise, we skip it for this restart.
optional double inprocessing_dtime_ratio = 273 [default = 0.2];
Definition at line 25553 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearInprocessingMinimizationDtime | ( | ) |
Parameters for an heuristic similar to the one described in "An effective learnt clause minimization approach for CDCL Sat Solvers", https://www.ijcai.org/proceedings/2017/0098.pdf This is the amount of dtime we should spend on this technique during each inprocessing phase. The minimization technique is the same as the one used to minimize core in max-sat. We also minimize problem clauses and not just the learned clause that we keep forever like in the paper.
optional double inprocessing_minimization_dtime = 275 [default = 1];
Definition at line 25701 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearInprocessingMinimizationUseAllOrderings | ( | ) |
optional bool inprocessing_minimization_use_all_orderings = 298 [default = false];
Definition at line 25781 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearInprocessingMinimizationUseConflictAnalysis | ( | ) |
optional bool inprocessing_minimization_use_conflict_analysis = 297 [default = true];
Definition at line 25741 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearInprocessingProbingDtime | ( | ) |
The amount of dtime we should spend on probing for each inprocessing round.
optional double inprocessing_probing_dtime = 274 [default = 1];
Definition at line 25609 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearInstantiateAllVariables | ( | ) |
If true, the solver will add a default integer branching strategy to the already defined search strategy. If not, some variable might still not be fixed at the end of the search. For now we assume these variable can just be set to their lower bound.
optional bool instantiate_all_variables = 106 [default = true];
Definition at line 33306 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearInterleaveBatchSize | ( | ) |
optional int32 interleave_batch_size = 134 [default = 0];
Definition at line 27346 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearInterleaveSearch | ( | ) |
Experimental. If this is true, then we interleave all our major search strategy and distribute the work amongst num_workers. The search is deterministic (independently of num_workers!), and we schedule and wait for interleave_batch_size task to be completed before synchronizing and scheduling the next batch of tasks.
optional bool interleave_search = 136 [default = false];
Definition at line 27306 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearKeepAllFeasibleSolutionsInPresolve | ( | ) |
If true, we disable the presolve reductions that remove feasible solutions from the search space. Such solution are usually dominated by a "better" solution that is kept, but depending on the situation, we might want to keep all solutions. A trivial example is when a variable is unused. If this is true, then the presolve will not fix it to an arbitrary value and it will stay in the search space.
optional bool keep_all_feasible_solutions_in_presolve = 173 [default = false];
Definition at line 33074 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearKeepSymmetryInPresolve | ( | ) |
Experimental. This will compute the symmetry of the problem once and for all. All presolve operations we do should keep the symmetry group intact or modify it properly. For now we have really little support for this. We will disable a bunch of presolve operations that could be supported.
optional bool keep_symmetry_in_presolve = 303 [default = false];
Definition at line 35166 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearLbRelaxNumWorkersThreshold | ( | ) |
Only use lb-relax if we have at least that many workers.
optional int32 lb_relax_num_workers_threshold = 296 [default = 16];
Definition at line 34034 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearLinearizationLevel | ( | ) |
A non-negative level indicating the type of constraints we consider in the LP relaxation. At level zero, no LP relaxation is used. At level 1, only the linear constraint and full encoding are added. At level 2, we also add all the Boolean constraints.
optional int32 linearization_level = 90 [default = 1];
Definition at line 35418 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearLinearSplitSize | ( | ) |
Linear constraints that are not pseudo-Boolean and that are longer than this size will be split into sqrt(size) intermediate sums in order to have faster propation in the CP engine.
optional int32 linear_split_size = 256 [default = 100];
Definition at line 35350 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearLnsInitialDeterministicLimit | ( | ) |
optional double lns_initial_deterministic_limit = 308 [default = 0.1];
Definition at line 33626 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearLnsInitialDifficulty | ( | ) |
Initial parameters for neighborhood generation.
optional double lns_initial_difficulty = 307 [default = 0.5];
Definition at line 33586 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearLogPrefix | ( | ) |
Add a prefix to all logs.
optional string log_prefix = 185 [default = ""];
Definition at line 23048 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearLogSearchProgress | ( | ) |
Whether the solver should log the search progress. This is the maing logging parameter and if this is false, none of the logging (callbacks, log_to_stdout, log_to_response, ...) will do anything.
optional bool log_search_progress = 41 [default = false];
Definition at line 22893 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearLogSubsolverStatistics | ( | ) |
Whether the solver should display per sub-solver search statistics. This is only useful is log_search_progress is set to true, and if the number of search workers is > 1. Note that in all case we display a bit of stats with one line per subsolver.
optional bool log_subsolver_statistics = 189 [default = false];
Definition at line 22961 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearLogToResponse | ( | ) |
Log to response proto.
optional bool log_to_response = 187 [default = false];
Definition at line 23177 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearLogToStdout | ( | ) |
Log to stdout.
optional bool log_to_stdout = 186 [default = true];
Definition at line 23121 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearLpDualTolerance | ( | ) |
optional double lp_dual_tolerance = 267 [default = 1e-07];
Definition at line 34886 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearLpPrimalTolerance | ( | ) |
The internal LP tolerances used by CP-SAT. These applies to the internal and scaled problem. If the domains of your variables are large it might be good to use lower tolerances. If your problem is binary with low coefficients, it might be good to use higher ones to speed-up the lp solves.
optional double lp_primal_tolerance = 266 [default = 1e-07];
Definition at line 34846 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxAllDiffCutSize | ( | ) |
Cut generator for all diffs can add too many cuts for large all_diff constraints. This parameter restricts the large all_diff constraints to have a cut generator.
optional int32 max_all_diff_cut_size = 148 [default = 64];
Definition at line 36182 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxAlldiffDomainSize | ( | ) |
Max domain size for all_different constraints to be expanded.
optional int32 max_alldiff_domain_size = 320 [default = 256];
Definition at line 24461 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxClauseActivityValue | ( | ) |
optional double max_clause_activity_value = 18 [default = 1e+20];
Definition at line 21368 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxConsecutiveInactiveCount | ( | ) |
If a constraint/cut in LP is not active for that many consecutive OPTIMAL solves, remove it from the LP. Note that it might be added again later if it become violated by the current LP solution.
optional int32 max_consecutive_inactive_count = 121 [default = 100];
Definition at line 36658 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxCutRoundsAtLevelZero | ( | ) |
Max number of time we perform cut generation and resolve the LP at level 0.
optional int32 max_cut_rounds_at_level_zero = 154 [default = 1];
Definition at line 36594 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxDeterministicTime | ( | ) |
Maximum time allowed in deterministic time to solve a problem. The deterministic time should be correlated with the real time used by the solver, the time unit being as close as possible to a second.
optional double max_deterministic_time = 67 [default = inf];
Definition at line 22221 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxDomainSizeWhenEncodingEqNeqConstraints | ( | ) |
When loading a*x + b*y ==/!= c when x and y are both fully encoded. The solver may decide to replace the linear equation by a set of clauses. This is triggered if the sizes of the domains of x and y are below the threshold.
optional int32 max_domain_size_when_encoding_eq_neq_constraints = 191 [default = 16];
Definition at line 35546 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaximumRegionsToSplitInDisconnectedNoOverlap2D | ( | ) |
Detects when the space where items of a no_overlap_2d constraint can placed is disjoint (ie., fixed boxes split the domain). When it is the case, we can introduce a boolean for each pair <item, component> encoding whether the item is in the component or not. Then we replace the original no_overlap_2d constraint by one no_overlap_2d constraint for each component, with the new booleans as the enforcement_literal of the intervals. This is equivalent to expanding the original no_overlap_2d constraint into a bin packing problem with each connected component being a bin. This heuristic is only done when the number of regions to split is less than this parameter and <= 1 disables it.
optional int32 maximum_regions_to_split_in_disconnected_no_overlap_2d = 315 [default = 0];
Definition at line 29754 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxIntegerRoundingScaling | ( | ) |
In the integer rounding procedure used for MIR and Gomory cut, the maximum "scaling" we use (must be positive). The lower this is, the lower the integer coefficients of the cut will be. Note that cut generated by lower values are not necessarily worse than cut generated by larger value. There is no strict dominance relationship. Setting this to 2 result in the "strong fractional rouding" of Letchford and Lodi.
optional int32 max_integer_rounding_scaling = 119 [default = 600];
Definition at line 36330 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxLinMaxSizeForExpansion | ( | ) |
If the number of expressions in the lin_max is less that the max size parameter, model expansion replaces target = max(xi) by linear constraint with the introduction of new booleans bi such that bi => target == xi. This is mainly for experimenting compared to a custom lin_max propagator.
optional int32 max_lin_max_size_for_expansion = 280 [default = 0];
Definition at line 24765 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxMemoryInMb | ( | ) |
Maximum memory allowed for the whole thread containing the solver. The solver will abort as soon as it detects that this limit is crossed. As a result, this limit is approximative, but usually the solver will not go too much over. TODO(user): This is only used by the pure SAT solver, generalize to CP-SAT.
optional int64 max_memory_in_mb = 40 [default = 10000];
Definition at line 22437 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxNumberOfConflicts | ( | ) |
Maximum number of conflicts allowed to solve a problem. TODO(user): Maybe change the way the conflict limit is enforced? currently it is enforced on each independent internal SAT solve, rather than on the overall number of conflicts across all solves. So in the context of an optimization problem, this is not really usable directly by a client.
optional int64 max_number_of_conflicts = 37 [default = 9223372036854775807];
Definition at line 22361 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxNumCuts | ( | ) |
The limit on the number of cuts in our cut pool. When this is reached we do not generate cuts anymore. TODO(user): We should probably remove this parameters, and just always generate cuts but only keep the best n or something.
optional int32 max_num_cuts = 91 [default = 10000];
Definition at line 35618 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxNumDeterministicBatches | ( | ) |
Stops after that number of batches has been scheduled. This only make sense when interleave_search is true.
optional int32 max_num_deterministic_batches = 291 [default = 0];
Definition at line 22281 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxNumIntervalsForTimetableEdgeFinding | ( | ) |
Max number of intervals for the timetable_edge_finding algorithm to propagate. A value of 0 disables the constraint.
optional int32 max_num_intervals_for_timetable_edge_finding = 260 [default = 100];
Definition at line 29046 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxPairsPairwiseReasoningInNoOverlap2D | ( | ) |
If the number of pairs to look is below this threshold, do an extra step of propagation in the no_overlap_2d constraint by looking at all pairs of intervals.
optional int32 max_pairs_pairwise_reasoning_in_no_overlap_2d = 276 [default = 1250];
Definition at line 29662 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxPresolveIterations | ( | ) |
In case of large reduction in a presolve iteration, we perform multiple presolve iterations. This parameter controls the maximum number of such presolve iterations.
optional int32 max_presolve_iterations = 138 [default = 3];
Definition at line 23965 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxSatAssumptionOrder | ( | ) |
optional .operations_research.sat.SatParameters.MaxSatAssumptionOrder max_sat_assumption_order = 51 [default = DEFAULT_ASSUMPTION_ORDER];
Definition at line 28226 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxSatReverseAssumptionOrder | ( | ) |
If true, adds the assumption in the reverse order of the one defined by max_sat_assumption_order.
optional bool max_sat_reverse_assumption_order = 52 [default = false];
Definition at line 28286 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxSatStratification | ( | ) |
optional .operations_research.sat.SatParameters.MaxSatStratificationAlgorithm max_sat_stratification = 53 [default = STRATIFICATION_DESCENT];
Definition at line 28326 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxSizeToCreatePrecedenceLiteralsInDisjunctive | ( | ) |
Create one literal for each disjunction of two pairs of tasks. This slows down the solve time, but improves the lower bound of the objective in the makespan case. This will be triggered if the number of intervals is less or equal than the parameter and if use_strong_propagation_in_disjunctive is true.
optional int32 max_size_to_create_precedence_literals_in_disjunctive = 229 [default = 60];
Definition at line 28574 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxTimeInSeconds | ( | ) |
Maximum time allowed in seconds to solve a problem. The counter will starts at the beginning of the Solve() call.
optional double max_time_in_seconds = 36 [default = inf];
Definition at line 22157 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMaxVariableActivityValue | ( | ) |
optional double max_variable_activity_value = 16 [default = 1e+100];
Definition at line 21120 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMergeAtMostOneWorkLimit | ( | ) |
optional double merge_at_most_one_work_limit = 146 [default = 100000000];
Definition at line 24997 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMergeNoOverlapWorkLimit | ( | ) |
During presolve, we use a maximum clique heuristic to merge together no-overlap constraints or at most one constraints. This code can be slow, so we have a limit in place on the number of explored nodes in the underlying graph. The internal limit is an int64, but we use double here to simplify manual input.
optional double merge_no_overlap_work_limit = 145 [default = 1000000000000];
Definition at line 24957 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMinimizationAlgorithm | ( | ) |
optional .operations_research.sat.SatParameters.ConflictMinimizationAlgorithm minimization_algorithm = 4 [default = RECURSIVE];
Definition at line 20472 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMinimizeReductionDuringPbResolution | ( | ) |
A different algorithm during PB resolution. It minimizes the number of calls to ReduceCoefficients() which can be time consuming. However, the search space will be different and if the coefficients are large, this may lead to integer overflows that could otherwise be prevented.
optional bool minimize_reduction_during_pb_resolution = 48 [default = false];
Definition at line 23313 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMinimizeSharedClauses | ( | ) |
Minimize and detect subsumption of shared clauses immediately after they are imported.
optional bool minimize_shared_clauses = 300 [default = true];
Definition at line 27634 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMinOrthogonalityForLpConstraints | ( | ) |
While adding constraints, skip the constraints which have orthogonality less than 'min_orthogonality_for_lp_constraints' with already added constraints during current call. Orthogonality is defined as 1 - cosine(vector angle between constraints). A value of zero disable this feature.
optional double min_orthogonality_for_lp_constraints = 115 [default = 0.05];
Definition at line 36538 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMipAutomaticallyScaleVariables | ( | ) |
If true, some continuous variable might be automatically scaled. For now, this is only the case where we detect that a variable is actually an integer multiple of a constant. For instance, variables of the form k * 0.5 are quite frequent, and if we detect this, we will scale such variable domain by 2 to make it implied integer.
optional bool mip_automatically_scale_variables = 166 [default = true];
Definition at line 37518 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMipCheckPrecision | ( | ) |
As explained in mip_precision and mip_max_activity_exponent, we cannot always reach the wanted precision during scaling. We use this threshold to enphasize in the logs when the precision seems bad.
optional double mip_check_precision = 128 [default = 0.0001];
Definition at line 37886 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMipComputeTrueObjectiveBound | ( | ) |
Even if we make big error when scaling the objective, we can always derive a correct lower bound on the original objective by using the exact lower bound on the scaled integer version of the objective. This should be fast, but if you don't care about having a precise lower bound, you can turn it off.
optional bool mip_compute_true_objective_bound = 198 [default = true];
Definition at line 37958 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMipDropTolerance | ( | ) |
Any value in the input mip with a magnitude lower than this will be set to zero. This is to avoid some issue in LP presolving.
optional double mip_drop_tolerance = 232 [default = 1e-16];
Definition at line 38166 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMipMaxActivityExponent | ( | ) |
To avoid integer overflow, we always force the maximum possible constraint activity (and objective value) according to the initial variable domain to be smaller than 2 to this given power. Because of this, we cannot always reach the "mip_wanted_precision" parameter above. This can go as high as 62, but some internal algo currently abort early if they might run into integer overflow, so it is better to keep it a bit lower than this.
optional int32 mip_max_activity_exponent = 127 [default = 53];
Definition at line 37822 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMipMaxBound | ( | ) |
We need to bound the maximum magnitude of the variables for CP-SAT, and that is the bound we use. If the MIP model expect larger variable value in the solution, then the converted model will likely not be relevant.
optional double mip_max_bound = 124 [default = 10000000];
Definition at line 37318 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMipMaxValidMagnitude | ( | ) |
Any finite values in the input MIP must be below this threshold, otherwise the model will be reported invalid. This is needed to avoid floating point overflow when evaluating bounds * coeff for instance. We are a bit more defensive, but in practice, users shouldn't use super large values in a MIP.
optional double mip_max_valid_magnitude = 199 [default = 1e+20];
Definition at line 38030 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMipPresolveLevel | ( | ) |
When solving a MIP, we do some basic floating point presolving before scaling the problem to integer to be handled by CP-SAT. This control how much of that presolve we do. It can help to better scale floating point model, but it is not always behaving nicely.
optional int32 mip_presolve_level = 261 [default = 2];
Definition at line 38234 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMipScaleLargeDomain | ( | ) |
If this is false, then mip_var_scaling is only applied to variables with "small" domain. If it is true, we scale all floating point variable independenlty of their domain.
optional bool mip_scale_large_domain = 225 [default = false];
Definition at line 37446 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMipTreatHighMagnitudeBoundsAsInfinity | ( | ) |
By default, any variable/constraint bound with a finite value and a magnitude greater than the mip_max_valid_magnitude will result with a invalid model. This flags change the behavior such that such bounds are silently transformed to +∞ or -∞. It is recommended to keep it at false, and create valid bounds.
optional bool mip_treat_high_magnitude_bounds_as_infinity = 278 [default = false];
Definition at line 38106 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMipVarScaling | ( | ) |
All continuous variable of the problem will be multiplied by this factor. By default, we don't do any variable scaling and rely on the MIP model to specify continuous variable domain with the wanted precision.
optional double mip_var_scaling = 125 [default = 1];
Definition at line 37382 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearMipWantedPrecision | ( | ) |
When scaling constraint with double coefficients to integer coefficients, we will multiply by a power of 2 and round the coefficients. We will choose the lowest power such that we have no potential overflow (see mip_max_activity_exponent) and the worst case constraint activity error does not exceed this threshold. Note that we also detect constraint with rational coefficients and scale them accordingly when it seems better instead of using a power of 2. We also relax all constraint bounds by this absolute value. For pure integer constraint, if this value if lower than one, this will not change anything. However it is needed when scaling MIP problems. If we manage to scale a constraint correctly, the maximum error we can make will be twice this value (once for the scaling error and once for the relaxed bounds). If we are not able to scale that well, we will display that fact but still scale as best as we can.
optional double mip_wanted_precision = 126 [default = 1e-06];
Definition at line 37738 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearName | ( | ) |
In some context, like in a portfolio of search, it makes sense to name a given parameters set for logging purpose.
optional string name = 171 [default = ""];
Definition at line 19770 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearNewConstraintsBatchSize | ( | ) |
Add that many lazy constraints (or cuts) at once in the LP. Note that at the beginning of the solve, we do add more than this.
optional int32 new_constraints_batch_size = 122 [default = 50];
Definition at line 36878 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearNewLinearPropagation | ( | ) |
The new linear propagation code treat all constraints at once and use an adaptation of Bellman-Ford-Tarjan to propagate constraint in a smarter order and potentially detect propagation cycle earlier.
optional bool new_linear_propagation = 224 [default = true];
Definition at line 35286 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearNoOverlap2DBooleanRelationsLimit | ( | ) |
If less than this number of boxes are present in a no-overlap 2d, we create 4 Booleans per pair of boxes: - Box 2 is after Box 1 on x. - Box 1 is after Box 2 on x. - Box 2 is after Box 1 on y. - Box 1 is after Box 2 on y. Note that at least one of them must be true, and at most one on x and one on y can be true. This can significantly help in closing small problem. The SAT reasoning can be a lot more powerful when we take decision on such positional relations.
optional int32 no_overlap_2d_boolean_relations_limit = 321 [default = 10];
Definition at line 29358 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearNumConflictsBeforeStrategyChanges | ( | ) |
After each restart, if the number of conflict since the last strategy change is greater that this, then we increment a "strategy_counter" that can be use to change the search strategy used by the following restarts.
optional int32 num_conflicts_before_strategy_changes = 68 [default = 0];
Definition at line 22037 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearNumFullSubsolvers | ( | ) |
We distinguish subsolvers that consume a full thread, and the ones that are always interleaved. If left at zero, we will fix this with a default formula that depends on num_workers. But if you start modifying what runs, you might want to fix that to a given value depending on the num_workers you use.
optional int32 num_full_subsolvers = 294 [default = 0];
Definition at line 26001 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearNumSearchWorkers | ( | ) |
optional int32 num_search_workers = 100 [default = 0];
Definition at line 25929 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearNumViolationLs | ( | ) |
This will create incomplete subsolvers (that are not LNS subsolvers) that use the feasibility jump code to find improving solution, treating the objective improvement as a hard constraint.
optional int32 num_violation_ls = 244 [default = 0];
Definition at line 32234 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearNumWorkers | ( | ) |
Specify the number of parallel workers (i.e. threads) to use during search. This should usually be lower than your number of available cpus + hyperthread in your machine. A value of 0 means the solver will try to use all cores on the machine. A number of 1 means no parallelism. Note that 'num_workers' is the preferred name, but if it is set to zero, we will still read the deprecated 'num_search_workers'. As of 2020-04-10, if you're using SAT via MPSolver (to solve integer programs) this field is overridden with a value of 8, if the field is not set *explicitly*. Thus, always set this field explicitly or via MPSolver::SetNumThreads().
optional int32 num_workers = 206 [default = 0];
Definition at line 25889 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearOnlyAddCutsAtLevelZero | ( | ) |
For the cut that can be generated at any level, this control if we only try to generate them at the root node.
optional bool only_add_cuts_at_level_zero = 92 [default = false];
Definition at line 35742 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearOnlySolveIp | ( | ) |
If one try to solve a MIP model with CP-SAT, because we assume all variable to be integer after scaling, we will not necessarily have the correct optimal. Note however that all feasible solutions are valid since we will just solve a more restricted version of the original problem. This parameters is here to prevent user to think the solution is optimal when it might not be. One will need to manually set this to false to solve a MIP model where the optimal might be different. Note that this is tested after some MIP presolve steps, so even if not all original variable are integer, we might end up with a pure IP after presolve and after implied integer detection.
optional bool only_solve_ip = 222 [default = false];
Definition at line 37618 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearOptimizeWithCore | ( | ) |
The default optimization method is a simple "linear scan", each time trying to find a better solution than the previous one. If this is true, then we use a core-based approach (like in max-SAT) when we try to increase the lower bound instead.
optional bool optimize_with_core = 83 [default = false];
Definition at line 31270 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearOptimizeWithLbTreeSearch | ( | ) |
Do a more conventional tree search (by opposition to SAT based one) where we keep all the explored node in a tree. This is meant to be used in a portfolio and focus on improving the objective lower bound. Keeping the whole tree allow us to report a better objective lower bound coming from the worst open node in the tree.
optional bool optimize_with_lb_tree_search = 188 [default = false];
Definition at line 31342 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearOptimizeWithMaxHs | ( | ) |
This has no effect if optimize_with_core is false. If true, use a different core-based algorithm similar to the max-HS algo for max-SAT. This is a hybrid MIP/CP approach and it uses a MIP solver in addition to the CP/SAT one. This is also related to the PhD work of tobyodavies@ "Automatic Logic-Based Benders Decomposition with MiniZinc" http://aaai.org/ocs/index.php/AAAI/AAAI17/paper/view/14489
optional bool optimize_with_max_hs = 85 [default = false];
Definition at line 31562 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPbCleanupIncrement | ( | ) |
Same as for the clauses, but for the learned pseudo-Boolean constraints.
optional int32 pb_cleanup_increment = 46 [default = 200];
Definition at line 20956 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPbCleanupRatio | ( | ) |
optional double pb_cleanup_ratio = 47 [default = 0.5];
Definition at line 20996 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPermutePresolveConstraintOrder | ( | ) |
optional bool permute_presolve_constraint_order = 179 [default = false];
Definition at line 22789 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPermuteVariableRandomly | ( | ) |
This is mainly here to test the solver variability. Note that in tests, if not explicitly set to false, all 3 options will be set to true so that clients do not rely on the solver returning a specific solution if they are many equivalent optimal solutions.
optional bool permute_variable_randomly = 178 [default = false];
Definition at line 22749 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPolarityExploitLsHints | ( | ) |
If true and we have first solution LS workers, tries in some phase to follow a LS solutions that violates has litle constraints as possible.
optional bool polarity_exploit_ls_hints = 309 [default = false];
Definition at line 20088 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPolarityRephaseIncrement | ( | ) |
If non-zero, then we change the polarity heuristic after that many number of conflicts in an arithmetically increasing fashion. So x the first time, 2 * x the second time, etc...
optional int32 polarity_rephase_increment = 168 [default = 1000];
Definition at line 20028 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPolishLpSolution | ( | ) |
Whether we try to do a few degenerate iteration at the end of an LP solve to minimize the fractionality of the integer variable in the basis. This helps on some problems, but not so much on others. It also cost of bit of time to do such polish step.
optional bool polish_lp_solution = 175 [default = false];
Definition at line 34774 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPreferredVariableOrder | ( | ) |
optional .operations_research.sat.SatParameters.VariableOrder preferred_variable_order = 1 [default = IN_ORDER];
Definition at line 19828 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPresolveBlockedClause | ( | ) |
Whether we use an heuristic to detect some basic case of blocked clause in the SAT presolve.
optional bool presolve_blocked_clause = 88 [default = true];
Definition at line 23777 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPresolveBvaThreshold | ( | ) |
Apply Bounded Variable Addition (BVA) if the number of clauses is reduced by stricly more than this threshold. The algorithm described in the paper uses 0, but quick experiments showed that 1 is a good value. It may not be worth it to add a new variable just to remove one clause.
optional int32 presolve_bva_threshold = 73 [default = 1];
Definition at line 23901 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPresolveBveClauseWeight | ( | ) |
During presolve, we apply BVE only if this weight times the number of clauses plus the number of clause literals is not increased.
optional int32 presolve_bve_clause_weight = 55 [default = 3];
Definition at line 23605 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPresolveBveThreshold | ( | ) |
During presolve, only try to perform the bounded variable elimination (BVE) of a variable x if the number of occurrences of x times the number of occurrences of not(x) is not greater than this parameter.
optional int32 presolve_bve_threshold = 54 [default = 500];
Definition at line 23465 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPresolveExtractIntegerEnforcement | ( | ) |
If true, we will extract from linear constraints, enforcement literals of the form "integer variable at bound => simplified constraint". This should always be beneficial except that we don't always handle them as efficiently as we could for now. This causes problem on manna81.mps (LP relaxation not as tight it seems) and on neos-3354841-apure.mps.gz (too many literals created this way).
optional bool presolve_extract_integer_enforcement = 174 [default = false];
Definition at line 25141 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPresolveInclusionWorkLimit | ( | ) |
A few presolve operations involve detecting constraints included in other constraint. Since there can be a quadratic number of such pairs, and processing them usually involve scanning them, the complexity of these operations can be big. This enforce a local deterministic limit on the number of entries scanned. Default is 1e8. A value of zero will disable these presolve rules completely.
optional int64 presolve_inclusion_work_limit = 201 [default = 100000000];
Definition at line 25221 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPresolveProbingDeterministicTimeLimit | ( | ) |
optional double presolve_probing_deterministic_time_limit = 57 [default = 30];
Definition at line 23717 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPresolveSubstitutionLevel | ( | ) |
How much substitution (also called free variable aggregation in MIP litterature) should we perform at presolve. This currently only concerns variable appearing only in linear constraints. For now the value 0 turns it off and any positive value performs substitution.
optional int32 presolve_substitution_level = 147 [default = 1];
Definition at line 25065 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPresolveUseBva | ( | ) |
Whether or not we use Bounded Variable Addition (BVA) in the presolve.
optional bool presolve_use_bva = 72 [default = true];
Definition at line 23833 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearProbingDeterministicTimeLimit | ( | ) |
The maximum "deterministic" time limit to spend in probing. A value of zero will disable the probing. TODO(user): Clean up. The first one is used in CP-SAT, the other in pure SAT presolve.
optional double probing_deterministic_time_limit = 226 [default = 1];
Definition at line 23677 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearProbingNumCombinationsLimit | ( | ) |
How many combinations of pairs or triplets of variables we want to scan.
optional int32 probing_num_combinations_limit = 272 [default = 20000];
Definition at line 30762 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPropagationLoopDetectionFactor | ( | ) |
Some search decisions might cause a really large number of propagations to happen when integer variables with large domains are only reduced by 1 at each step. If we propagate more than the number of variable times this parameters we try to take counter-measure. Setting this to 0.0 disable this feature. TODO(user): Setting this to something like 10 helps in most cases, but the code is currently buggy and can cause the solve to enter a bad state where no progress is made.
optional double propagation_loop_detection_factor = 221 [default = 10];
Definition at line 28414 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPseudoCostReliabilityThreshold | ( | ) |
The solver ignores the pseudo costs of variables with number of recordings less than this threshold.
optional int64 pseudo_cost_reliability_threshold = 123 [default = 100];
Definition at line 31202 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearPushAllTasksTowardStart | ( | ) |
Experimental code: specify if the objective pushes all tasks toward the start of the schedule.
optional bool push_all_tasks_toward_start = 262 [default = false];
Definition at line 34310 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRandomBranchesRatio | ( | ) |
A number between 0 and 1 that indicates the proportion of branching variables that are selected randomly instead of choosing the first variable from the given variable_ordering strategy.
optional double random_branches_ratio = 32 [default = 0];
Definition at line 20224 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRandomizeSearch | ( | ) |
Randomize fixed search.
optional bool randomize_search = 103 [default = false];
Definition at line 34186 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRandomPolarityRatio | ( | ) |
The proportion of polarity chosen at random. Note that this take precedence over the phase saving heuristic. This is different from initial_polarity:POLARITY_RANDOM because it will select a new random polarity each time the variable is branched upon instead of selecting one initially and then always taking this choice.
optional double random_polarity_ratio = 45 [default = 0];
Definition at line 20160 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRandomSeed | ( | ) |
At the beginning of each solve, the random number generator used in some part of the solver is reinitialized to this seed. If you change the random seed, the solver may make different choices during the solving process. For some problems, the running time may vary a lot depending on small change in the solving algorithm. Running the solver with different seeds enables to have more robust benchmarks when evaluating new features.
optional int32 random_seed = 31 [default = 1];
Definition at line 22681 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRelativeGapLimit | ( | ) |
optional double relative_gap_limit = 160 [default = 0];
Definition at line 22601 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRemoveFixedVariablesEarly | ( | ) |
If cp_model_presolve is true and there is a large proportion of fixed variable after the first model copy, remap all the model to a dense set of variable before the full presolve even starts. This should help for LNS on large models.
optional bool remove_fixed_variables_early = 310 [default = true];
Definition at line 24201 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRepairHint | ( | ) |
If true, the solver tries to repair the solution given in the hint. This search terminates after the 'hint_conflict_limit' is reached and the solver switches to regular search. If false, then we do a FIXED_SEARCH using the hint until the hint_conflict_limit is reached.
optional bool repair_hint = 167 [default = false];
Definition at line 30526 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRestartAlgorithms | ( | ) |
The restart strategies will change each time the strategy_counter is increased. The current strategy will simply be the one at index strategy_counter modulo the number of strategy. Note that if this list includes a NO_RESTART, nothing will change when it is reached because the strategy_counter will only increment after a restart. The idea of switching of search strategy tailored for SAT/UNSAT comes from Chanseok Oh with his COMiniSatPS solver, see http://cs.nyu.edu/~chanseok/. But more generally, it seems REALLY beneficial to try different strategy.
repeated .operations_research.sat.SatParameters.RestartAlgorithm restart_algorithms = 61;
Definition at line 21533 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRestartDlAverageRatio | ( | ) |
In the moving average restart algorithms, a restart is triggered if the window average times this ratio is greater that the global average.
optional double restart_dl_average_ratio = 63 [default = 1];
Definition at line 21789 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRestartLbdAverageRatio | ( | ) |
optional double restart_lbd_average_ratio = 71 [default = 1];
Definition at line 21829 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRestartPeriod | ( | ) |
Restart period for the FIXED_RESTART strategy. This is also the multiplier used by the LUBY_RESTART strategy.
optional int32 restart_period = 30 [default = 50];
Definition at line 21673 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRestartRunningWindowSize | ( | ) |
Size of the window for the moving average restarts.
optional int32 restart_running_window_size = 62 [default = 50];
Definition at line 21729 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRootLpIterations | ( | ) |
Even at the root node, we do not want to spend too much time on the LP if it is "difficult". So we solve it in "chunks" of that many iterations. The solve will be continued down in the tree or the next time we go back to the root node.
optional int32 root_lp_iterations = 227 [default = 2000];
Definition at line 36466 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRoutingCutDpEffort | ( | ) |
The amount of "effort" to spend in dynamic programming for computing routing cuts. This is in term of basic operations needed by the algorithm in the worst case, so a value like 1e8 should take less than a second to compute.
optional double routing_cut_dp_effort = 314 [default = 10000000];
Definition at line 30302 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRoutingCutMaxInfeasiblePathLength | ( | ) |
If the length of an infeasible path is less than this value, a cut will be added to exclude it.
optional int32 routing_cut_max_infeasible_path_length = 317 [default = 6];
Definition at line 30362 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRoutingCutSubsetSizeForBinaryRelationBound | ( | ) |
If the size of a subset of nodes of a RoutesConstraint is less than this value, use linear constraints of size 1 and 2 (such as capacity and time window constraints) enforced by the arc literals to compute cuts for this subset (unless the subset size is less than routing_cut_subset_size_for_tight_binary_relation_bound, in which case the corresponding algorithm is used instead). The algorithm for these cuts has a O(n^3) complexity, where n is the subset size. Hence the value of this parameter should not be too large (e.g. 10 or 20).
optional int32 routing_cut_subset_size_for_binary_relation_bound = 312 [default = 0];
Definition at line 30038 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRoutingCutSubsetSizeForExactBinaryRelationBound | ( | ) |
Similar to above, but with an even stronger algorithm in O(n!). We try to be defensive and abort early or not run that often. Still the value of that parameter shouldn't really be much more than 10.
optional int32 routing_cut_subset_size_for_exact_binary_relation_bound = 316 [default = 8];
Definition at line 30166 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRoutingCutSubsetSizeForShortestPathsBound | ( | ) |
Similar to routing_cut_subset_size_for_exact_binary_relation_bound but use a bound based on shortest path distances (which respect triangular inequality). This allows to derive bounds that are valid for any superset of a given subset. This is slow, so it shouldn't really be larger than 10.
optional int32 routing_cut_subset_size_for_shortest_paths_bound = 318 [default = 8];
Definition at line 30234 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearRoutingCutSubsetSizeForTightBinaryRelationBound | ( | ) |
Similar to above, but with a different algorithm producing better cuts, at the price of a higher O(2^n) complexity, where n is the subset size. Hence the value of this parameter should be small (e.g. less than 10).
optional int32 routing_cut_subset_size_for_tight_binary_relation_bound = 313 [default = 0];
Definition at line 30102 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearSaveLpBasisInLbTreeSearch | ( | ) |
Experimental. Save the current LP basis at each node of the search tree so that when we jump around, we can load it and reduce the number of LP iterations needed. It currently works okay if we do not change the lp with cuts or simplification... More work is needed to make it robust in all cases.
optional bool save_lp_basis_in_lb_tree_search = 284 [default = false];
Definition at line 31418 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearSearchBranching | ( | ) |
optional .operations_research.sat.SatParameters.SearchBranching search_branching = 82 [default = AUTOMATIC_SEARCH];
Definition at line 30402 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearSearchRandomVariablePoolSize | ( | ) |
Search randomization will collect the top 'search_random_variable_pool_size' valued variables, and pick one randomly. The value of the variable is specific to each strategy.
optional int64 search_random_variable_pool_size = 104 [default = 0];
Definition at line 34250 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearShareBinaryClauses | ( | ) |
Allows sharing of new learned binary clause between workers.
optional bool share_binary_clauses = 203 [default = true];
Definition at line 27514 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearSharedTreeBalanceTolerance | ( | ) |
How much deeper compared to the ideal max depth of the tree is considered "balanced" enough to still accept a split. Without such a tolerance, sometimes the tree can only be split by a single worker, and they may not generate a split for some time. In contrast, with a tolerance of 1, at least half of all workers should be able to split the tree as soon as a split becomes required. This only has an effect on SPLIT_STRATEGY_BALANCED_TREE and SPLIT_STRATEGY_DISCREPANCY.
optional int32 shared_tree_balance_tolerance = 305 [default = 1];
Definition at line 32910 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearSharedTreeMaxNodesPerWorker | ( | ) |
In order to limit total shared memory and communication overhead, limit the total number of nodes that may be generated in the shared tree. If the shared tree runs out of unassigned leaves, workers act as portfolio workers. Note: this limit includes interior nodes, not just leaves.
optional int32 shared_tree_max_nodes_per_worker = 238 [default = 10000];
Definition at line 32790 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearSharedTreeNumWorkers | ( | ) |
Enables shared tree search. If positive, start this many complete worker threads to explore a shared search tree. These workers communicate objective bounds and simple decision nogoods relating to the shared prefix of the tree, and will avoid exploring the same subtrees as one another. Specifying a negative number uses a heuristic to select an appropriate number of shared tree workeres based on the total number of workers.
optional int32 shared_tree_num_workers = 235 [default = 0];
Definition at line 32430 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearSharedTreeOpenLeavesPerWorker | ( | ) |
How many open leaf nodes should the shared tree maintain per worker.
optional double shared_tree_open_leaves_per_worker = 281 [default = 2];
Definition at line 32722 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearSharedTreeSplitStrategy | ( | ) |
optional .operations_research.sat.SatParameters.SharedTreeSplitStrategy shared_tree_split_strategy = 239 [default = SPLIT_STRATEGY_AUTO];
Definition at line 32830 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearSharedTreeWorkerEnablePhaseSharing | ( | ) |
If true, shared tree workers share their target phase when returning an assigned subtree for the next worker to use.
optional bool shared_tree_worker_enable_phase_sharing = 304 [default = true];
Definition at line 32666 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearSharedTreeWorkerEnableTrailSharing | ( | ) |
If true, workers share more of the information from their local trail. Specifically, literals implied by the shared tree decisions.
optional bool shared_tree_worker_enable_trail_sharing = 295 [default = true];
Definition at line 32606 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearSharedTreeWorkerMinRestartsPerSubtree | ( | ) |
Minimum restarts before a worker will replace a subtree that looks "bad" based on the average LBD of learned clauses.
optional int32 shared_tree_worker_min_restarts_per_subtree = 282 [default = 1];
Definition at line 32546 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearShareGlueClauses | ( | ) |
Allows sharing of short glue clauses between workers. Implicitly disabled if share_binary_clauses is false.
optional bool share_glue_clauses = 285 [default = false];
Definition at line 27574 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearShareGlueClausesDtime | ( | ) |
The amount of dtime between each export of shared glue clauses.
optional double share_glue_clauses_dtime = 322 [default = 1];
Definition at line 27690 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearShareLevelZeroBounds | ( | ) |
Allows sharing of the bounds of modified variables at level 0.
optional bool share_level_zero_bounds = 114 [default = true];
Definition at line 27458 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearShareObjectiveBounds | ( | ) |
Allows objective sharing between workers.
optional bool share_objective_bounds = 113 [default = true];
Definition at line 27402 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearShavingDeterministicTimeInProbingSearch | ( | ) |
Add a shaving phase (where the solver tries to prove that the lower or upper bound of a variable are infeasible) to the probing search. (<= 0 disables it).
optional double shaving_deterministic_time_in_probing_search = 204 [default = 0.001];
Definition at line 30826 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearShavingSearchDeterministicTime | ( | ) |
Specifies the amount of deterministic time spent of each try at shaving a bound in the shaving search.
optional double shaving_search_deterministic_time = 205 [default = 0.1];
Definition at line 30886 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearShavingSearchThreshold | ( | ) |
Specifies the threshold between two modes in the shaving procedure. If the range of the variable/objective is less than this threshold, then the shaving procedure will try to remove values one by one. Otherwise, it will try to remove one range at a time.
optional int64 shaving_search_threshold = 290 [default = 64];
Definition at line 30954 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearSolutionPoolSize | ( | ) |
Size of the top-n different solutions kept by the solver. This parameter must be > 0. Currently this only impact the "base" solution chosen for a LNS fragment.
optional int32 solution_pool_size = 193 [default = 3];
Definition at line 33802 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearStopAfterFirstSolution | ( | ) |
For an optimization problem, stop the solver as soon as we have a solution.
optional bool stop_after_first_solution = 98 [default = false];
Definition at line 33430 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearStopAfterPresolve | ( | ) |
Mainly used when improving the presolver. When true, stops the solver after the presolve is complete (or after loading and root level propagation).
optional bool stop_after_presolve = 149 [default = false];
Definition at line 33490 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearStopAfterRootPropagation | ( | ) |
optional bool stop_after_root_propagation = 252 [default = false];
Definition at line 33530 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearStrategyChangeIncreaseRatio | ( | ) |
The parameter num_conflicts_before_strategy_changes is increased by that much after each strategy change.
optional double strategy_change_increase_ratio = 69 [default = 0];
Definition at line 22097 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearSubsolverParams | ( | ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Definition at line 27093 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearSubsolvers | ( | ) |
In multi-thread, the solver can be mainly seen as a portfolio of solvers with different parameters. This field indicates the names of the parameters that are used in multithread. This only applies to "full" subsolvers. See cp_model_search.cc to see a list of the names and the default value (if left empty) that looks like: - default_lp (linearization_level:1) - fixed (only if fixed search specified or scheduling) - no_lp (linearization_level:0) - max_lp (linearization_level:2) - pseudo_costs (only if objective, change search heuristic) - reduced_costs (only if objective, change search heuristic) - quick_restart (kind of probing) - quick_restart_no_lp (kind of probing with linearization_level:0) - lb_tree_search (to improve lower bound, MIP like tree search) - probing (continuous probing and shaving) Also, note that some set of parameters will be ignored if they do not make sense. For instance if there is no objective, pseudo_cost or reduced_cost search will be ignored. Core based search will only work if the objective has many terms. If there is no fixed strategy fixed will be ignored. And so on. The order is important, as only the first num_full_subsolvers will be scheduled. You can see in the log which one are selected for a given run.
repeated string subsolvers = 207;
Definition at line 26320 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearSubsumptionDuringConflictAnalysis | ( | ) |
At a really low cost, during the 1-UIP conflict computation, it is easy to detect if some of the involved reasons are subsumed by the current conflict. When this is true, such clauses are detached and later removed from the problem.
optional bool subsumption_during_conflict_analysis = 56 [default = true];
Definition at line 20580 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearSymmetryDetectionDeterministicTimeLimit | ( | ) |
Deterministic time limit for symmetry detection.
optional double symmetry_detection_deterministic_time_limit = 302 [default = 1];
Definition at line 35222 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearSymmetryLevel | ( | ) |
Whether we try to automatically detect the symmetries in a model and exploit them. Currently, at level 1 we detect them in presolve and try to fix Booleans. At level 2, we also do some form of dynamic symmetry breaking during search. At level 3, we also detect symmetries for very large models, which can be slow. At level 4, we try to break as much symmetry as possible in presolve.
optional int32 symmetry_level = 183 [default = 2];
Definition at line 35026 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearTableCompressionLevel | ( | ) |
How much we try to "compress" a table constraint. Compressing more leads to less Booleans and faster propagation but can reduced the quality of the lp relaxation. Values goes from 0 to 3 where we always try to fully compress a table. At 2, we try to automatically decide if it is worth it.
optional int32 table_compression_level = 217 [default = 2];
Definition at line 24345 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseAbslRandom | ( | ) |
optional bool use_absl_random = 180 [default = false];
Definition at line 22829 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseAllDifferentForCircuit | ( | ) |
Turn on extra propagation for the circuit constraint. This can be quite slow.
optional bool use_all_different_for_circuit = 311 [default = false];
Definition at line 29954 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseAreaEnergeticReasoningInNoOverlap2D | ( | ) |
When this is true, the no_overlap_2d constraint is reinforced with an energetic reasoning that uses an area-based energy. This can be combined with the two other overlap heuristics above.
optional bool use_area_energetic_reasoning_in_no_overlap_2d = 271 [default = false];
Definition at line 29558 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseBlockingRestart | ( | ) |
Block a moving restart algorithm if the trail size of the current conflict is greater than the multiplier times the moving average of the trail size at the previous conflicts.
optional bool use_blocking_restart = 64 [default = false];
Definition at line 21893 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseCombinedNoOverlap | ( | ) |
This can be beneficial if there is a lot of no-overlap constraints but a relatively low number of different intervals in the problem. Like 1000 intervals, but 1M intervals in the no-overlap constraints covering them.
optional bool use_combined_no_overlap = 133 [default = false];
Definition at line 34518 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseConservativeScaleOverloadChecker | ( | ) |
Enable a heuristic to solve cumulative constraints using a modified energy constraint. We modify the usual energy definition by applying a super-additive function (also called "conservative scale" or "dual-feasible function") to the demand and the durations of the tasks. This heuristic is fast but for most problems it does not help much to find a solution.
optional bool use_conservative_scale_overload_checker = 286 [default = false];
Definition at line 28906 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseDisjunctiveConstraintInCumulative | ( | ) |
When this is true, the cumulative constraint is reinforced with propagators from the disjunctive constraint to improve the inference on a set of tasks that are disjunctive at the root of the problem. This additional level supplements the default level of reasoning. Propagators of the cumulative constraint will not be used at all if all the tasks are disjunctive at root node. This always result in better propagation, but it is usually slow, so depending on the problem, turning this off may lead to a faster solution.
optional bool use_disjunctive_constraint_in_cumulative = 80 [default = true];
Definition at line 29254 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseDualSchedulingHeuristics | ( | ) |
When set, it activates a few scheduling parameters to improve the lower bound of scheduling problems. This is only effective with multiple workers as it modifies the reduced_cost, lb_tree_search, and probing workers.
optional bool use_dual_scheduling_heuristics = 214 [default = true];
Definition at line 29894 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseDynamicPrecedenceInCumulative | ( | ) |
optional bool use_dynamic_precedence_in_cumulative = 268 [default = false];
Definition at line 28746 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseDynamicPrecedenceInDisjunctive | ( | ) |
Whether we try to branch on decision "interval A before interval B" rather than on intervals bounds. This usually works better, but slow down a bit the time to find the first solution. These parameters are still EXPERIMENTAL, the result should be correct, but it some corner cases, they can cause some failing CHECK in the solver.
optional bool use_dynamic_precedence_in_disjunctive = 263 [default = false];
Definition at line 28706 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseEnergeticReasoningInNoOverlap2D | ( | ) |
When this is true, the no_overlap_2d constraint is reinforced with energetic reasoning. This additional level supplements the default level of reasoning.
optional bool use_energetic_reasoning_in_no_overlap_2d = 213 [default = false];
Definition at line 29494 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseErwaHeuristic | ( | ) |
Whether we use the ERWA (Exponential Recency Weighted Average) heuristic as described in "Learning Rate Based Branching Heuristic for SAT solvers", J.H.Liang, V. Ganesh, P. Poupart, K.Czarnecki, SAT 2016.
optional bool use_erwa_heuristic = 75 [default = false];
Definition at line 20288 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseExactLpReason | ( | ) |
The solver usually exploit the LP relaxation of a model. If this option is true, then whatever is infered by the LP will be used like an heuristic to compute EXACT propagation on the IP. So with this option, there is no numerical imprecision issues.
optional bool use_exact_lp_reason = 109 [default = true];
Definition at line 34454 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseExtendedProbing | ( | ) |
Use extended probing (probe bool_or, at_most_one, exactly_one).
optional bool use_extended_probing = 269 [default = true];
Definition at line 30706 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseFeasibilityJump | ( | ) |
Parameters for an heuristic similar to the one described in the paper: "Feasibility Jump: an LP-free Lagrangian MIP heuristic", Bjørnar Luteberget, Giorgio Sartor, 2023, Mathematical Programming Computation.
optional bool use_feasibility_jump = 265 [default = true];
Definition at line 31626 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseFeasibilityPump | ( | ) |
Adds a feasibility pump subsolver along with lns subsolvers.
optional bool use_feasibility_pump = 164 [default = true];
Definition at line 33914 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseHardPrecedencesInCumulative | ( | ) |
If true, detect and create constraint for integer variable that are "after" a set of intervals in the same cumulative constraint. Experimental: by default we just use "direct" precedences. If exploit_all_precedences is true, we explore the full precedence graph. This assumes we have a DAG otherwise it fails.
optional bool use_hard_precedences_in_cumulative = 215 [default = false];
Definition at line 29122 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseImpliedBounds | ( | ) |
Stores and exploits "implied-bounds" in the solver. That is, relations of the form literal => (var >= bound). This is currently used to derive stronger cuts.
optional bool use_implied_bounds = 144 [default = true];
Definition at line 34706 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseLbRelaxLns | ( | ) |
Turns on neighborhood generator based on local branching LP. Based on Huang et al., "Local Branching Relaxation Heuristics for Integer Linear Programs", 2023.
optional bool use_lb_relax_lns = 255 [default = true];
Definition at line 33978 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseLinear3ForNoOverlap2DPrecedences | ( | ) |
When set, this activates a propagator for the no_overlap_2d constraint that
uses any eventual linear constraints of the model in the form
`{start interval 1} - {end interval 2} + c*w <= ub` to detect that two
intervals must overlap in one dimension for some values of `w`. This is
particularly useful for problems where the distance between two boxes is
part of the model.
optional bool use_linear3_for_no_overlap_2d_precedences = 323 [default = true];
Definition at line 29830 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseLns | ( | ) |
Testing parameters used to disable all lns workers.
optional bool use_lns = 283 [default = true];
Definition at line 33682 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseLnsOnly | ( | ) |
Experimental parameters to disable everything but lns.
optional bool use_lns_only = 101 [default = false];
Definition at line 33738 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseLsOnly | ( | ) |
Disable every other type of subsolver, setting this turns CP-SAT into a pure local-search solver.
optional bool use_ls_only = 240 [default = false];
Definition at line 31686 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseObjectiveLbSearch | ( | ) |
If true, search will search in ascending max objective value (when minimizing) starting from the lower bound of the objective.
optional bool use_objective_lb_search = 228 [default = false];
Definition at line 31014 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseObjectiveShavingSearch | ( | ) |
This search differs from the previous search as it will not use assumptions to bound the objective, and it will recreate a full model with the hardcoded objective value.
optional bool use_objective_shaving_search = 253 [default = false];
Definition at line 31078 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseOptimizationHints | ( | ) |
For an optimization problem, whether we follow some hints in order to find a better first solution. For a variable with hint, the solver will always try to follow the hint. It will revert to the variable_branching default otherwise.
optional bool use_optimization_hints = 35 [default = true];
Definition at line 27994 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseOptionalVariables | ( | ) |
If true, we automatically detect variables whose constraint are always enforced by the same literal and we mark them as optional. This allows to propagate them as if they were present in some situation. TODO(user): This is experimental and seems to lead to wrong optimal in some situation. It should however gives correct solutions. Fix.
optional bool use_optional_variables = 108 [default = false];
Definition at line 34386 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseOverloadCheckerInCumulative | ( | ) |
When this is true, the cumulative constraint is reinforced with overload checking, i.e., an additional level of reasoning based on energy. This additional level supplements the default level of reasoning as well as timetable edge finding. This always result in better propagation, but it is usually slow, so depending on the problem, turning this off may lead to a faster solution.
optional bool use_overload_checker_in_cumulative = 78 [default = false];
Definition at line 28826 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUsePbResolution | ( | ) |
Whether to use pseudo-Boolean resolution to analyze a conflict. Note that this option only make sense if your problem is modelized using pseudo-Boolean constraints. If you only have clauses, this shouldn't change anything (except slow the solver down).
optional bool use_pb_resolution = 43 [default = false];
Definition at line 23245 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUsePhaseSaving | ( | ) |
If this is true, then the polarity of a variable will be the last value it was assigned to, or its default polarity if it was never assigned since the call to ResetDecisionHeuristic(). Actually, we use a newer version where we follow the last value in the longest non-conflicting partial assignment in the current phase. This is called 'literal phase saving'. For details see 'A Lightweight Component Caching Scheme for Satisfiability Solvers' K. Pipatsrisawat and A.Darwiche, In 10th International Conference on Theory and Applications of Satisfiability Testing, 2007.
optional bool use_phase_saving = 44 [default = true];
Definition at line 19964 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUsePrecedencesInDisjunctiveConstraint | ( | ) |
When this is true, then a disjunctive constraint will try to use the precedence relations between time intervals to propagate their bounds further. For instance if task A and B are both before C and task A and B are in disjunction, then we can deduce that task C must start after duration(A) + duration(B) instead of simply max(duration(A), duration(B)), provided that the start time for all task was currently zero. This always result in better propagation, but it is usually slow, so depending on the problem, turning this off may lead to a faster solution.
optional bool use_precedences_in_disjunctive_constraint = 74 [default = true];
Definition at line 28502 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseProbingSearch | ( | ) |
If true, search will continuously probe Boolean variables, and integer variable bounds. This parameter is set to true in parallel on the probing worker.
optional bool use_probing_search = 176 [default = false];
Definition at line 30650 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseRinsLns | ( | ) |
Turns on relaxation induced neighborhood generator.
optional bool use_rins_lns = 129 [default = true];
Definition at line 33858 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseSatInprocessing | ( | ) |
Enable or disable "inprocessing" which is some SAT presolving done at each restart to the root level.
optional bool use_sat_inprocessing = 163 [default = true];
Definition at line 25489 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseSharedTreeSearch | ( | ) |
Set on shared subtree workers. Users should not set this directly.
optional bool use_shared_tree_search = 236 [default = false];
Definition at line 32486 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseStrongPropagationInDisjunctive | ( | ) |
Enable stronger and more expensive propagation on no_overlap constraint.
optional bool use_strong_propagation_in_disjunctive = 230 [default = false];
Definition at line 28630 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseSymmetryInLp | ( | ) |
When we have symmetry, it is possible to "fold" all variables from the same orbit into a single variable, while having the same power of LP relaxation. This can help significantly on symmetric problem. However there is currently a bit of overhead as the rest of the solver need to do some translation between the folded LP and the rest of the problem.
optional bool use_symmetry_in_lp = 301 [default = false];
Definition at line 35098 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseTimetableEdgeFindingInCumulative | ( | ) |
When this is true, the cumulative constraint is reinforced with timetable edge finding, i.e., an additional level of reasoning based on the conjunction of energy and mandatory parts. This additional level supplements the default level of reasoning as well as overload_checker. This always result in better propagation, but it is usually slow, so depending on the problem, turning this off may lead to a faster solution.
optional bool use_timetable_edge_finding_in_cumulative = 79 [default = false];
Definition at line 28986 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseTimetablingInNoOverlap2D | ( | ) |
When this is true, the no_overlap_2d constraint is reinforced with propagators from the cumulative constraints. It consists of ignoring the position of rectangles in one position and projecting the no_overlap_2d on the other dimension to create a cumulative constraint. This is done on both axis. This additional level supplements the default level of reasoning.
optional bool use_timetabling_in_no_overlap_2d = 200 [default = false];
Definition at line 29430 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearUseTryEdgeReasoningInNoOverlap2D | ( | ) |
optional bool use_try_edge_reasoning_in_no_overlap_2d = 299 [default = false];
Definition at line 29598 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearVariableActivityDecay | ( | ) |
Each time a conflict is found, the activities of some variables are increased by one. Then, the activity of all variables are multiplied by variable_activity_decay. To implement this efficiently, the activity of all the variables is not decayed at each conflict. Instead, the activity increment is multiplied by 1 / decay. When an activity reach max_variable_activity_value, all the activity are multiplied by 1 / max_variable_activity_value.
optional double variable_activity_decay = 15 [default = 0.8];
Definition at line 21080 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearVariablesShavingLevel | ( | ) |
This search takes all Boolean or integer variables, and maximize or minimize them in order to reduce their domain. -1 is automatic, otherwise value 0 disables it, and 1, 2, or 3 changes something.
optional int32 variables_shaving_level = 289 [default = -1];
Definition at line 31142 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearViolationLsCompoundMoveProbability | ( | ) |
Probability of using compound move search each restart. TODO(user): Add reference to paper when published.
optional double violation_ls_compound_move_probability = 259 [default = 0.5];
Definition at line 32350 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.clearViolationLsPerturbationPeriod | ( | ) |
How long violation_ls should wait before perturbating a solution.
optional int32 violation_ls_perturbation_period = 249 [default = 100];
Definition at line 32290 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getAbsoluteGapLimit | ( | ) |
Stop the search when the gap between the best feasible objective (O) and our best objective bound (B) is smaller than a limit. The exact definition is: - Absolute: abs(O - B) - Relative: abs(O - B) / max(1, abs(O)). Important: The relative gap depends on the objective offset! If you artificially shift the objective, you will get widely different value of the relative gap. Note that if the gap is reached, the search status will be OPTIMAL. But one can check the best objective bound to see the actual gap. If the objective is integer, then any absolute gap < 1 will lead to a true optimal. If the objective is floating point, a gap of zero make little sense so is is why we use a non-zero default value. At the end of the search, we will display a warning if OPTIMAL is reported yet the gap is greater than this absolute gap.
optional double absolute_gap_limit = 159 [default = 0.0001];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22500 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getAddCgCuts | ( | ) |
Whether we generate and add Chvatal-Gomory cuts to the LP at root node. Note that for now, this is not heavily tuned.
optional bool add_cg_cuts = 117 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35845 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getAddCliqueCuts | ( | ) |
Whether we generate clique cuts from the binary implication graph. Note that as the search goes on, this graph will contains new binary clauses learned by the SAT engine.
optional bool add_clique_cuts = 172 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36027 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getAddLinMaxCuts | ( | ) |
For the lin max constraints, generates the cuts described in "Strong mixed-integer programming formulations for trained neural networks" by Ross Anderson et. (https://arxiv.org/pdf/1811.01988.pdf)
optional bool add_lin_max_cuts = 152 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36215 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getAddLpConstraintsLazily | ( | ) |
If true, we start by an empty LP, and only add constraints not satisfied by the current LP solution batch by batch. A constraint that is only added like this is known as a "lazy" constraint in the literature, except that we currently consider all constraints as lazy here.
optional bool add_lp_constraints_lazily = 112 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36365 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getAddMirCuts | ( | ) |
Whether we generate MIR cuts at root node. Note that for now, this is not heavily tuned.
optional bool add_mir_cuts = 120 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35905 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getAddObjectiveCut | ( | ) |
When the LP objective is fractional, do we add the cut that forces the linear objective expression to be greater or equal to this fractional value rounded up? We can always do that since our objective is integer, and combined with MIR heuristic to reduce the coefficient of such cut, it can help.
optional bool add_objective_cut = 197 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35779 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getAddRltCuts | ( | ) |
Whether we generate RLT cuts. This is still experimental but can help on binary problem with a lot of clauses of size 3.
optional bool add_rlt_cuts = 279 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36089 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getAddZeroHalfCuts | ( | ) |
Whether we generate Zero-Half cuts at root node. Note that for now, this is not heavily tuned.
optional bool add_zero_half_cuts = 169 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35965 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getAlsoBumpVariablesInConflictReasons | ( | ) |
When this is true, then the variables that appear in any of the reason of the variables in a conflict have their activity bumped. This is addition to the variables in the conflict, and the one that were used during conflict resolution.
optional bool also_bump_variables_in_conflict_reasons = 77 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20399 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getAtMostOneMaxExpansionSize | ( | ) |
All at_most_one constraints with a size <= param will be replaced by a quadratic number of binary implications.
optional int32 at_most_one_max_expansion_size = 270 [default = 3];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34549 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getAutoDetectGreaterThanAtLeastOneOf | ( | ) |
If true, then the precedences propagator try to detect for each variable if it has a set of "optional incoming arc" for which at least one of them is present. This is usually useful to have but can be slow on model with a lot of precedence.
optional bool auto_detect_greater_than_at_least_one_of = 95 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33341 of file SatParameters.java.
| com.google.ortools.sat.SatParameters.BinaryMinizationAlgorithm com.google.ortools.sat.SatParameters.Builder.getBinaryMinimizationAlgorithm | ( | ) |
optional .operations_research.sat.SatParameters.BinaryMinizationAlgorithm binary_minimization_algorithm = 34 [default = BINARY_MINIMIZATION_FIRST];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20492 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getBinarySearchNumConflicts | ( | ) |
If non-negative, perform a binary search on the objective variable in order to find an [min, max] interval outside of which the solver proved unsat/sat under this amount of conflict. This can quickly reduce the objective domain on some problems.
optional int32 binary_search_num_conflicts = 99 [default = -1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31453 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getBlockingRestartMultiplier | ( | ) |
optional double blocking_restart_multiplier = 66 [default = 1.4];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21954 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getBlockingRestartWindowSize | ( | ) |
optional int32 blocking_restart_window_size = 65 [default = 5000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21914 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getBooleanEncodingLevel | ( | ) |
A non-negative level indicating how much we should try to fully encode Integer variables as Boolean.
optional int32 boolean_encoding_level = 107 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35449 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getCatchSigintSignal | ( | ) |
Indicates if the CP-SAT layer should catch Control-C (SIGINT) signals when calling solve. If set, catching the SIGINT signal will terminate the search gracefully, as if a time limit was reached.
optional bool catch_sigint_signal = 135 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34611 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getClauseActivityDecay | ( | ) |
Clause activity parameters (same effect as the one on the variables).
optional double clause_activity_decay = 17 [default = 0.999];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21301 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getClauseCleanupLbdBound | ( | ) |
All the clauses with a LBD (literal blocks distance) lower or equal to this parameters will always be kept.
optional int32 clause_cleanup_lbd_bound = 59 [default = 5];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20831 of file SatParameters.java.
| com.google.ortools.sat.SatParameters.ClauseOrdering com.google.ortools.sat.SatParameters.Builder.getClauseCleanupOrdering | ( | ) |
optional .operations_research.sat.SatParameters.ClauseOrdering clause_cleanup_ordering = 60 [default = CLAUSE_ACTIVITY];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20880 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getClauseCleanupPeriod | ( | ) |
Trigger a cleanup when this number of "deletable" clauses is learned.
optional int32 clause_cleanup_period = 11 [default = 10000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20609 of file SatParameters.java.
| com.google.ortools.sat.SatParameters.ClauseProtection com.google.ortools.sat.SatParameters.Builder.getClauseCleanupProtection | ( | ) |
optional .operations_research.sat.SatParameters.ClauseProtection clause_cleanup_protection = 58 [default = PROTECTION_NONE];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20780 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getClauseCleanupRatio | ( | ) |
During a cleanup, if clause_cleanup_target is 0, we will delete the clause_cleanup_ratio of "deletable" clauses instead of aiming for a fixed target of clauses to keep.
optional double clause_cleanup_ratio = 190 [default = 0.5];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20729 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getClauseCleanupTarget | ( | ) |
During a cleanup, we will always keep that number of "deletable" clauses. Note that this doesn't include the "protected" clauses.
optional int32 clause_cleanup_target = 13 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20667 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getConvertIntervals | ( | ) |
Temporary flag util the feature is more mature. This convert intervals to the newer proto format that support affine start/var/end instead of just variables.
optional bool convert_intervals = 177 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34919 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getCoreMinimizationLevel | ( | ) |
If positive, we spend some effort on each core: - At level 1, we use a simple heuristic to try to minimize an UNSAT core. - At level 2, we use propagation to minimize the core but also identify literal in at most one relationship in this core.
optional int32 core_minimization_level = 50 [default = 2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28029 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getCountAssumptionLevelsInLbd | ( | ) |
Whether or not the assumption levels are taken into account during the LBD computation. According to the reference below, not counting them improves the solver in some situation. Note that this only impact solves under assumptions. Gilles Audemard, Jean-Marie Lagniez, Laurent Simon, "Improving Glucose for Incremental SAT Solving with Assumptions: Application to MUS Extraction" Theory and Applications of Satisfiability Testing - SAT 2013, Lecture Notes in Computer Science Volume 7962, 2013, pp 309-317.
optional bool count_assumption_levels_in_lbd = 49 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23358 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getCoverOptimization | ( | ) |
If true, when the max-sat algo find a core, we compute the minimal number of literals in the core that needs to be true to have a feasible solution. This is also called core exhaustion in more recent max-SAT papers.
optional bool cover_optimization = 89 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28155 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getCpModelPresolve | ( | ) |
Whether we presolve the cp_model before solving it.
optional bool cp_model_presolve = 86 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23994 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getCpModelProbingLevel | ( | ) |
How much effort do we spend on probing. 0 disables it completely.
optional int32 cp_model_probing_level = 110 [default = 2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24050 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getCpModelUseSatPresolve | ( | ) |
Whether we also use the sat presolve when cp_model_presolve is true.
optional bool cp_model_use_sat_presolve = 93 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24106 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getCutActiveCountDecay | ( | ) |
optional double cut_active_count_decay = 156 [default = 0.8];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36743 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getCutCleanupTarget | ( | ) |
Target number of constraints to remove during cleanup.
optional int32 cut_cleanup_target = 157 [default = 1000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36791 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getCutLevel | ( | ) |
Control the global cut effort. Zero will turn off all cut. For now we just have one level. Note also that most cuts are only used at linearization level >= 2.
optional int32 cut_level = 196 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35651 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getCutMaxActiveCountValue | ( | ) |
These parameters are similar to sat clause management activity parameters. They are effective only if the number of generated cuts exceed the storage limit. Default values are based on a few experiments on miplib instances.
optional double cut_max_active_count_value = 155 [default = 10000000000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36691 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getDebugCrashIfPresolveBreaksHint | ( | ) |
Crash if presolve breaks a feasible hint.
optional bool debug_crash_if_presolve_breaks_hint = 306 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27899 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getDebugCrashOnBadHint | ( | ) |
Crash if we do not manage to complete the hint into a full solution.
optional bool debug_crash_on_bad_hint = 195 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27843 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getDebugMaxNumPresolveOperations | ( | ) |
If positive, try to stop just after that many presolve rules have been applied. This is mainly useful for debugging presolve.
optional int32 debug_max_num_presolve_operations = 151 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27785 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getDebugPostsolveWithFullSolver | ( | ) |
We have two different postsolve code. The default one should be better and it allows for a more powerful presolve, but it can be useful to postsolve using the full solver instead.
optional bool debug_postsolve_with_full_solver = 162 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27723 of file SatParameters.java.
| com.google.ortools.sat.SatParameters com.google.ortools.sat.SatParameters.Builder.getDefaultInstanceForType | ( | ) |
Definition at line 16036 of file SatParameters.java.
| java.lang.String com.google.ortools.sat.SatParameters.Builder.getDefaultRestartAlgorithms | ( | ) |
optional string default_restart_algorithms = 70 [default = "LUBY_RESTART,LBD_MOVING_AVERAGE_RESTART,DL_MOVING_AVERAGE_RESTART"];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21552 of file SatParameters.java.
| com.google.protobuf.ByteString com.google.ortools.sat.SatParameters.Builder.getDefaultRestartAlgorithmsBytes | ( | ) |
optional string default_restart_algorithms = 70 [default = "LUBY_RESTART,LBD_MOVING_AVERAGE_RESTART,DL_MOVING_AVERAGE_RESTART"];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21571 of file SatParameters.java.
|
static |
Definition at line 15707 of file SatParameters.java.
| com.google.protobuf.Descriptors.Descriptor com.google.ortools.sat.SatParameters.Builder.getDescriptorForType | ( | ) |
Definition at line 16031 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getDetectLinearizedProduct | ( | ) |
Infer products of Boolean or of Boolean time IntegerVariable from the linear constrainst in the problem. This can be used in some cuts, altough for now we don't really exploit it.
optional bool detect_linearized_product = 277 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37223 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getDetectTableWithCost | ( | ) |
If true, we detect variable that are unique to a table constraint and only there to encode a cost on each tuple. This is usually the case when a WCSP (weighted constraint program) is encoded into CP-SAT format. This can lead to a dramatic speed-up for such problems but is still experimental at this point.
optional bool detect_table_with_cost = 216 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24240 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getDisableConstraintExpansion | ( | ) |
If true, it disable all constraint expansion. This should only be used to test the presolve of expanded constraints.
optional bool disable_constraint_expansion = 181 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24796 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getDiversifyLnsParams | ( | ) |
If true, registers more lns subsolvers with different parameters.
optional bool diversify_lns_params = 137 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34103 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getEncodeComplexLinearConstraintWithInteger | ( | ) |
Linear constraint with a complex right hand side (more than a single interval) need to be expanded, there is a couple of way to do that.
optional bool encode_complex_linear_constraint_with_integer = 223 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24856 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getEncodeCumulativeAsReservoir | ( | ) |
Encore cumulative with fixed demands and capacity as a reservoir constraint. The only reason you might want to do that is to test the reservoir propagation code!
optional bool encode_cumulative_as_reservoir = 287 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24662 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getEnumerateAllSolutions | ( | ) |
Whether we enumerate all solutions of a problem without objective. Note that setting this to true automatically disable some presolve reduction that can remove feasible solution. That is it has the same effect as setting keep_all_feasible_solutions_in_presolve. TODO(user): Do not do that and let the user choose what behavior is best by setting keep_all_feasible_solutions_in_presolve ?
optional bool enumerate_all_solutions = 87 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32951 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getExpandAlldiffConstraints | ( | ) |
If true, expand all_different constraints that are not permutations. Permutations (#Variables = #Values) are always expanded.
optional bool expand_alldiff_constraints = 170 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24376 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getExpandReservoirConstraints | ( | ) |
If true, expand the reservoir constraints by creating booleans for all possible precedences between event and encoding the constraint.
optional bool expand_reservoir_constraints = 182 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24492 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getExpandReservoirUsingCircuit | ( | ) |
Mainly useful for testing. If this and expand_reservoir_constraints is true, we use a different encoding of the reservoir constraint using circuit instead of precedences. Note that this is usually slower, but can exercise different part of the solver. Note that contrary to the precedence encoding, this easily support variable demands. WARNING: with this encoding, the constraint takes a slightly different meaning. There must exist a permutation of the events occurring at the same time such that the level is within the reservoir after each of these events (in this permuted order). So we cannot have +100 and -100 at the same time if the level must be between 0 and 10 (as authorized by the reservoir constraint).
optional bool expand_reservoir_using_circuit = 288 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24576 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getExploitAllLpSolution | ( | ) |
If true and the Lp relaxation of the problem has a solution, try to exploit it. This is same as above except in this case the lp solution might not be an integer solution.
optional bool exploit_all_lp_solution = 116 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36979 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getExploitAllPrecedences | ( | ) |
optional bool exploit_all_precedences = 220 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29143 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getExploitBestSolution | ( | ) |
When branching on a variable, follow the last best solution value.
optional bool exploit_best_solution = 130 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37039 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getExploitIntegerLpSolution | ( | ) |
If true and the Lp relaxation of the problem has an integer optimal solution, try to exploit it. Note that since the LP relaxation may not contain all the constraints, such a solution is not necessarily a solution of the full problem.
optional bool exploit_integer_lp_solution = 94 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36913 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getExploitObjective | ( | ) |
When branching an a variable that directly affect the objective, branch on the value that lead to the best objective first.
optional bool exploit_objective = 131 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37161 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getExploitRelaxationSolution | ( | ) |
When branching on a variable, follow the last best relaxation solution value. We use the relaxation with the tightest bound on the objective as the best relaxation solution.
optional bool exploit_relaxation_solution = 161 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37099 of file SatParameters.java.
| java.lang.String com.google.ortools.sat.SatParameters.Builder.getExtraSubsolvers | ( | int | index | ) |
A convenient way to add more workers types. These will be added at the beginning of the list.
repeated string extra_subsolvers = 219;
| index | The index of the element to return. |
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26414 of file SatParameters.java.
| com.google.protobuf.ByteString com.google.ortools.sat.SatParameters.Builder.getExtraSubsolversBytes | ( | int | index | ) |
A convenient way to add more workers types. These will be added at the beginning of the list.
repeated string extra_subsolvers = 219;
| index | The index of the value to return. |
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26428 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getExtraSubsolversCount | ( | ) |
A convenient way to add more workers types. These will be added at the beginning of the list.
repeated string extra_subsolvers = 219;
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26401 of file SatParameters.java.
| com.google.protobuf.ProtocolStringList com.google.ortools.sat.SatParameters.Builder.getExtraSubsolversList | ( | ) |
A convenient way to add more workers types. These will be added at the beginning of the list.
repeated string extra_subsolvers = 219;
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26388 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getFeasibilityJumpBatchDtime | ( | ) |
How much dtime for each LS batch.
optional double feasibility_jump_batch_dtime = 292 [default = 0.1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31891 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getFeasibilityJumpDecay | ( | ) |
On each restart, we randomly choose if we use decay (with this parameter) or no decay.
optional double feasibility_jump_decay = 242 [default = 0.95];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31717 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getFeasibilityJumpEnableRestarts | ( | ) |
When stagnating, feasibility jump will either restart from a default solution (with some possible randomization), or randomly pertubate the current solution. This parameter selects the first option.
optional bool feasibility_jump_enable_restarts = 250 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32071 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getFeasibilityJumpLinearizationLevel | ( | ) |
How much do we linearize the problem in the local search code.
optional int32 feasibility_jump_linearization_level = 257 [default = 2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31775 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getFeasibilityJumpMaxExpandedConstraintSize | ( | ) |
Maximum size of no_overlap or no_overlap_2d constraint for a quadratic expansion. This might look a lot, but by expanding such constraint, we get a linear time evaluation per single variable moves instead of a slow O(n log n) one.
optional int32 feasibility_jump_max_expanded_constraint_size = 264 [default = 500];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32137 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getFeasibilityJumpRestartFactor | ( | ) |
This is a factor that directly influence the work before each restart. Increasing it leads to longer restart.
optional int32 feasibility_jump_restart_factor = 258 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31833 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getFeasibilityJumpVarPerburbationRangeRatio | ( | ) |
Max distance between the default value and the pertubated value relative to the range of the domain of the variable.
optional double feasibility_jump_var_perburbation_range_ratio = 248 [default = 0.2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32009 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getFeasibilityJumpVarRandomizationProbability | ( | ) |
Probability for a variable to have a non default value upon restarts or perturbations.
optional double feasibility_jump_var_randomization_probability = 247 [default = 0.05];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31949 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getFillAdditionalSolutionsInResponse | ( | ) |
If true, the final response addition_solutions field will be filled with all solutions from our solutions pool. Note that if both this field and enumerate_all_solutions is true, we will copy to the pool all of the solution found. So if solution_pool_size is big enough, you can get all solutions this way instead of using the solution callback. Note that this only affect the "final" solution, not the one passed to the solution callbacks.
optional bool fill_additional_solutions_in_response = 194 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33193 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getFillTightenedDomainsInResponse | ( | ) |
If true, add information about the derived variable domains to the CpSolverResponse. It is an option because it makes the response slighly bigger and there is a bit more work involved during the postsolve to construct it, but it should still have a low overhead. See the tightened_variables field in CpSolverResponse for more details.
optional bool fill_tightened_domains_in_response = 132 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33111 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getFilterSatPostsolveClauses | ( | ) |
Internal parameter. During BVE, if we eliminate a variable x, by default we will push all clauses containing x and all clauses containing not(x) to the postsolve. However, it is possible to write the postsolve code so that only one such set is needed. The idea is that, if we push the set containing a literal l, is to set l to false except if it is needed to satisfy one of the clause in the set. This is always beneficial, but for historical reason, not all our postsolve algorithm support this.
optional bool filter_sat_postsolve_clauses = 324 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23506 of file SatParameters.java.
| java.lang.String com.google.ortools.sat.SatParameters.Builder.getFilterSubsolvers | ( | int | index | ) |
repeated string filter_subsolvers = 293;
| index | The index of the element to return. |
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26772 of file SatParameters.java.
| com.google.protobuf.ByteString com.google.ortools.sat.SatParameters.Builder.getFilterSubsolversBytes | ( | int | index | ) |
repeated string filter_subsolvers = 293;
| index | The index of the value to return. |
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26781 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getFilterSubsolversCount | ( | ) |
repeated string filter_subsolvers = 293;
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26764 of file SatParameters.java.
| com.google.protobuf.ProtocolStringList com.google.ortools.sat.SatParameters.Builder.getFilterSubsolversList | ( | ) |
repeated string filter_subsolvers = 293;
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26756 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getFindBigLinearOverlap | ( | ) |
Try to find large "rectangle" in the linear constraint matrix with identical lines. If such rectangle is big enough, we can introduce a new integer variable corresponding to the common expression and greatly reduce the number of non-zero.
optional bool find_big_linear_overlap = 234 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25396 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getFindMultipleCores | ( | ) |
Whether we try to find more independent cores for a given set of assumptions in the core based max-SAT algorithms.
optional bool find_multiple_cores = 84 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28093 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getFixVariablesToTheirHintedValue | ( | ) |
If true, variables appearing in the solution hints will be fixed to their hinted value.
optional bool fix_variables_to_their_hinted_value = 192 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30557 of file SatParameters.java.
| com.google.ortools.sat.SatParameters.FPRoundingMethod com.google.ortools.sat.SatParameters.Builder.getFpRounding | ( | ) |
optional .operations_research.sat.SatParameters.FPRoundingMethod fp_rounding = 165 [default = PROPAGATION_ASSISTED];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34054 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getGlucoseDecayIncrement | ( | ) |
optional double glucose_decay_increment = 23 [default = 0.01];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21213 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getGlucoseDecayIncrementPeriod | ( | ) |
optional int32 glucose_decay_increment_period = 24 [default = 5000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21253 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getGlucoseMaxDecay | ( | ) |
The activity starts at 0.8 and increment by 0.01 every 5000 conflicts until 0.95. This "hack" seems to work well and comes from: Glucose 2.3 in the SAT 2013 Competition - SAT Competition 2013 http://edacc4.informatik.uni-ulm.de/SC13/solver-description-download/136
optional double glucose_max_decay = 22 [default = 0.95];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21157 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getHintConflictLimit | ( | ) |
Conflict limit used in the phase that exploit the solution hint.
optional int32 hint_conflict_limit = 153 [default = 10];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30431 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getIgnoreNames | ( | ) |
If true, we don't keep names in our internal copy of the user given model.
optional bool ignore_names = 202 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25250 of file SatParameters.java.
| java.lang.String com.google.ortools.sat.SatParameters.Builder.getIgnoreSubsolvers | ( | int | index | ) |
Rather than fully specifying subsolvers, it is often convenient to just remove the ones that are not useful on a given problem or only keep specific ones for testing. Each string is interpreted as a "glob", so we support '*' and '?'. The way this work is that we will only accept a name that match a filter pattern (if non-empty) and do not match an ignore pattern. Note also that these fields work on LNS or LS names even if these are currently not specified via the subsolvers field.
repeated string ignore_subsolvers = 209;
| index | The index of the element to return. |
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26590 of file SatParameters.java.
| com.google.protobuf.ByteString com.google.ortools.sat.SatParameters.Builder.getIgnoreSubsolversBytes | ( | int | index | ) |
Rather than fully specifying subsolvers, it is often convenient to just remove the ones that are not useful on a given problem or only keep specific ones for testing. Each string is interpreted as a "glob", so we support '*' and '?'. The way this work is that we will only accept a name that match a filter pattern (if non-empty) and do not match an ignore pattern. Note also that these fields work on LNS or LS names even if these are currently not specified via the subsolvers field.
repeated string ignore_subsolvers = 209;
| index | The index of the value to return. |
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26611 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getIgnoreSubsolversCount | ( | ) |
Rather than fully specifying subsolvers, it is often convenient to just remove the ones that are not useful on a given problem or only keep specific ones for testing. Each string is interpreted as a "glob", so we support '*' and '?'. The way this work is that we will only accept a name that match a filter pattern (if non-empty) and do not match an ignore pattern. Note also that these fields work on LNS or LS names even if these are currently not specified via the subsolvers field.
repeated string ignore_subsolvers = 209;
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26570 of file SatParameters.java.
| com.google.protobuf.ProtocolStringList com.google.ortools.sat.SatParameters.Builder.getIgnoreSubsolversList | ( | ) |
Rather than fully specifying subsolvers, it is often convenient to just remove the ones that are not useful on a given problem or only keep specific ones for testing. Each string is interpreted as a "glob", so we support '*' and '?'. The way this work is that we will only accept a name that match a filter pattern (if non-empty) and do not match an ignore pattern. Note also that these fields work on LNS or LS names even if these are currently not specified via the subsolvers field.
repeated string ignore_subsolvers = 209;
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26550 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getInferAllDiffs | ( | ) |
Run a max-clique code amongst all the x != y we can find and try to infer set of variables that are all different. This allows to close neos16.mps for instance. Note that we only run this code if there is no all_diff already in the model so that if a user want to add some all_diff, we assume it is well done and do not try to add more. This will also detect and add no_overlap constraints, if all the relations x != y have "offsets" between them. I.e. x > y + offset.
optional bool infer_all_diffs = 233 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25320 of file SatParameters.java.
| com.google.ortools.sat.SatParameters.Polarity com.google.ortools.sat.SatParameters.Builder.getInitialPolarity | ( | ) |
optional .operations_research.sat.SatParameters.Polarity initial_polarity = 2 [default = POLARITY_FALSE];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 19848 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getInitialVariablesActivity | ( | ) |
The initial value of the variables activity. A non-zero value only make sense when use_erwa_heuristic is true. Experiments with a value of 1e-2 together with the ERWA heuristic showed slighthly better result than simply using zero. The idea is that when the "learning rate" of a variable becomes lower than this value, then we prefer to branch on never explored before variables. This is not in the ERWA paper.
optional double initial_variables_activity = 76 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20327 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getInprocessingDtimeRatio | ( | ) |
Proportion of deterministic time we should spend on inprocessing. At each "restart", if the proportion is below this ratio, we will do some inprocessing, otherwise, we skip it for this restart.
optional double inprocessing_dtime_ratio = 273 [default = 0.2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25522 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getInprocessingMinimizationDtime | ( | ) |
Parameters for an heuristic similar to the one described in "An effective learnt clause minimization approach for CDCL Sat Solvers", https://www.ijcai.org/proceedings/2017/0098.pdf This is the amount of dtime we should spend on this technique during each inprocessing phase. The minimization technique is the same as the one used to minimize core in max-sat. We also minimize problem clauses and not just the learned clause that we keep forever like in the paper.
optional double inprocessing_minimization_dtime = 275 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25656 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getInprocessingMinimizationUseAllOrderings | ( | ) |
optional bool inprocessing_minimization_use_all_orderings = 298 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25762 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getInprocessingMinimizationUseConflictAnalysis | ( | ) |
optional bool inprocessing_minimization_use_conflict_analysis = 297 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25722 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getInprocessingProbingDtime | ( | ) |
The amount of dtime we should spend on probing for each inprocessing round.
optional double inprocessing_probing_dtime = 274 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25582 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getInstantiateAllVariables | ( | ) |
If true, the solver will add a default integer branching strategy to the already defined search strategy. If not, some variable might still not be fixed at the end of the search. For now we assume these variable can just be set to their lower bound.
optional bool instantiate_all_variables = 106 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33273 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getInterleaveBatchSize | ( | ) |
optional int32 interleave_batch_size = 134 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27327 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getInterleaveSearch | ( | ) |
Experimental. If this is true, then we interleave all our major search strategy and distribute the work amongst num_workers. The search is deterministic (independently of num_workers!), and we schedule and wait for interleave_batch_size task to be completed before synchronizing and scheduling the next batch of tasks.
optional bool interleave_search = 136 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27269 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getKeepAllFeasibleSolutionsInPresolve | ( | ) |
If true, we disable the presolve reductions that remove feasible solutions from the search space. Such solution are usually dominated by a "better" solution that is kept, but depending on the situation, we might want to keep all solutions. A trivial example is when a variable is unused. If this is true, then the presolve will not fix it to an arbitrary value and it will stay in the search space.
optional bool keep_all_feasible_solutions_in_presolve = 173 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33033 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getKeepSymmetryInPresolve | ( | ) |
Experimental. This will compute the symmetry of the problem once and for all. All presolve operations we do should keep the symmetry group intact or modify it properly. For now we have really little support for this. We will disable a bunch of presolve operations that could be supported.
optional bool keep_symmetry_in_presolve = 303 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35133 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getLbRelaxNumWorkersThreshold | ( | ) |
Only use lb-relax if we have at least that many workers.
optional int32 lb_relax_num_workers_threshold = 296 [default = 16];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34007 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getLinearizationLevel | ( | ) |
A non-negative level indicating the type of constraints we consider in the LP relaxation. At level zero, no LP relaxation is used. At level 1, only the linear constraint and full encoding are added. At level 2, we also add all the Boolean constraints.
optional int32 linearization_level = 90 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35385 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getLinearSplitSize | ( | ) |
Linear constraints that are not pseudo-Boolean and that are longer than this size will be split into sqrt(size) intermediate sums in order to have faster propation in the CP engine.
optional int32 linear_split_size = 256 [default = 100];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35319 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getLnsInitialDeterministicLimit | ( | ) |
optional double lns_initial_deterministic_limit = 308 [default = 0.1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33607 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getLnsInitialDifficulty | ( | ) |
Initial parameters for neighborhood generation.
optional double lns_initial_difficulty = 307 [default = 0.5];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33559 of file SatParameters.java.
| java.lang.String com.google.ortools.sat.SatParameters.Builder.getLogPrefix | ( | ) |
Add a prefix to all logs.
optional string log_prefix = 185 [default = ""];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22988 of file SatParameters.java.
| com.google.protobuf.ByteString com.google.ortools.sat.SatParameters.Builder.getLogPrefixBytes | ( | ) |
Add a prefix to all logs.
optional string log_prefix = 185 [default = ""];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23011 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getLogSearchProgress | ( | ) |
Whether the solver should log the search progress. This is the maing logging parameter and if this is false, none of the logging (callbacks, log_to_stdout, log_to_response, ...) will do anything.
optional bool log_search_progress = 41 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22862 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getLogSubsolverStatistics | ( | ) |
Whether the solver should display per sub-solver search statistics. This is only useful is log_search_progress is set to true, and if the number of search workers is > 1. Note that in all case we display a bit of stats with one line per subsolver.
optional bool log_subsolver_statistics = 189 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22928 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getLogToResponse | ( | ) |
Log to response proto.
optional bool log_to_response = 187 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23150 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getLogToStdout | ( | ) |
Log to stdout.
optional bool log_to_stdout = 186 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23094 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getLpDualTolerance | ( | ) |
optional double lp_dual_tolerance = 267 [default = 1e-07];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34867 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getLpPrimalTolerance | ( | ) |
The internal LP tolerances used by CP-SAT. These applies to the internal and scaled problem. If the domains of your variables are large it might be good to use lower tolerances. If your problem is binary with low coefficients, it might be good to use higher ones to speed-up the lp solves.
optional double lp_primal_tolerance = 266 [default = 1e-07];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34811 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getMaxAllDiffCutSize | ( | ) |
Cut generator for all diffs can add too many cuts for large all_diff constraints. This parameter restricts the large all_diff constraints to have a cut generator.
optional int32 max_all_diff_cut_size = 148 [default = 64];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36151 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getMaxAlldiffDomainSize | ( | ) |
Max domain size for all_different constraints to be expanded.
optional int32 max_alldiff_domain_size = 320 [default = 256];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24434 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getMaxClauseActivityValue | ( | ) |
optional double max_clause_activity_value = 18 [default = 1e+20];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21349 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getMaxConsecutiveInactiveCount | ( | ) |
If a constraint/cut in LP is not active for that many consecutive OPTIMAL solves, remove it from the LP. Note that it might be added again later if it become violated by the current LP solution.
optional int32 max_consecutive_inactive_count = 121 [default = 100];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36627 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getMaxCutRoundsAtLevelZero | ( | ) |
Max number of time we perform cut generation and resolve the LP at level 0.
optional int32 max_cut_rounds_at_level_zero = 154 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36567 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getMaxDeterministicTime | ( | ) |
Maximum time allowed in deterministic time to solve a problem. The deterministic time should be correlated with the real time used by the solver, the time unit being as close as possible to a second.
optional double max_deterministic_time = 67 [default = inf];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22190 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getMaxDomainSizeWhenEncodingEqNeqConstraints | ( | ) |
When loading a*x + b*y ==/!= c when x and y are both fully encoded. The solver may decide to replace the linear equation by a set of clauses. This is triggered if the sizes of the domains of x and y are below the threshold.
optional int32 max_domain_size_when_encoding_eq_neq_constraints = 191 [default = 16];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35513 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getMaximumRegionsToSplitInDisconnectedNoOverlap2D | ( | ) |
Detects when the space where items of a no_overlap_2d constraint can placed is disjoint (ie., fixed boxes split the domain). When it is the case, we can introduce a boolean for each pair <item, component> encoding whether the item is in the component or not. Then we replace the original no_overlap_2d constraint by one no_overlap_2d constraint for each component, with the new booleans as the enforcement_literal of the intervals. This is equivalent to expanding the original no_overlap_2d constraint into a bin packing problem with each connected component being a bin. This heuristic is only done when the number of regions to split is less than this parameter and <= 1 disables it.
optional int32 maximum_regions_to_split_in_disconnected_no_overlap_2d = 315 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29709 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getMaxIntegerRoundingScaling | ( | ) |
In the integer rounding procedure used for MIR and Gomory cut, the maximum "scaling" we use (must be positive). The lower this is, the lower the integer coefficients of the cut will be. Note that cut generated by lower values are not necessarily worse than cut generated by larger value. There is no strict dominance relationship. Setting this to 2 result in the "strong fractional rouding" of Letchford and Lodi.
optional int32 max_integer_rounding_scaling = 119 [default = 600];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36289 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getMaxLinMaxSizeForExpansion | ( | ) |
If the number of expressions in the lin_max is less that the max size parameter, model expansion replaces target = max(xi) by linear constraint with the introduction of new booleans bi such that bi => target == xi. This is mainly for experimenting compared to a custom lin_max propagator.
optional int32 max_lin_max_size_for_expansion = 280 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24730 of file SatParameters.java.
| long com.google.ortools.sat.SatParameters.Builder.getMaxMemoryInMb | ( | ) |
Maximum memory allowed for the whole thread containing the solver. The solver will abort as soon as it detects that this limit is crossed. As a result, this limit is approximative, but usually the solver will not go too much over. TODO(user): This is only used by the pure SAT solver, generalize to CP-SAT.
optional int64 max_memory_in_mb = 40 [default = 10000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22400 of file SatParameters.java.
| long com.google.ortools.sat.SatParameters.Builder.getMaxNumberOfConflicts | ( | ) |
Maximum number of conflicts allowed to solve a problem. TODO(user): Maybe change the way the conflict limit is enforced? currently it is enforced on each independent internal SAT solve, rather than on the overall number of conflicts across all solves. So in the context of an optimization problem, this is not really usable directly by a client.
optional int64 max_number_of_conflicts = 37 [default = 9223372036854775807];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22322 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getMaxNumCuts | ( | ) |
The limit on the number of cuts in our cut pool. When this is reached we do not generate cuts anymore. TODO(user): We should probably remove this parameters, and just always generate cuts but only keep the best n or something.
optional int32 max_num_cuts = 91 [default = 10000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35583 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getMaxNumDeterministicBatches | ( | ) |
Stops after that number of batches has been scheduled. This only make sense when interleave_search is true.
optional int32 max_num_deterministic_batches = 291 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22252 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getMaxNumIntervalsForTimetableEdgeFinding | ( | ) |
Max number of intervals for the timetable_edge_finding algorithm to propagate. A value of 0 disables the constraint.
optional int32 max_num_intervals_for_timetable_edge_finding = 260 [default = 100];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29017 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getMaxPairsPairwiseReasoningInNoOverlap2D | ( | ) |
If the number of pairs to look is below this threshold, do an extra step of propagation in the no_overlap_2d constraint by looking at all pairs of intervals.
optional int32 max_pairs_pairwise_reasoning_in_no_overlap_2d = 276 [default = 1250];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29631 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getMaxPresolveIterations | ( | ) |
In case of large reduction in a presolve iteration, we perform multiple presolve iterations. This parameter controls the maximum number of such presolve iterations.
optional int32 max_presolve_iterations = 138 [default = 3];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23934 of file SatParameters.java.
| com.google.ortools.sat.SatParameters.MaxSatAssumptionOrder com.google.ortools.sat.SatParameters.Builder.getMaxSatAssumptionOrder | ( | ) |
optional .operations_research.sat.SatParameters.MaxSatAssumptionOrder max_sat_assumption_order = 51 [default = DEFAULT_ASSUMPTION_ORDER];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28206 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getMaxSatReverseAssumptionOrder | ( | ) |
If true, adds the assumption in the reverse order of the one defined by max_sat_assumption_order.
optional bool max_sat_reverse_assumption_order = 52 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28257 of file SatParameters.java.
| com.google.ortools.sat.SatParameters.MaxSatStratificationAlgorithm com.google.ortools.sat.SatParameters.Builder.getMaxSatStratification | ( | ) |
optional .operations_research.sat.SatParameters.MaxSatStratificationAlgorithm max_sat_stratification = 53 [default = STRATIFICATION_DESCENT];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28306 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getMaxSizeToCreatePrecedenceLiteralsInDisjunctive | ( | ) |
Create one literal for each disjunction of two pairs of tasks. This slows down the solve time, but improves the lower bound of the objective in the makespan case. This will be triggered if the number of intervals is less or equal than the parameter and if use_strong_propagation_in_disjunctive is true.
optional int32 max_size_to_create_precedence_literals_in_disjunctive = 229 [default = 60];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28539 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getMaxTimeInSeconds | ( | ) |
Maximum time allowed in seconds to solve a problem. The counter will starts at the beginning of the Solve() call.
optional double max_time_in_seconds = 36 [default = inf];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22128 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getMaxVariableActivityValue | ( | ) |
optional double max_variable_activity_value = 16 [default = 1e+100];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21101 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getMergeAtMostOneWorkLimit | ( | ) |
optional double merge_at_most_one_work_limit = 146 [default = 100000000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24978 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getMergeNoOverlapWorkLimit | ( | ) |
During presolve, we use a maximum clique heuristic to merge together no-overlap constraints or at most one constraints. This code can be slow, so we have a limit in place on the number of explored nodes in the underlying graph. The internal limit is an int64, but we use double here to simplify manual input.
optional double merge_no_overlap_work_limit = 145 [default = 1000000000000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24922 of file SatParameters.java.
| com.google.ortools.sat.SatParameters.ConflictMinimizationAlgorithm com.google.ortools.sat.SatParameters.Builder.getMinimizationAlgorithm | ( | ) |
optional .operations_research.sat.SatParameters.ConflictMinimizationAlgorithm minimization_algorithm = 4 [default = RECURSIVE];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20452 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getMinimizeReductionDuringPbResolution | ( | ) |
A different algorithm during PB resolution. It minimizes the number of calls to ReduceCoefficients() which can be time consuming. However, the search space will be different and if the coefficients are large, this may lead to integer overflows that could otherwise be prevented.
optional bool minimize_reduction_during_pb_resolution = 48 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23280 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getMinimizeSharedClauses | ( | ) |
Minimize and detect subsumption of shared clauses immediately after they are imported.
optional bool minimize_shared_clauses = 300 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27605 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getMinOrthogonalityForLpConstraints | ( | ) |
While adding constraints, skip the constraints which have orthogonality less than 'min_orthogonality_for_lp_constraints' with already added constraints during current call. Orthogonality is defined as 1 - cosine(vector angle between constraints). A value of zero disable this feature.
optional double min_orthogonality_for_lp_constraints = 115 [default = 0.05];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36503 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getMipAutomaticallyScaleVariables | ( | ) |
If true, some continuous variable might be automatically scaled. For now, this is only the case where we detect that a variable is actually an integer multiple of a constant. For instance, variables of the form k * 0.5 are quite frequent, and if we detect this, we will scale such variable domain by 2 to make it implied integer.
optional bool mip_automatically_scale_variables = 166 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37483 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getMipCheckPrecision | ( | ) |
As explained in mip_precision and mip_max_activity_exponent, we cannot always reach the wanted precision during scaling. We use this threshold to enphasize in the logs when the precision seems bad.
optional double mip_check_precision = 128 [default = 0.0001];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37855 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getMipComputeTrueObjectiveBound | ( | ) |
Even if we make big error when scaling the objective, we can always derive a correct lower bound on the original objective by using the exact lower bound on the scaled integer version of the objective. This should be fast, but if you don't care about having a precise lower bound, you can turn it off.
optional bool mip_compute_true_objective_bound = 198 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37923 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getMipDropTolerance | ( | ) |
Any value in the input mip with a magnitude lower than this will be set to zero. This is to avoid some issue in LP presolving.
optional double mip_drop_tolerance = 232 [default = 1e-16];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 38137 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getMipMaxActivityExponent | ( | ) |
To avoid integer overflow, we always force the maximum possible constraint activity (and objective value) according to the initial variable domain to be smaller than 2 to this given power. Because of this, we cannot always reach the "mip_wanted_precision" parameter above. This can go as high as 62, but some internal algo currently abort early if they might run into integer overflow, so it is better to keep it a bit lower than this.
optional int32 mip_max_activity_exponent = 127 [default = 53];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37781 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getMipMaxBound | ( | ) |
We need to bound the maximum magnitude of the variables for CP-SAT, and that is the bound we use. If the MIP model expect larger variable value in the solution, then the converted model will likely not be relevant.
optional double mip_max_bound = 124 [default = 10000000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37287 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getMipMaxValidMagnitude | ( | ) |
Any finite values in the input MIP must be below this threshold, otherwise the model will be reported invalid. This is needed to avoid floating point overflow when evaluating bounds * coeff for instance. We are a bit more defensive, but in practice, users shouldn't use super large values in a MIP.
optional double mip_max_valid_magnitude = 199 [default = 1e+20];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37995 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getMipPresolveLevel | ( | ) |
When solving a MIP, we do some basic floating point presolving before scaling the problem to integer to be handled by CP-SAT. This control how much of that presolve we do. It can help to better scale floating point model, but it is not always behaving nicely.
optional int32 mip_presolve_level = 261 [default = 2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 38201 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getMipScaleLargeDomain | ( | ) |
If this is false, then mip_var_scaling is only applied to variables with "small" domain. If it is true, we scale all floating point variable independenlty of their domain.
optional bool mip_scale_large_domain = 225 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37415 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getMipTreatHighMagnitudeBoundsAsInfinity | ( | ) |
By default, any variable/constraint bound with a finite value and a magnitude greater than the mip_max_valid_magnitude will result with a invalid model. This flags change the behavior such that such bounds are silently transformed to +∞ or -∞. It is recommended to keep it at false, and create valid bounds.
optional bool mip_treat_high_magnitude_bounds_as_infinity = 278 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 38069 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getMipVarScaling | ( | ) |
All continuous variable of the problem will be multiplied by this factor. By default, we don't do any variable scaling and rely on the MIP model to specify continuous variable domain with the wanted precision.
optional double mip_var_scaling = 125 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37351 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getMipWantedPrecision | ( | ) |
When scaling constraint with double coefficients to integer coefficients, we will multiply by a power of 2 and round the coefficients. We will choose the lowest power such that we have no potential overflow (see mip_max_activity_exponent) and the worst case constraint activity error does not exceed this threshold. Note that we also detect constraint with rational coefficients and scale them accordingly when it seems better instead of using a power of 2. We also relax all constraint bounds by this absolute value. For pure integer constraint, if this value if lower than one, this will not change anything. However it is needed when scaling MIP problems. If we manage to scale a constraint correctly, the maximum error we can make will be twice this value (once for the scaling error and once for the relaxed bounds). If we are not able to scale that well, we will display that fact but still scale as best as we can.
optional double mip_wanted_precision = 126 [default = 1e-06];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37679 of file SatParameters.java.
| java.lang.String com.google.ortools.sat.SatParameters.Builder.getName | ( | ) |
In some context, like in a portfolio of search, it makes sense to name a given parameters set for logging purpose.
optional string name = 171 [default = ""];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 19707 of file SatParameters.java.
| com.google.protobuf.ByteString com.google.ortools.sat.SatParameters.Builder.getNameBytes | ( | ) |
In some context, like in a portfolio of search, it makes sense to name a given parameters set for logging purpose.
optional string name = 171 [default = ""];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 19731 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getNewConstraintsBatchSize | ( | ) |
Add that many lazy constraints (or cuts) at once in the LP. Note that at the beginning of the solve, we do add more than this.
optional int32 new_constraints_batch_size = 122 [default = 50];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36849 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getNewLinearPropagation | ( | ) |
The new linear propagation code treat all constraints at once and use an adaptation of Bellman-Ford-Tarjan to propagate constraint in a smarter order and potentially detect propagation cycle earlier.
optional bool new_linear_propagation = 224 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35255 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getNoOverlap2DBooleanRelationsLimit | ( | ) |
If less than this number of boxes are present in a no-overlap 2d, we create 4 Booleans per pair of boxes: - Box 2 is after Box 1 on x. - Box 1 is after Box 2 on x. - Box 2 is after Box 1 on y. - Box 1 is after Box 2 on y. Note that at least one of them must be true, and at most one on x and one on y can be true. This can significantly help in closing small problem. The SAT reasoning can be a lot more powerful when we take decision on such positional relations.
optional int32 no_overlap_2d_boolean_relations_limit = 321 [default = 10];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29307 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getNumConflictsBeforeStrategyChanges | ( | ) |
After each restart, if the number of conflict since the last strategy change is greater that this, then we increment a "strategy_counter" that can be use to change the search strategy used by the following restarts.
optional int32 num_conflicts_before_strategy_changes = 68 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22006 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getNumFullSubsolvers | ( | ) |
We distinguish subsolvers that consume a full thread, and the ones that are always interleaved. If left at zero, we will fix this with a default formula that depends on num_workers. But if you start modifying what runs, you might want to fix that to a given value depending on the num_workers you use.
optional int32 num_full_subsolvers = 294 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25966 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getNumSearchWorkers | ( | ) |
optional int32 num_search_workers = 100 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25910 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getNumViolationLs | ( | ) |
This will create incomplete subsolvers (that are not LNS subsolvers) that use the feasibility jump code to find improving solution, treating the objective improvement as a hard constraint.
optional int32 num_violation_ls = 244 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32203 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getNumWorkers | ( | ) |
Specify the number of parallel workers (i.e. threads) to use during search. This should usually be lower than your number of available cpus + hyperthread in your machine. A value of 0 means the solver will try to use all cores on the machine. A number of 1 means no parallelism. Note that 'num_workers' is the preferred name, but if it is set to zero, we will still read the deprecated 'num_search_workers'. As of 2020-04-10, if you're using SAT via MPSolver (to solve integer programs) this field is overridden with a value of 8, if the field is not set *explicitly*. Thus, always set this field explicitly or via MPSolver::SetNumThreads().
optional int32 num_workers = 206 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25836 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getOnlyAddCutsAtLevelZero | ( | ) |
For the cut that can be generated at any level, this control if we only try to generate them at the root node.
optional bool only_add_cuts_at_level_zero = 92 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35713 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getOnlySolveIp | ( | ) |
If one try to solve a MIP model with CP-SAT, because we assume all variable to be integer after scaling, we will not necessarily have the correct optimal. Note however that all feasible solutions are valid since we will just solve a more restricted version of the original problem. This parameters is here to prevent user to think the solution is optimal when it might not be. One will need to manually set this to false to solve a MIP model where the optimal might be different. Note that this is tested after some MIP presolve steps, so even if not all original variable are integer, we might end up with a pure IP after presolve and after implied integer detection.
optional bool only_solve_ip = 222 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37569 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getOptimizeWithCore | ( | ) |
The default optimization method is a simple "linear scan", each time trying to find a better solution than the previous one. If this is true, then we use a core-based approach (like in max-SAT) when we try to increase the lower bound instead.
optional bool optimize_with_core = 83 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31237 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getOptimizeWithLbTreeSearch | ( | ) |
Do a more conventional tree search (by opposition to SAT based one) where we keep all the explored node in a tree. This is meant to be used in a portfolio and focus on improving the objective lower bound. Keeping the whole tree allow us to report a better objective lower bound coming from the worst open node in the tree.
optional bool optimize_with_lb_tree_search = 188 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31307 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getOptimizeWithMaxHs | ( | ) |
This has no effect if optimize_with_core is false. If true, use a different core-based algorithm similar to the max-HS algo for max-SAT. This is a hybrid MIP/CP approach and it uses a MIP solver in addition to the CP/SAT one. This is also related to the PhD work of tobyodavies@ "Automatic Logic-Based Benders Decomposition with MiniZinc" http://aaai.org/ocs/index.php/AAAI/AAAI17/paper/view/14489
optional bool optimize_with_max_hs = 85 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31525 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getPbCleanupIncrement | ( | ) |
Same as for the clauses, but for the learned pseudo-Boolean constraints.
optional int32 pb_cleanup_increment = 46 [default = 200];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20929 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getPbCleanupRatio | ( | ) |
optional double pb_cleanup_ratio = 47 [default = 0.5];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20977 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getPermutePresolveConstraintOrder | ( | ) |
optional bool permute_presolve_constraint_order = 179 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22770 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getPermuteVariableRandomly | ( | ) |
This is mainly here to test the solver variability. Note that in tests, if not explicitly set to false, all 3 options will be set to true so that clients do not rely on the solver returning a specific solution if they are many equivalent optimal solutions.
optional bool permute_variable_randomly = 178 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22716 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getPolarityExploitLsHints | ( | ) |
If true and we have first solution LS workers, tries in some phase to follow a LS solutions that violates has litle constraints as possible.
optional bool polarity_exploit_ls_hints = 309 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20059 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getPolarityRephaseIncrement | ( | ) |
If non-zero, then we change the polarity heuristic after that many number of conflicts in an arithmetically increasing fashion. So x the first time, 2 * x the second time, etc...
optional int32 polarity_rephase_increment = 168 [default = 1000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 19997 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getPolishLpSolution | ( | ) |
Whether we try to do a few degenerate iteration at the end of an LP solve to minimize the fractionality of the integer variable in the basis. This helps on some problems, but not so much on others. It also cost of bit of time to do such polish step.
optional bool polish_lp_solution = 175 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34741 of file SatParameters.java.
| com.google.ortools.sat.SatParameters.VariableOrder com.google.ortools.sat.SatParameters.Builder.getPreferredVariableOrder | ( | ) |
optional .operations_research.sat.SatParameters.VariableOrder preferred_variable_order = 1 [default = IN_ORDER];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 19808 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getPresolveBlockedClause | ( | ) |
Whether we use an heuristic to detect some basic case of blocked clause in the SAT presolve.
optional bool presolve_blocked_clause = 88 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23748 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getPresolveBvaThreshold | ( | ) |
Apply Bounded Variable Addition (BVA) if the number of clauses is reduced by stricly more than this threshold. The algorithm described in the paper uses 0, but quick experiments showed that 1 is a good value. It may not be worth it to add a new variable just to remove one clause.
optional int32 presolve_bva_threshold = 73 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23868 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getPresolveBveClauseWeight | ( | ) |
During presolve, we apply BVE only if this weight times the number of clauses plus the number of clause literals is not increased.
optional int32 presolve_bve_clause_weight = 55 [default = 3];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23576 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getPresolveBveThreshold | ( | ) |
During presolve, only try to perform the bounded variable elimination (BVE) of a variable x if the number of occurrences of x times the number of occurrences of not(x) is not greater than this parameter.
optional int32 presolve_bve_threshold = 54 [default = 500];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23434 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getPresolveExtractIntegerEnforcement | ( | ) |
If true, we will extract from linear constraints, enforcement literals of the form "integer variable at bound => simplified constraint". This should always be beneficial except that we don't always handle them as efficiently as we could for now. This causes problem on manna81.mps (LP relaxation not as tight it seems) and on neos-3354841-apure.mps.gz (too many literals created this way).
optional bool presolve_extract_integer_enforcement = 174 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25104 of file SatParameters.java.
| long com.google.ortools.sat.SatParameters.Builder.getPresolveInclusionWorkLimit | ( | ) |
A few presolve operations involve detecting constraints included in other constraint. Since there can be a quadratic number of such pairs, and processing them usually involve scanning them, the complexity of these operations can be big. This enforce a local deterministic limit on the number of entries scanned. Default is 1e8. A value of zero will disable these presolve rules completely.
optional int64 presolve_inclusion_work_limit = 201 [default = 100000000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25182 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getPresolveProbingDeterministicTimeLimit | ( | ) |
optional double presolve_probing_deterministic_time_limit = 57 [default = 30];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23698 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getPresolveSubstitutionLevel | ( | ) |
How much substitution (also called free variable aggregation in MIP litterature) should we perform at presolve. This currently only concerns variable appearing only in linear constraints. For now the value 0 turns it off and any positive value performs substitution.
optional int32 presolve_substitution_level = 147 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25032 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getPresolveUseBva | ( | ) |
Whether or not we use Bounded Variable Addition (BVA) in the presolve.
optional bool presolve_use_bva = 72 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23806 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getProbingDeterministicTimeLimit | ( | ) |
The maximum "deterministic" time limit to spend in probing. A value of zero will disable the probing. TODO(user): Clean up. The first one is used in CP-SAT, the other in pure SAT presolve.
optional double probing_deterministic_time_limit = 226 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23642 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getProbingNumCombinationsLimit | ( | ) |
How many combinations of pairs or triplets of variables we want to scan.
optional int32 probing_num_combinations_limit = 272 [default = 20000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30735 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getPropagationLoopDetectionFactor | ( | ) |
Some search decisions might cause a really large number of propagations to happen when integer variables with large domains are only reduced by 1 at each step. If we propagate more than the number of variable times this parameters we try to take counter-measure. Setting this to 0.0 disable this feature. TODO(user): Setting this to something like 10 helps in most cases, but the code is currently buggy and can cause the solve to enter a bad state where no progress is made.
optional double propagation_loop_detection_factor = 221 [default = 10];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28371 of file SatParameters.java.
| long com.google.ortools.sat.SatParameters.Builder.getPseudoCostReliabilityThreshold | ( | ) |
The solver ignores the pseudo costs of variables with number of recordings less than this threshold.
optional int64 pseudo_cost_reliability_threshold = 123 [default = 100];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31173 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getPushAllTasksTowardStart | ( | ) |
Experimental code: specify if the objective pushes all tasks toward the start of the schedule.
optional bool push_all_tasks_toward_start = 262 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34281 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getRandomBranchesRatio | ( | ) |
A number between 0 and 1 that indicates the proportion of branching variables that are selected randomly instead of choosing the first variable from the given variable_ordering strategy.
optional double random_branches_ratio = 32 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20193 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getRandomizeSearch | ( | ) |
Randomize fixed search.
optional bool randomize_search = 103 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34159 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getRandomPolarityRatio | ( | ) |
The proportion of polarity chosen at random. Note that this take precedence over the phase saving heuristic. This is different from initial_polarity:POLARITY_RANDOM because it will select a new random polarity each time the variable is branched upon instead of selecting one initially and then always taking this choice.
optional double random_polarity_ratio = 45 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20125 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getRandomSeed | ( | ) |
At the beginning of each solve, the random number generator used in some part of the solver is reinitialized to this seed. If you change the random seed, the solver may make different choices during the solving process. For some problems, the running time may vary a lot depending on small change in the solving algorithm. Running the solver with different seeds enables to have more robust benchmarks when evaluating new features.
optional int32 random_seed = 31 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22642 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getRelativeGapLimit | ( | ) |
optional double relative_gap_limit = 160 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22582 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getRemoveFixedVariablesEarly | ( | ) |
If cp_model_presolve is true and there is a large proportion of fixed variable after the first model copy, remap all the model to a dense set of variable before the full presolve even starts. This should help for LNS on large models.
optional bool remove_fixed_variables_early = 310 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24168 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getRepairHint | ( | ) |
If true, the solver tries to repair the solution given in the hint. This search terminates after the 'hint_conflict_limit' is reached and the solver switches to regular search. If false, then we do a FIXED_SEARCH using the hint until the hint_conflict_limit is reached.
optional bool repair_hint = 167 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30493 of file SatParameters.java.
| com.google.ortools.sat.SatParameters.RestartAlgorithm com.google.ortools.sat.SatParameters.Builder.getRestartAlgorithms | ( | int | index | ) |
The restart strategies will change each time the strategy_counter is increased. The current strategy will simply be the one at index strategy_counter modulo the number of strategy. Note that if this list includes a NO_RESTART, nothing will change when it is reached because the strategy_counter will only increment after a restart. The idea of switching of search strategy tailored for SAT/UNSAT comes from Chanseok Oh with his COMiniSatPS solver, see http://cs.nyu.edu/~chanseok/. But more generally, it seems REALLY beneficial to try different strategy.
repeated .operations_research.sat.SatParameters.RestartAlgorithm restart_algorithms = 61;
| index | The index of the element to return. |
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21438 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getRestartAlgorithmsCount | ( | ) |
The restart strategies will change each time the strategy_counter is increased. The current strategy will simply be the one at index strategy_counter modulo the number of strategy. Note that if this list includes a NO_RESTART, nothing will change when it is reached because the strategy_counter will only increment after a restart. The idea of switching of search strategy tailored for SAT/UNSAT comes from Chanseok Oh with his COMiniSatPS solver, see http://cs.nyu.edu/~chanseok/. But more generally, it seems REALLY beneficial to try different strategy.
repeated .operations_research.sat.SatParameters.RestartAlgorithm restart_algorithms = 61;
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21418 of file SatParameters.java.
| java.util.List< com.google.ortools.sat.SatParameters.RestartAlgorithm > com.google.ortools.sat.SatParameters.Builder.getRestartAlgorithmsList | ( | ) |
The restart strategies will change each time the strategy_counter is increased. The current strategy will simply be the one at index strategy_counter modulo the number of strategy. Note that if this list includes a NO_RESTART, nothing will change when it is reached because the strategy_counter will only increment after a restart. The idea of switching of search strategy tailored for SAT/UNSAT comes from Chanseok Oh with his COMiniSatPS solver, see http://cs.nyu.edu/~chanseok/. But more generally, it seems REALLY beneficial to try different strategy.
repeated .operations_research.sat.SatParameters.RestartAlgorithm restart_algorithms = 61;
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21398 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getRestartDlAverageRatio | ( | ) |
In the moving average restart algorithms, a restart is triggered if the window average times this ratio is greater that the global average.
optional double restart_dl_average_ratio = 63 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21760 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getRestartLbdAverageRatio | ( | ) |
optional double restart_lbd_average_ratio = 71 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21810 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getRestartPeriod | ( | ) |
Restart period for the FIXED_RESTART strategy. This is also the multiplier used by the LUBY_RESTART strategy.
optional int32 restart_period = 30 [default = 50];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21644 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getRestartRunningWindowSize | ( | ) |
Size of the window for the moving average restarts.
optional int32 restart_running_window_size = 62 [default = 50];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21702 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getRootLpIterations | ( | ) |
Even at the root node, we do not want to spend too much time on the LP if it is "difficult". So we solve it in "chunks" of that many iterations. The solve will be continued down in the tree or the next time we go back to the root node.
optional int32 root_lp_iterations = 227 [default = 2000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36433 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getRoutingCutDpEffort | ( | ) |
The amount of "effort" to spend in dynamic programming for computing routing cuts. This is in term of basic operations needed by the algorithm in the worst case, so a value like 1e8 should take less than a second to compute.
optional double routing_cut_dp_effort = 314 [default = 10000000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30269 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getRoutingCutMaxInfeasiblePathLength | ( | ) |
If the length of an infeasible path is less than this value, a cut will be added to exclude it.
optional int32 routing_cut_max_infeasible_path_length = 317 [default = 6];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30333 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getRoutingCutSubsetSizeForBinaryRelationBound | ( | ) |
If the size of a subset of nodes of a RoutesConstraint is less than this value, use linear constraints of size 1 and 2 (such as capacity and time window constraints) enforced by the arc literals to compute cuts for this subset (unless the subset size is less than routing_cut_subset_size_for_tight_binary_relation_bound, in which case the corresponding algorithm is used instead). The algorithm for these cuts has a O(n^3) complexity, where n is the subset size. Hence the value of this parameter should not be too large (e.g. 10 or 20).
optional int32 routing_cut_subset_size_for_binary_relation_bound = 312 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29997 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getRoutingCutSubsetSizeForExactBinaryRelationBound | ( | ) |
Similar to above, but with an even stronger algorithm in O(n!). We try to be defensive and abort early or not run that often. Still the value of that parameter shouldn't really be much more than 10.
optional int32 routing_cut_subset_size_for_exact_binary_relation_bound = 316 [default = 8];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30135 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getRoutingCutSubsetSizeForShortestPathsBound | ( | ) |
Similar to routing_cut_subset_size_for_exact_binary_relation_bound but use a bound based on shortest path distances (which respect triangular inequality). This allows to derive bounds that are valid for any superset of a given subset. This is slow, so it shouldn't really be larger than 10.
optional int32 routing_cut_subset_size_for_shortest_paths_bound = 318 [default = 8];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30201 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getRoutingCutSubsetSizeForTightBinaryRelationBound | ( | ) |
Similar to above, but with a different algorithm producing better cuts, at the price of a higher O(2^n) complexity, where n is the subset size. Hence the value of this parameter should be small (e.g. less than 10).
optional int32 routing_cut_subset_size_for_tight_binary_relation_bound = 313 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30071 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getSaveLpBasisInLbTreeSearch | ( | ) |
Experimental. Save the current LP basis at each node of the search tree so that when we jump around, we can load it and reduce the number of LP iterations needed. It currently works okay if we do not change the lp with cuts or simplification... More work is needed to make it robust in all cases.
optional bool save_lp_basis_in_lb_tree_search = 284 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31381 of file SatParameters.java.
| com.google.ortools.sat.SatParameters.SearchBranching com.google.ortools.sat.SatParameters.Builder.getSearchBranching | ( | ) |
optional .operations_research.sat.SatParameters.SearchBranching search_branching = 82 [default = AUTOMATIC_SEARCH];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30382 of file SatParameters.java.
| long com.google.ortools.sat.SatParameters.Builder.getSearchRandomVariablePoolSize | ( | ) |
Search randomization will collect the top 'search_random_variable_pool_size' valued variables, and pick one randomly. The value of the variable is specific to each strategy.
optional int64 search_random_variable_pool_size = 104 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34219 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getShareBinaryClauses | ( | ) |
Allows sharing of new learned binary clause between workers.
optional bool share_binary_clauses = 203 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27487 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getSharedTreeBalanceTolerance | ( | ) |
How much deeper compared to the ideal max depth of the tree is considered "balanced" enough to still accept a split. Without such a tolerance, sometimes the tree can only be split by a single worker, and they may not generate a split for some time. In contrast, with a tolerance of 1, at least half of all workers should be able to split the tree as soon as a split becomes required. This only has an effect on SPLIT_STRATEGY_BALANCED_TREE and SPLIT_STRATEGY_DISCREPANCY.
optional int32 shared_tree_balance_tolerance = 305 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32871 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getSharedTreeMaxNodesPerWorker | ( | ) |
In order to limit total shared memory and communication overhead, limit the total number of nodes that may be generated in the shared tree. If the shared tree runs out of unassigned leaves, workers act as portfolio workers. Note: this limit includes interior nodes, not just leaves.
optional int32 shared_tree_max_nodes_per_worker = 238 [default = 10000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32757 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getSharedTreeNumWorkers | ( | ) |
Enables shared tree search. If positive, start this many complete worker threads to explore a shared search tree. These workers communicate objective bounds and simple decision nogoods relating to the shared prefix of the tree, and will avoid exploring the same subtrees as one another. Specifying a negative number uses a heuristic to select an appropriate number of shared tree workeres based on the total number of workers.
optional int32 shared_tree_num_workers = 235 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32391 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getSharedTreeOpenLeavesPerWorker | ( | ) |
How many open leaf nodes should the shared tree maintain per worker.
optional double shared_tree_open_leaves_per_worker = 281 [default = 2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32695 of file SatParameters.java.
| com.google.ortools.sat.SatParameters.SharedTreeSplitStrategy com.google.ortools.sat.SatParameters.Builder.getSharedTreeSplitStrategy | ( | ) |
optional .operations_research.sat.SatParameters.SharedTreeSplitStrategy shared_tree_split_strategy = 239 [default = SPLIT_STRATEGY_AUTO];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32810 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getSharedTreeWorkerEnablePhaseSharing | ( | ) |
If true, shared tree workers share their target phase when returning an assigned subtree for the next worker to use.
optional bool shared_tree_worker_enable_phase_sharing = 304 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32637 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getSharedTreeWorkerEnableTrailSharing | ( | ) |
If true, workers share more of the information from their local trail. Specifically, literals implied by the shared tree decisions.
optional bool shared_tree_worker_enable_trail_sharing = 295 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32577 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getSharedTreeWorkerMinRestartsPerSubtree | ( | ) |
Minimum restarts before a worker will replace a subtree that looks "bad" based on the average LBD of learned clauses.
optional int32 shared_tree_worker_min_restarts_per_subtree = 282 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32517 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getShareGlueClauses | ( | ) |
Allows sharing of short glue clauses between workers. Implicitly disabled if share_binary_clauses is false.
optional bool share_glue_clauses = 285 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27545 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getShareGlueClausesDtime | ( | ) |
The amount of dtime between each export of shared glue clauses.
optional double share_glue_clauses_dtime = 322 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27663 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getShareLevelZeroBounds | ( | ) |
Allows sharing of the bounds of modified variables at level 0.
optional bool share_level_zero_bounds = 114 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27431 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getShareObjectiveBounds | ( | ) |
Allows objective sharing between workers.
optional bool share_objective_bounds = 113 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27375 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getShavingDeterministicTimeInProbingSearch | ( | ) |
Add a shaving phase (where the solver tries to prove that the lower or upper bound of a variable are infeasible) to the probing search. (<= 0 disables it).
optional double shaving_deterministic_time_in_probing_search = 204 [default = 0.001];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30795 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getShavingSearchDeterministicTime | ( | ) |
Specifies the amount of deterministic time spent of each try at shaving a bound in the shaving search.
optional double shaving_search_deterministic_time = 205 [default = 0.1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30857 of file SatParameters.java.
| long com.google.ortools.sat.SatParameters.Builder.getShavingSearchThreshold | ( | ) |
Specifies the threshold between two modes in the shaving procedure. If the range of the variable/objective is less than this threshold, then the shaving procedure will try to remove values one by one. Otherwise, it will try to remove one range at a time.
optional int64 shaving_search_threshold = 290 [default = 64];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30921 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getSolutionPoolSize | ( | ) |
Size of the top-n different solutions kept by the solver. This parameter must be > 0. Currently this only impact the "base" solution chosen for a LNS fragment.
optional int32 solution_pool_size = 193 [default = 3];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33771 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getStopAfterFirstSolution | ( | ) |
For an optimization problem, stop the solver as soon as we have a solution.
optional bool stop_after_first_solution = 98 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33403 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getStopAfterPresolve | ( | ) |
Mainly used when improving the presolver. When true, stops the solver after the presolve is complete (or after loading and root level propagation).
optional bool stop_after_presolve = 149 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33461 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getStopAfterRootPropagation | ( | ) |
optional bool stop_after_root_propagation = 252 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33511 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getStrategyChangeIncreaseRatio | ( | ) |
The parameter num_conflicts_before_strategy_changes is increased by that much after each strategy change.
optional double strategy_change_increase_ratio = 69 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22068 of file SatParameters.java.
| com.google.ortools.sat.SatParameters com.google.ortools.sat.SatParameters.Builder.getSubsolverParams | ( | int | index | ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26912 of file SatParameters.java.
| com.google.ortools.sat.SatParameters.Builder com.google.ortools.sat.SatParameters.Builder.getSubsolverParamsBuilder | ( | int | index | ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Definition at line 27135 of file SatParameters.java.
| java.util.List< com.google.ortools.sat.SatParameters.Builder > com.google.ortools.sat.SatParameters.Builder.getSubsolverParamsBuilderList | ( | ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Definition at line 27219 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getSubsolverParamsCount | ( | ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26894 of file SatParameters.java.
| java.util.List< com.google.ortools.sat.SatParameters > com.google.ortools.sat.SatParameters.Builder.getSubsolverParamsList | ( | ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26876 of file SatParameters.java.
| com.google.ortools.sat.SatParametersOrBuilder com.google.ortools.sat.SatParameters.Builder.getSubsolverParamsOrBuilder | ( | int | index | ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27150 of file SatParameters.java.
| java.util.List<? extends com.google.ortools.sat.SatParametersOrBuilder > com.google.ortools.sat.SatParameters.Builder.getSubsolverParamsOrBuilderList | ( | ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27169 of file SatParameters.java.
| java.lang.String com.google.ortools.sat.SatParameters.Builder.getSubsolvers | ( | int | index | ) |
In multi-thread, the solver can be mainly seen as a portfolio of solvers with different parameters. This field indicates the names of the parameters that are used in multithread. This only applies to "full" subsolvers. See cp_model_search.cc to see a list of the names and the default value (if left empty) that looks like: - default_lp (linearization_level:1) - fixed (only if fixed search specified or scheduling) - no_lp (linearization_level:0) - max_lp (linearization_level:2) - pseudo_costs (only if objective, change search heuristic) - reduced_costs (only if objective, change search heuristic) - quick_restart (kind of probing) - quick_restart_no_lp (kind of probing with linearization_level:0) - lb_tree_search (to improve lower bound, MIP like tree search) - probing (continuous probing and shaving) Also, note that some set of parameters will be ignored if they do not make sense. For instance if there is no objective, pseudo_cost or reduced_cost search will be ignored. Core based search will only work if the objective has many terms. If there is no fixed strategy fixed will be ignored. And so on. The order is important, as only the first num_full_subsolvers will be scheduled. You can see in the log which one are selected for a given run.
repeated string subsolvers = 207;
| index | The index of the element to return. |
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26121 of file SatParameters.java.
| com.google.protobuf.ByteString com.google.ortools.sat.SatParameters.Builder.getSubsolversBytes | ( | int | index | ) |
In multi-thread, the solver can be mainly seen as a portfolio of solvers with different parameters. This field indicates the names of the parameters that are used in multithread. This only applies to "full" subsolvers. See cp_model_search.cc to see a list of the names and the default value (if left empty) that looks like: - default_lp (linearization_level:1) - fixed (only if fixed search specified or scheduling) - no_lp (linearization_level:0) - max_lp (linearization_level:2) - pseudo_costs (only if objective, change search heuristic) - reduced_costs (only if objective, change search heuristic) - quick_restart (kind of probing) - quick_restart_no_lp (kind of probing with linearization_level:0) - lb_tree_search (to improve lower bound, MIP like tree search) - probing (continuous probing and shaving) Also, note that some set of parameters will be ignored if they do not make sense. For instance if there is no objective, pseudo_cost or reduced_cost search will be ignored. Core based search will only work if the objective has many terms. If there is no fixed strategy fixed will be ignored. And so on. The order is important, as only the first num_full_subsolvers will be scheduled. You can see in the log which one are selected for a given run.
repeated string subsolvers = 207;
| index | The index of the value to return. |
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26158 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getSubsolversCount | ( | ) |
In multi-thread, the solver can be mainly seen as a portfolio of solvers with different parameters. This field indicates the names of the parameters that are used in multithread. This only applies to "full" subsolvers. See cp_model_search.cc to see a list of the names and the default value (if left empty) that looks like: - default_lp (linearization_level:1) - fixed (only if fixed search specified or scheduling) - no_lp (linearization_level:0) - max_lp (linearization_level:2) - pseudo_costs (only if objective, change search heuristic) - reduced_costs (only if objective, change search heuristic) - quick_restart (kind of probing) - quick_restart_no_lp (kind of probing with linearization_level:0) - lb_tree_search (to improve lower bound, MIP like tree search) - probing (continuous probing and shaving) Also, note that some set of parameters will be ignored if they do not make sense. For instance if there is no objective, pseudo_cost or reduced_cost search will be ignored. Core based search will only work if the objective has many terms. If there is no fixed strategy fixed will be ignored. And so on. The order is important, as only the first num_full_subsolvers will be scheduled. You can see in the log which one are selected for a given run.
repeated string subsolvers = 207;
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26085 of file SatParameters.java.
| com.google.protobuf.ProtocolStringList com.google.ortools.sat.SatParameters.Builder.getSubsolversList | ( | ) |
In multi-thread, the solver can be mainly seen as a portfolio of solvers with different parameters. This field indicates the names of the parameters that are used in multithread. This only applies to "full" subsolvers. See cp_model_search.cc to see a list of the names and the default value (if left empty) that looks like: - default_lp (linearization_level:1) - fixed (only if fixed search specified or scheduling) - no_lp (linearization_level:0) - max_lp (linearization_level:2) - pseudo_costs (only if objective, change search heuristic) - reduced_costs (only if objective, change search heuristic) - quick_restart (kind of probing) - quick_restart_no_lp (kind of probing with linearization_level:0) - lb_tree_search (to improve lower bound, MIP like tree search) - probing (continuous probing and shaving) Also, note that some set of parameters will be ignored if they do not make sense. For instance if there is no objective, pseudo_cost or reduced_cost search will be ignored. Core based search will only work if the objective has many terms. If there is no fixed strategy fixed will be ignored. And so on. The order is important, as only the first num_full_subsolvers will be scheduled. You can see in the log which one are selected for a given run.
repeated string subsolvers = 207;
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 26049 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getSubsumptionDuringConflictAnalysis | ( | ) |
At a really low cost, during the 1-UIP conflict computation, it is easy to detect if some of the involved reasons are subsumed by the current conflict. When this is true, such clauses are detached and later removed from the problem.
optional bool subsumption_during_conflict_analysis = 56 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20547 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getSymmetryDetectionDeterministicTimeLimit | ( | ) |
Deterministic time limit for symmetry detection.
optional double symmetry_detection_deterministic_time_limit = 302 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35195 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getSymmetryLevel | ( | ) |
Whether we try to automatically detect the symmetries in a model and exploit them. Currently, at level 1 we detect them in presolve and try to fix Booleans. At level 2, we also do some form of dynamic symmetry breaking during search. At level 3, we also detect symmetries for very large models, which can be slow. At level 4, we try to break as much symmetry as possible in presolve.
optional int32 symmetry_level = 183 [default = 2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34989 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getTableCompressionLevel | ( | ) |
How much we try to "compress" a table constraint. Compressing more leads to less Booleans and faster propagation but can reduced the quality of the lp relaxation. Values goes from 0 to 3 where we always try to fully compress a table. At 2, we try to automatically decide if it is worth it.
optional int32 table_compression_level = 217 [default = 2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24312 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseAbslRandom | ( | ) |
optional bool use_absl_random = 180 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22810 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseAllDifferentForCircuit | ( | ) |
Turn on extra propagation for the circuit constraint. This can be quite slow.
optional bool use_all_different_for_circuit = 311 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29925 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseAreaEnergeticReasoningInNoOverlap2D | ( | ) |
When this is true, the no_overlap_2d constraint is reinforced with an energetic reasoning that uses an area-based energy. This can be combined with the two other overlap heuristics above.
optional bool use_area_energetic_reasoning_in_no_overlap_2d = 271 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29527 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseBlockingRestart | ( | ) |
Block a moving restart algorithm if the trail size of the current conflict is greater than the multiplier times the moving average of the trail size at the previous conflicts.
optional bool use_blocking_restart = 64 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21862 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseCombinedNoOverlap | ( | ) |
This can be beneficial if there is a lot of no-overlap constraints but a relatively low number of different intervals in the problem. Like 1000 intervals, but 1M intervals in the no-overlap constraints covering them.
optional bool use_combined_no_overlap = 133 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34487 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseConservativeScaleOverloadChecker | ( | ) |
Enable a heuristic to solve cumulative constraints using a modified energy constraint. We modify the usual energy definition by applying a super-additive function (also called "conservative scale" or "dual-feasible function") to the demand and the durations of the tasks. This heuristic is fast but for most problems it does not help much to find a solution.
optional bool use_conservative_scale_overload_checker = 286 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28867 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseDisjunctiveConstraintInCumulative | ( | ) |
When this is true, the cumulative constraint is reinforced with propagators from the disjunctive constraint to improve the inference on a set of tasks that are disjunctive at the root of the problem. This additional level supplements the default level of reasoning. Propagators of the cumulative constraint will not be used at all if all the tasks are disjunctive at root node. This always result in better propagation, but it is usually slow, so depending on the problem, turning this off may lead to a faster solution.
optional bool use_disjunctive_constraint_in_cumulative = 80 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29209 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseDualSchedulingHeuristics | ( | ) |
When set, it activates a few scheduling parameters to improve the lower bound of scheduling problems. This is only effective with multiple workers as it modifies the reduced_cost, lb_tree_search, and probing workers.
optional bool use_dual_scheduling_heuristics = 214 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29863 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseDynamicPrecedenceInCumulative | ( | ) |
optional bool use_dynamic_precedence_in_cumulative = 268 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28727 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseDynamicPrecedenceInDisjunctive | ( | ) |
Whether we try to branch on decision "interval A before interval B" rather than on intervals bounds. This usually works better, but slow down a bit the time to find the first solution. These parameters are still EXPERIMENTAL, the result should be correct, but it some corner cases, they can cause some failing CHECK in the solver.
optional bool use_dynamic_precedence_in_disjunctive = 263 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28669 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseEnergeticReasoningInNoOverlap2D | ( | ) |
When this is true, the no_overlap_2d constraint is reinforced with energetic reasoning. This additional level supplements the default level of reasoning.
optional bool use_energetic_reasoning_in_no_overlap_2d = 213 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29463 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseErwaHeuristic | ( | ) |
Whether we use the ERWA (Exponential Recency Weighted Average) heuristic as described in "Learning Rate Based Branching Heuristic for SAT solvers", J.H.Liang, V. Ganesh, P. Poupart, K.Czarnecki, SAT 2016.
optional bool use_erwa_heuristic = 75 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20257 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseExactLpReason | ( | ) |
The solver usually exploit the LP relaxation of a model. If this option is true, then whatever is infered by the LP will be used like an heuristic to compute EXACT propagation on the IP. So with this option, there is no numerical imprecision issues.
optional bool use_exact_lp_reason = 109 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34421 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseExtendedProbing | ( | ) |
Use extended probing (probe bool_or, at_most_one, exactly_one).
optional bool use_extended_probing = 269 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30679 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseFeasibilityJump | ( | ) |
Parameters for an heuristic similar to the one described in the paper: "Feasibility Jump: an LP-free Lagrangian MIP heuristic", Bjørnar Luteberget, Giorgio Sartor, 2023, Mathematical Programming Computation.
optional bool use_feasibility_jump = 265 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31595 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseFeasibilityPump | ( | ) |
Adds a feasibility pump subsolver along with lns subsolvers.
optional bool use_feasibility_pump = 164 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33887 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseHardPrecedencesInCumulative | ( | ) |
If true, detect and create constraint for integer variable that are "after" a set of intervals in the same cumulative constraint. Experimental: by default we just use "direct" precedences. If exploit_all_precedences is true, we explore the full precedence graph. This assumes we have a DAG otherwise it fails.
optional bool use_hard_precedences_in_cumulative = 215 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29085 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseImpliedBounds | ( | ) |
Stores and exploits "implied-bounds" in the solver. That is, relations of the form literal => (var >= bound). This is currently used to derive stronger cuts.
optional bool use_implied_bounds = 144 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34675 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseLbRelaxLns | ( | ) |
Turns on neighborhood generator based on local branching LP. Based on Huang et al., "Local Branching Relaxation Heuristics for Integer Linear Programs", 2023.
optional bool use_lb_relax_lns = 255 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33947 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseLinear3ForNoOverlap2DPrecedences | ( | ) |
When set, this activates a propagator for the no_overlap_2d constraint that
uses any eventual linear constraints of the model in the form
`{start interval 1} - {end interval 2} + c*w <= ub` to detect that two
intervals must overlap in one dimension for some values of `w`. This is
particularly useful for problems where the distance between two boxes is
part of the model.
optional bool use_linear3_for_no_overlap_2d_precedences = 323 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29793 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseLns | ( | ) |
Testing parameters used to disable all lns workers.
optional bool use_lns = 283 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33655 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseLnsOnly | ( | ) |
Experimental parameters to disable everything but lns.
optional bool use_lns_only = 101 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33711 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseLsOnly | ( | ) |
Disable every other type of subsolver, setting this turns CP-SAT into a pure local-search solver.
optional bool use_ls_only = 240 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31657 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseObjectiveLbSearch | ( | ) |
If true, search will search in ascending max objective value (when minimizing) starting from the lower bound of the objective.
optional bool use_objective_lb_search = 228 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30985 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseObjectiveShavingSearch | ( | ) |
This search differs from the previous search as it will not use assumptions to bound the objective, and it will recreate a full model with the hardcoded objective value.
optional bool use_objective_shaving_search = 253 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31047 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseOptimizationHints | ( | ) |
For an optimization problem, whether we follow some hints in order to find a better first solution. For a variable with hint, the solver will always try to follow the hint. It will revert to the variable_branching default otherwise.
optional bool use_optimization_hints = 35 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27961 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseOptionalVariables | ( | ) |
If true, we automatically detect variables whose constraint are always enforced by the same literal and we mark them as optional. This allows to propagate them as if they were present in some situation. TODO(user): This is experimental and seems to lead to wrong optimal in some situation. It should however gives correct solutions. Fix.
optional bool use_optional_variables = 108 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34349 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseOverloadCheckerInCumulative | ( | ) |
When this is true, the cumulative constraint is reinforced with overload checking, i.e., an additional level of reasoning based on energy. This additional level supplements the default level of reasoning as well as timetable edge finding. This always result in better propagation, but it is usually slow, so depending on the problem, turning this off may lead to a faster solution.
optional bool use_overload_checker_in_cumulative = 78 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28787 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUsePbResolution | ( | ) |
Whether to use pseudo-Boolean resolution to analyze a conflict. Note that this option only make sense if your problem is modelized using pseudo-Boolean constraints. If you only have clauses, this shouldn't change anything (except slow the solver down).
optional bool use_pb_resolution = 43 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23212 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUsePhaseSaving | ( | ) |
If this is true, then the polarity of a variable will be the last value it was assigned to, or its default polarity if it was never assigned since the call to ResetDecisionHeuristic(). Actually, we use a newer version where we follow the last value in the longest non-conflicting partial assignment in the current phase. This is called 'literal phase saving'. For details see 'A Lightweight Component Caching Scheme for Satisfiability Solvers' K. Pipatsrisawat and A.Darwiche, In 10th International Conference on Theory and Applications of Satisfiability Testing, 2007.
optional bool use_phase_saving = 44 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 19917 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUsePrecedencesInDisjunctiveConstraint | ( | ) |
When this is true, then a disjunctive constraint will try to use the precedence relations between time intervals to propagate their bounds further. For instance if task A and B are both before C and task A and B are in disjunction, then we can deduce that task C must start after duration(A) + duration(B) instead of simply max(duration(A), duration(B)), provided that the start time for all task was currently zero. This always result in better propagation, but it is usually slow, so depending on the problem, turning this off may lead to a faster solution.
optional bool use_precedences_in_disjunctive_constraint = 74 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28459 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseProbingSearch | ( | ) |
If true, search will continuously probe Boolean variables, and integer variable bounds. This parameter is set to true in parallel on the probing worker.
optional bool use_probing_search = 176 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30619 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseRinsLns | ( | ) |
Turns on relaxation induced neighborhood generator.
optional bool use_rins_lns = 129 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33831 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseSatInprocessing | ( | ) |
Enable or disable "inprocessing" which is some SAT presolving done at each restart to the root level.
optional bool use_sat_inprocessing = 163 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25460 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseSharedTreeSearch | ( | ) |
Set on shared subtree workers. Users should not set this directly.
optional bool use_shared_tree_search = 236 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32459 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseStrongPropagationInDisjunctive | ( | ) |
Enable stronger and more expensive propagation on no_overlap constraint.
optional bool use_strong_propagation_in_disjunctive = 230 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28603 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseSymmetryInLp | ( | ) |
When we have symmetry, it is possible to "fold" all variables from the same orbit into a single variable, while having the same power of LP relaxation. This can help significantly on symmetric problem. However there is currently a bit of overhead as the rest of the solver need to do some translation between the folded LP and the rest of the problem.
optional bool use_symmetry_in_lp = 301 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35063 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseTimetableEdgeFindingInCumulative | ( | ) |
When this is true, the cumulative constraint is reinforced with timetable edge finding, i.e., an additional level of reasoning based on the conjunction of energy and mandatory parts. This additional level supplements the default level of reasoning as well as overload_checker. This always result in better propagation, but it is usually slow, so depending on the problem, turning this off may lead to a faster solution.
optional bool use_timetable_edge_finding_in_cumulative = 79 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28947 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseTimetablingInNoOverlap2D | ( | ) |
When this is true, the no_overlap_2d constraint is reinforced with propagators from the cumulative constraints. It consists of ignoring the position of rectangles in one position and projecting the no_overlap_2d on the other dimension to create a cumulative constraint. This is done on both axis. This additional level supplements the default level of reasoning.
optional bool use_timetabling_in_no_overlap_2d = 200 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29395 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.getUseTryEdgeReasoningInNoOverlap2D | ( | ) |
optional bool use_try_edge_reasoning_in_no_overlap_2d = 299 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29579 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getVariableActivityDecay | ( | ) |
Each time a conflict is found, the activities of some variables are increased by one. Then, the activity of all variables are multiplied by variable_activity_decay. To implement this efficiently, the activity of all the variables is not decayed at each conflict. Instead, the activity increment is multiplied by 1 / decay. When an activity reach max_variable_activity_value, all the activity are multiplied by 1 / max_variable_activity_value.
optional double variable_activity_decay = 15 [default = 0.8];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21039 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getVariablesShavingLevel | ( | ) |
This search takes all Boolean or integer variables, and maximize or minimize them in order to reduce their domain. -1 is automatic, otherwise value 0 disables it, and 1, 2, or 3 changes something.
optional int32 variables_shaving_level = 289 [default = -1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31111 of file SatParameters.java.
| double com.google.ortools.sat.SatParameters.Builder.getViolationLsCompoundMoveProbability | ( | ) |
Probability of using compound move search each restart. TODO(user): Add reference to paper when published.
optional double violation_ls_compound_move_probability = 259 [default = 0.5];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32321 of file SatParameters.java.
| int com.google.ortools.sat.SatParameters.Builder.getViolationLsPerturbationPeriod | ( | ) |
How long violation_ls should wait before perturbating a solution.
optional int32 violation_ls_perturbation_period = 249 [default = 100];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32263 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasAbsoluteGapLimit | ( | ) |
Stop the search when the gap between the best feasible objective (O) and our best objective bound (B) is smaller than a limit. The exact definition is: - Absolute: abs(O - B) - Relative: abs(O - B) / max(1, abs(O)). Important: The relative gap depends on the objective offset! If you artificially shift the objective, you will get widely different value of the relative gap. Note that if the gap is reached, the search status will be OPTIMAL. But one can check the best objective bound to see the actual gap. If the objective is integer, then any absolute gap < 1 will lead to a true optimal. If the objective is floating point, a gap of zero make little sense so is is why we use a non-zero default value. At the end of the search, we will display a warning if OPTIMAL is reported yet the gap is greater than this absolute gap.
optional double absolute_gap_limit = 159 [default = 0.0001];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22471 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasAddCgCuts | ( | ) |
Whether we generate and add Chvatal-Gomory cuts to the LP at root node. Note that for now, this is not heavily tuned.
optional bool add_cg_cuts = 117 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35832 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasAddCliqueCuts | ( | ) |
Whether we generate clique cuts from the binary implication graph. Note that as the search goes on, this graph will contains new binary clauses learned by the SAT engine.
optional bool add_clique_cuts = 172 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36013 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasAddLinMaxCuts | ( | ) |
For the lin max constraints, generates the cuts described in "Strong mixed-integer programming formulations for trained neural networks" by Ross Anderson et. (https://arxiv.org/pdf/1811.01988.pdf)
optional bool add_lin_max_cuts = 152 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36201 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasAddLpConstraintsLazily | ( | ) |
If true, we start by an empty LP, and only add constraints not satisfied by the current LP solution batch by batch. A constraint that is only added like this is known as a "lazy" constraint in the literature, except that we currently consider all constraints as lazy here.
optional bool add_lp_constraints_lazily = 112 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36350 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasAddMirCuts | ( | ) |
Whether we generate MIR cuts at root node. Note that for now, this is not heavily tuned.
optional bool add_mir_cuts = 120 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35892 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasAddObjectiveCut | ( | ) |
When the LP objective is fractional, do we add the cut that forces the linear objective expression to be greater or equal to this fractional value rounded up? We can always do that since our objective is integer, and combined with MIR heuristic to reduce the coefficient of such cut, it can help.
optional bool add_objective_cut = 197 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35763 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasAddRltCuts | ( | ) |
Whether we generate RLT cuts. This is still experimental but can help on binary problem with a lot of clauses of size 3.
optional bool add_rlt_cuts = 279 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36076 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasAddZeroHalfCuts | ( | ) |
Whether we generate Zero-Half cuts at root node. Note that for now, this is not heavily tuned.
optional bool add_zero_half_cuts = 169 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35952 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasAlsoBumpVariablesInConflictReasons | ( | ) |
When this is true, then the variables that appear in any of the reason of the variables in a conflict have their activity bumped. This is addition to the variables in the conflict, and the one that were used during conflict resolution.
optional bool also_bump_variables_in_conflict_reasons = 77 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20384 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasAtMostOneMaxExpansionSize | ( | ) |
All at_most_one constraints with a size <= param will be replaced by a quadratic number of binary implications.
optional int32 at_most_one_max_expansion_size = 270 [default = 3];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34536 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasAutoDetectGreaterThanAtLeastOneOf | ( | ) |
If true, then the precedences propagator try to detect for each variable if it has a set of "optional incoming arc" for which at least one of them is present. This is usually useful to have but can be slow on model with a lot of precedence.
optional bool auto_detect_greater_than_at_least_one_of = 95 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33326 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasBinaryMinimizationAlgorithm | ( | ) |
optional .operations_research.sat.SatParameters.BinaryMinizationAlgorithm binary_minimization_algorithm = 34 [default = BINARY_MINIMIZATION_FIRST];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20484 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasBinarySearchNumConflicts | ( | ) |
If non-negative, perform a binary search on the objective variable in order to find an [min, max] interval outside of which the solver proved unsat/sat under this amount of conflict. This can quickly reduce the objective domain on some problems.
optional int32 binary_search_num_conflicts = 99 [default = -1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31438 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasBlockingRestartMultiplier | ( | ) |
optional double blocking_restart_multiplier = 66 [default = 1.4];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21946 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasBlockingRestartWindowSize | ( | ) |
optional int32 blocking_restart_window_size = 65 [default = 5000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21906 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasBooleanEncodingLevel | ( | ) |
A non-negative level indicating how much we should try to fully encode Integer variables as Boolean.
optional int32 boolean_encoding_level = 107 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35436 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasCatchSigintSignal | ( | ) |
Indicates if the CP-SAT layer should catch Control-C (SIGINT) signals when calling solve. If set, catching the SIGINT signal will terminate the search gracefully, as if a time limit was reached.
optional bool catch_sigint_signal = 135 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34597 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasClauseActivityDecay | ( | ) |
Clause activity parameters (same effect as the one on the variables).
optional double clause_activity_decay = 17 [default = 0.999];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21289 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasClauseCleanupLbdBound | ( | ) |
All the clauses with a LBD (literal blocks distance) lower or equal to this parameters will always be kept.
optional int32 clause_cleanup_lbd_bound = 59 [default = 5];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20818 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasClauseCleanupOrdering | ( | ) |
optional .operations_research.sat.SatParameters.ClauseOrdering clause_cleanup_ordering = 60 [default = CLAUSE_ACTIVITY];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20872 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasClauseCleanupPeriod | ( | ) |
Trigger a cleanup when this number of "deletable" clauses is learned.
optional int32 clause_cleanup_period = 11 [default = 10000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20597 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasClauseCleanupProtection | ( | ) |
optional .operations_research.sat.SatParameters.ClauseProtection clause_cleanup_protection = 58 [default = PROTECTION_NONE];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20772 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasClauseCleanupRatio | ( | ) |
During a cleanup, if clause_cleanup_target is 0, we will delete the clause_cleanup_ratio of "deletable" clauses instead of aiming for a fixed target of clauses to keep.
optional double clause_cleanup_ratio = 190 [default = 0.5];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20715 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasClauseCleanupTarget | ( | ) |
During a cleanup, we will always keep that number of "deletable" clauses. Note that this doesn't include the "protected" clauses.
optional int32 clause_cleanup_target = 13 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20654 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasConvertIntervals | ( | ) |
Temporary flag util the feature is more mature. This convert intervals to the newer proto format that support affine start/var/end instead of just variables.
optional bool convert_intervals = 177 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34905 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasCoreMinimizationLevel | ( | ) |
If positive, we spend some effort on each core: - At level 1, we use a simple heuristic to try to minimize an UNSAT core. - At level 2, we use propagation to minimize the core but also identify literal in at most one relationship in this core.
optional int32 core_minimization_level = 50 [default = 2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28014 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasCountAssumptionLevelsInLbd | ( | ) |
Whether or not the assumption levels are taken into account during the LBD computation. According to the reference below, not counting them improves the solver in some situation. Note that this only impact solves under assumptions. Gilles Audemard, Jean-Marie Lagniez, Laurent Simon, "Improving Glucose for Incremental SAT Solving with Assumptions: Application to MUS Extraction" Theory and Applications of Satisfiability Testing - SAT 2013, Lecture Notes in Computer Science Volume 7962, 2013, pp 309-317.
optional bool count_assumption_levels_in_lbd = 49 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23338 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasCoverOptimization | ( | ) |
If true, when the max-sat algo find a core, we compute the minimal number of literals in the core that needs to be true to have a feasible solution. This is also called core exhaustion in more recent max-SAT papers.
optional bool cover_optimization = 89 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28141 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasCpModelPresolve | ( | ) |
Whether we presolve the cp_model before solving it.
optional bool cp_model_presolve = 86 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23982 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasCpModelProbingLevel | ( | ) |
How much effort do we spend on probing. 0 disables it completely.
optional int32 cp_model_probing_level = 110 [default = 2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24038 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasCpModelUseSatPresolve | ( | ) |
Whether we also use the sat presolve when cp_model_presolve is true.
optional bool cp_model_use_sat_presolve = 93 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24094 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasCutActiveCountDecay | ( | ) |
optional double cut_active_count_decay = 156 [default = 0.8];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36735 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasCutCleanupTarget | ( | ) |
Target number of constraints to remove during cleanup.
optional int32 cut_cleanup_target = 157 [default = 1000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36779 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasCutLevel | ( | ) |
Control the global cut effort. Zero will turn off all cut. For now we just have one level. Note also that most cuts are only used at linearization level >= 2.
optional int32 cut_level = 196 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35637 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasCutMaxActiveCountValue | ( | ) |
These parameters are similar to sat clause management activity parameters. They are effective only if the number of generated cuts exceed the storage limit. Default values are based on a few experiments on miplib instances.
optional double cut_max_active_count_value = 155 [default = 10000000000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36677 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasDebugCrashIfPresolveBreaksHint | ( | ) |
Crash if presolve breaks a feasible hint.
optional bool debug_crash_if_presolve_breaks_hint = 306 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27887 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasDebugCrashOnBadHint | ( | ) |
Crash if we do not manage to complete the hint into a full solution.
optional bool debug_crash_on_bad_hint = 195 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27831 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasDebugMaxNumPresolveOperations | ( | ) |
If positive, try to stop just after that many presolve rules have been applied. This is mainly useful for debugging presolve.
optional int32 debug_max_num_presolve_operations = 151 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27772 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasDebugPostsolveWithFullSolver | ( | ) |
We have two different postsolve code. The default one should be better and it allows for a more powerful presolve, but it can be useful to postsolve using the full solver instead.
optional bool debug_postsolve_with_full_solver = 162 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27709 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasDefaultRestartAlgorithms | ( | ) |
optional string default_restart_algorithms = 70 [default = "LUBY_RESTART,LBD_MOVING_AVERAGE_RESTART,DL_MOVING_AVERAGE_RESTART"];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21545 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasDetectLinearizedProduct | ( | ) |
Infer products of Boolean or of Boolean time IntegerVariable from the linear constrainst in the problem. This can be used in some cuts, altough for now we don't really exploit it.
optional bool detect_linearized_product = 277 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37209 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasDetectTableWithCost | ( | ) |
If true, we detect variable that are unique to a table constraint and only there to encode a cost on each tuple. This is usually the case when a WCSP (weighted constraint program) is encoded into CP-SAT format. This can lead to a dramatic speed-up for such problems but is still experimental at this point.
optional bool detect_table_with_cost = 216 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24223 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasDisableConstraintExpansion | ( | ) |
If true, it disable all constraint expansion. This should only be used to test the presolve of expanded constraints.
optional bool disable_constraint_expansion = 181 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24783 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasDiversifyLnsParams | ( | ) |
If true, registers more lns subsolvers with different parameters.
optional bool diversify_lns_params = 137 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34091 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasEncodeComplexLinearConstraintWithInteger | ( | ) |
Linear constraint with a complex right hand side (more than a single interval) need to be expanded, there is a couple of way to do that.
optional bool encode_complex_linear_constraint_with_integer = 223 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24843 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasEncodeCumulativeAsReservoir | ( | ) |
Encore cumulative with fixed demands and capacity as a reservoir constraint. The only reason you might want to do that is to test the reservoir propagation code!
optional bool encode_cumulative_as_reservoir = 287 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24648 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasEnumerateAllSolutions | ( | ) |
Whether we enumerate all solutions of a problem without objective. Note that setting this to true automatically disable some presolve reduction that can remove feasible solution. That is it has the same effect as setting keep_all_feasible_solutions_in_presolve. TODO(user): Do not do that and let the user choose what behavior is best by setting keep_all_feasible_solutions_in_presolve ?
optional bool enumerate_all_solutions = 87 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32933 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasExpandAlldiffConstraints | ( | ) |
If true, expand all_different constraints that are not permutations. Permutations (#Variables = #Values) are always expanded.
optional bool expand_alldiff_constraints = 170 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24363 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasExpandReservoirConstraints | ( | ) |
If true, expand the reservoir constraints by creating booleans for all possible precedences between event and encoding the constraint.
optional bool expand_reservoir_constraints = 182 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24479 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasExpandReservoirUsingCircuit | ( | ) |
Mainly useful for testing. If this and expand_reservoir_constraints is true, we use a different encoding of the reservoir constraint using circuit instead of precedences. Note that this is usually slower, but can exercise different part of the solver. Note that contrary to the precedence encoding, this easily support variable demands. WARNING: with this encoding, the constraint takes a slightly different meaning. There must exist a permutation of the events occurring at the same time such that the level is within the reservoir after each of these events (in this permuted order). So we cannot have +100 and -100 at the same time if the level must be between 0 and 10 (as authorized by the reservoir constraint).
optional bool expand_reservoir_using_circuit = 288 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24551 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasExploitAllLpSolution | ( | ) |
If true and the Lp relaxation of the problem has a solution, try to exploit it. This is same as above except in this case the lp solution might not be an integer solution.
optional bool exploit_all_lp_solution = 116 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36965 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasExploitAllPrecedences | ( | ) |
optional bool exploit_all_precedences = 220 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29135 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasExploitBestSolution | ( | ) |
When branching on a variable, follow the last best solution value.
optional bool exploit_best_solution = 130 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37027 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasExploitIntegerLpSolution | ( | ) |
If true and the Lp relaxation of the problem has an integer optimal solution, try to exploit it. Note that since the LP relaxation may not contain all the constraints, such a solution is not necessarily a solution of the full problem.
optional bool exploit_integer_lp_solution = 94 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36898 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasExploitObjective | ( | ) |
When branching an a variable that directly affect the objective, branch on the value that lead to the best objective first.
optional bool exploit_objective = 131 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37148 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasExploitRelaxationSolution | ( | ) |
When branching on a variable, follow the last best relaxation solution value. We use the relaxation with the tightest bound on the objective as the best relaxation solution.
optional bool exploit_relaxation_solution = 161 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37085 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasFeasibilityJumpBatchDtime | ( | ) |
How much dtime for each LS batch.
optional double feasibility_jump_batch_dtime = 292 [default = 0.1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31879 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasFeasibilityJumpDecay | ( | ) |
On each restart, we randomly choose if we use decay (with this parameter) or no decay.
optional double feasibility_jump_decay = 242 [default = 0.95];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31704 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasFeasibilityJumpEnableRestarts | ( | ) |
When stagnating, feasibility jump will either restart from a default solution (with some possible randomization), or randomly pertubate the current solution. This parameter selects the first option.
optional bool feasibility_jump_enable_restarts = 250 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32057 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasFeasibilityJumpLinearizationLevel | ( | ) |
How much do we linearize the problem in the local search code.
optional int32 feasibility_jump_linearization_level = 257 [default = 2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31763 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasFeasibilityJumpMaxExpandedConstraintSize | ( | ) |
Maximum size of no_overlap or no_overlap_2d constraint for a quadratic expansion. This might look a lot, but by expanding such constraint, we get a linear time evaluation per single variable moves instead of a slow O(n log n) one.
optional int32 feasibility_jump_max_expanded_constraint_size = 264 [default = 500];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32122 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasFeasibilityJumpRestartFactor | ( | ) |
This is a factor that directly influence the work before each restart. Increasing it leads to longer restart.
optional int32 feasibility_jump_restart_factor = 258 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31820 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasFeasibilityJumpVarPerburbationRangeRatio | ( | ) |
Max distance between the default value and the pertubated value relative to the range of the domain of the variable.
optional double feasibility_jump_var_perburbation_range_ratio = 248 [default = 0.2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31996 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasFeasibilityJumpVarRandomizationProbability | ( | ) |
Probability for a variable to have a non default value upon restarts or perturbations.
optional double feasibility_jump_var_randomization_probability = 247 [default = 0.05];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31936 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasFillAdditionalSolutionsInResponse | ( | ) |
If true, the final response addition_solutions field will be filled with all solutions from our solutions pool. Note that if both this field and enumerate_all_solutions is true, we will copy to the pool all of the solution found. So if solution_pool_size is big enough, you can get all solutions this way instead of using the solution callback. Note that this only affect the "final" solution, not the one passed to the solution callbacks.
optional bool fill_additional_solutions_in_response = 194 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33172 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasFillTightenedDomainsInResponse | ( | ) |
If true, add information about the derived variable domains to the CpSolverResponse. It is an option because it makes the response slighly bigger and there is a bit more work involved during the postsolve to construct it, but it should still have a low overhead. See the tightened_variables field in CpSolverResponse for more details.
optional bool fill_tightened_domains_in_response = 132 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33095 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasFilterSatPostsolveClauses | ( | ) |
Internal parameter. During BVE, if we eliminate a variable x, by default we will push all clauses containing x and all clauses containing not(x) to the postsolve. However, it is possible to write the postsolve code so that only one such set is needed. The idea is that, if we push the set containing a literal l, is to set l to false except if it is needed to satisfy one of the clause in the set. This is always beneficial, but for historical reason, not all our postsolve algorithm support this.
optional bool filter_sat_postsolve_clauses = 324 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23488 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasFindBigLinearOverlap | ( | ) |
Try to find large "rectangle" in the linear constraint matrix with identical lines. If such rectangle is big enough, we can introduce a new integer variable corresponding to the common expression and greatly reduce the number of non-zero.
optional bool find_big_linear_overlap = 234 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25381 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasFindMultipleCores | ( | ) |
Whether we try to find more independent cores for a given set of assumptions in the core based max-SAT algorithms.
optional bool find_multiple_cores = 84 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28080 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasFixVariablesToTheirHintedValue | ( | ) |
If true, variables appearing in the solution hints will be fixed to their hinted value.
optional bool fix_variables_to_their_hinted_value = 192 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30544 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasFpRounding | ( | ) |
optional .operations_research.sat.SatParameters.FPRoundingMethod fp_rounding = 165 [default = PROPAGATION_ASSISTED];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34046 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasGlucoseDecayIncrement | ( | ) |
optional double glucose_decay_increment = 23 [default = 0.01];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21205 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasGlucoseDecayIncrementPeriod | ( | ) |
optional int32 glucose_decay_increment_period = 24 [default = 5000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21245 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasGlucoseMaxDecay | ( | ) |
The activity starts at 0.8 and increment by 0.01 every 5000 conflicts until 0.95. This "hack" seems to work well and comes from: Glucose 2.3 in the SAT 2013 Competition - SAT Competition 2013 http://edacc4.informatik.uni-ulm.de/SC13/solver-description-download/136
optional double glucose_max_decay = 22 [default = 0.95];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21141 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasHintConflictLimit | ( | ) |
Conflict limit used in the phase that exploit the solution hint.
optional int32 hint_conflict_limit = 153 [default = 10];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30419 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasIgnoreNames | ( | ) |
If true, we don't keep names in our internal copy of the user given model.
optional bool ignore_names = 202 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25238 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasInferAllDiffs | ( | ) |
Run a max-clique code amongst all the x != y we can find and try to infer set of variables that are all different. This allows to close neos16.mps for instance. Note that we only run this code if there is no all_diff already in the model so that if a user want to add some all_diff, we assume it is well done and do not try to add more. This will also detect and add no_overlap constraints, if all the relations x != y have "offsets" between them. I.e. x > y + offset.
optional bool infer_all_diffs = 233 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25301 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasInitialPolarity | ( | ) |
optional .operations_research.sat.SatParameters.Polarity initial_polarity = 2 [default = POLARITY_FALSE];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 19840 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasInitialVariablesActivity | ( | ) |
The initial value of the variables activity. A non-zero value only make sense when use_erwa_heuristic is true. Experiments with a value of 1e-2 together with the ERWA heuristic showed slighthly better result than simply using zero. The idea is that when the "learning rate" of a variable becomes lower than this value, then we prefer to branch on never explored before variables. This is not in the ERWA paper.
optional double initial_variables_activity = 76 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20310 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasInprocessingDtimeRatio | ( | ) |
Proportion of deterministic time we should spend on inprocessing. At each "restart", if the proportion is below this ratio, we will do some inprocessing, otherwise, we skip it for this restart.
optional double inprocessing_dtime_ratio = 273 [default = 0.2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25508 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasInprocessingMinimizationDtime | ( | ) |
Parameters for an heuristic similar to the one described in "An effective learnt clause minimization approach for CDCL Sat Solvers", https://www.ijcai.org/proceedings/2017/0098.pdf This is the amount of dtime we should spend on this technique during each inprocessing phase. The minimization technique is the same as the one used to minimize core in max-sat. We also minimize problem clauses and not just the learned clause that we keep forever like in the paper.
optional double inprocessing_minimization_dtime = 275 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25635 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasInprocessingMinimizationUseAllOrderings | ( | ) |
optional bool inprocessing_minimization_use_all_orderings = 298 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25754 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasInprocessingMinimizationUseConflictAnalysis | ( | ) |
optional bool inprocessing_minimization_use_conflict_analysis = 297 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25714 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasInprocessingProbingDtime | ( | ) |
The amount of dtime we should spend on probing for each inprocessing round.
optional double inprocessing_probing_dtime = 274 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25570 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasInstantiateAllVariables | ( | ) |
If true, the solver will add a default integer branching strategy to the already defined search strategy. If not, some variable might still not be fixed at the end of the search. For now we assume these variable can just be set to their lower bound.
optional bool instantiate_all_variables = 106 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33258 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasInterleaveBatchSize | ( | ) |
optional int32 interleave_batch_size = 134 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27319 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasInterleaveSearch | ( | ) |
Experimental. If this is true, then we interleave all our major search strategy and distribute the work amongst num_workers. The search is deterministic (independently of num_workers!), and we schedule and wait for interleave_batch_size task to be completed before synchronizing and scheduling the next batch of tasks.
optional bool interleave_search = 136 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27252 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasKeepAllFeasibleSolutionsInPresolve | ( | ) |
If true, we disable the presolve reductions that remove feasible solutions from the search space. Such solution are usually dominated by a "better" solution that is kept, but depending on the situation, we might want to keep all solutions. A trivial example is when a variable is unused. If this is true, then the presolve will not fix it to an arbitrary value and it will stay in the search space.
optional bool keep_all_feasible_solutions_in_presolve = 173 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33014 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasKeepSymmetryInPresolve | ( | ) |
Experimental. This will compute the symmetry of the problem once and for all. All presolve operations we do should keep the symmetry group intact or modify it properly. For now we have really little support for this. We will disable a bunch of presolve operations that could be supported.
optional bool keep_symmetry_in_presolve = 303 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35118 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasLbRelaxNumWorkersThreshold | ( | ) |
Only use lb-relax if we have at least that many workers.
optional int32 lb_relax_num_workers_threshold = 296 [default = 16];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33995 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasLinearizationLevel | ( | ) |
A non-negative level indicating the type of constraints we consider in the LP relaxation. At level zero, no LP relaxation is used. At level 1, only the linear constraint and full encoding are added. At level 2, we also add all the Boolean constraints.
optional int32 linearization_level = 90 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35370 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasLinearSplitSize | ( | ) |
Linear constraints that are not pseudo-Boolean and that are longer than this size will be split into sqrt(size) intermediate sums in order to have faster propation in the CP engine.
optional int32 linear_split_size = 256 [default = 100];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35305 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasLnsInitialDeterministicLimit | ( | ) |
optional double lns_initial_deterministic_limit = 308 [default = 0.1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33599 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasLnsInitialDifficulty | ( | ) |
Initial parameters for neighborhood generation.
optional double lns_initial_difficulty = 307 [default = 0.5];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33547 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasLogPrefix | ( | ) |
Add a prefix to all logs.
optional string log_prefix = 185 [default = ""];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22977 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasLogSearchProgress | ( | ) |
Whether the solver should log the search progress. This is the maing logging parameter and if this is false, none of the logging (callbacks, log_to_stdout, log_to_response, ...) will do anything.
optional bool log_search_progress = 41 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22848 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasLogSubsolverStatistics | ( | ) |
Whether the solver should display per sub-solver search statistics. This is only useful is log_search_progress is set to true, and if the number of search workers is > 1. Note that in all case we display a bit of stats with one line per subsolver.
optional bool log_subsolver_statistics = 189 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22913 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasLogToResponse | ( | ) |
Log to response proto.
optional bool log_to_response = 187 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23138 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasLogToStdout | ( | ) |
Log to stdout.
optional bool log_to_stdout = 186 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23082 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasLpDualTolerance | ( | ) |
optional double lp_dual_tolerance = 267 [default = 1e-07];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34859 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasLpPrimalTolerance | ( | ) |
The internal LP tolerances used by CP-SAT. These applies to the internal and scaled problem. If the domains of your variables are large it might be good to use lower tolerances. If your problem is binary with low coefficients, it might be good to use higher ones to speed-up the lp solves.
optional double lp_primal_tolerance = 266 [default = 1e-07];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34795 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxAllDiffCutSize | ( | ) |
Cut generator for all diffs can add too many cuts for large all_diff constraints. This parameter restricts the large all_diff constraints to have a cut generator.
optional int32 max_all_diff_cut_size = 148 [default = 64];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36137 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxAlldiffDomainSize | ( | ) |
Max domain size for all_different constraints to be expanded.
optional int32 max_alldiff_domain_size = 320 [default = 256];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24422 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxClauseActivityValue | ( | ) |
optional double max_clause_activity_value = 18 [default = 1e+20];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21341 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxConsecutiveInactiveCount | ( | ) |
If a constraint/cut in LP is not active for that many consecutive OPTIMAL solves, remove it from the LP. Note that it might be added again later if it become violated by the current LP solution.
optional int32 max_consecutive_inactive_count = 121 [default = 100];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36613 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxCutRoundsAtLevelZero | ( | ) |
Max number of time we perform cut generation and resolve the LP at level 0.
optional int32 max_cut_rounds_at_level_zero = 154 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36555 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxDeterministicTime | ( | ) |
Maximum time allowed in deterministic time to solve a problem. The deterministic time should be correlated with the real time used by the solver, the time unit being as close as possible to a second.
optional double max_deterministic_time = 67 [default = inf];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22176 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxDomainSizeWhenEncodingEqNeqConstraints | ( | ) |
When loading a*x + b*y ==/!= c when x and y are both fully encoded. The solver may decide to replace the linear equation by a set of clauses. This is triggered if the sizes of the domains of x and y are below the threshold.
optional int32 max_domain_size_when_encoding_eq_neq_constraints = 191 [default = 16];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35498 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaximumRegionsToSplitInDisconnectedNoOverlap2D | ( | ) |
Detects when the space where items of a no_overlap_2d constraint can placed is disjoint (ie., fixed boxes split the domain). When it is the case, we can introduce a boolean for each pair <item, component> encoding whether the item is in the component or not. Then we replace the original no_overlap_2d constraint by one no_overlap_2d constraint for each component, with the new booleans as the enforcement_literal of the intervals. This is equivalent to expanding the original no_overlap_2d constraint into a bin packing problem with each connected component being a bin. This heuristic is only done when the number of regions to split is less than this parameter and <= 1 disables it.
optional int32 maximum_regions_to_split_in_disconnected_no_overlap_2d = 315 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29688 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxIntegerRoundingScaling | ( | ) |
In the integer rounding procedure used for MIR and Gomory cut, the maximum "scaling" we use (must be positive). The lower this is, the lower the integer coefficients of the cut will be. Note that cut generated by lower values are not necessarily worse than cut generated by larger value. There is no strict dominance relationship. Setting this to 2 result in the "strong fractional rouding" of Letchford and Lodi.
optional int32 max_integer_rounding_scaling = 119 [default = 600];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36270 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxLinMaxSizeForExpansion | ( | ) |
If the number of expressions in the lin_max is less that the max size parameter, model expansion replaces target = max(xi) by linear constraint with the introduction of new booleans bi such that bi => target == xi. This is mainly for experimenting compared to a custom lin_max propagator.
optional int32 max_lin_max_size_for_expansion = 280 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24714 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxMemoryInMb | ( | ) |
Maximum memory allowed for the whole thread containing the solver. The solver will abort as soon as it detects that this limit is crossed. As a result, this limit is approximative, but usually the solver will not go too much over. TODO(user): This is only used by the pure SAT solver, generalize to CP-SAT.
optional int64 max_memory_in_mb = 40 [default = 10000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22383 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxNumberOfConflicts | ( | ) |
Maximum number of conflicts allowed to solve a problem. TODO(user): Maybe change the way the conflict limit is enforced? currently it is enforced on each independent internal SAT solve, rather than on the overall number of conflicts across all solves. So in the context of an optimization problem, this is not really usable directly by a client.
optional int64 max_number_of_conflicts = 37 [default = 9223372036854775807];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22304 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxNumCuts | ( | ) |
The limit on the number of cuts in our cut pool. When this is reached we do not generate cuts anymore. TODO(user): We should probably remove this parameters, and just always generate cuts but only keep the best n or something.
optional int32 max_num_cuts = 91 [default = 10000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35567 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxNumDeterministicBatches | ( | ) |
Stops after that number of batches has been scheduled. This only make sense when interleave_search is true.
optional int32 max_num_deterministic_batches = 291 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22239 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxNumIntervalsForTimetableEdgeFinding | ( | ) |
Max number of intervals for the timetable_edge_finding algorithm to propagate. A value of 0 disables the constraint.
optional int32 max_num_intervals_for_timetable_edge_finding = 260 [default = 100];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29004 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxPairsPairwiseReasoningInNoOverlap2D | ( | ) |
If the number of pairs to look is below this threshold, do an extra step of propagation in the no_overlap_2d constraint by looking at all pairs of intervals.
optional int32 max_pairs_pairwise_reasoning_in_no_overlap_2d = 276 [default = 1250];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29617 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxPresolveIterations | ( | ) |
In case of large reduction in a presolve iteration, we perform multiple presolve iterations. This parameter controls the maximum number of such presolve iterations.
optional int32 max_presolve_iterations = 138 [default = 3];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23920 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxSatAssumptionOrder | ( | ) |
optional .operations_research.sat.SatParameters.MaxSatAssumptionOrder max_sat_assumption_order = 51 [default = DEFAULT_ASSUMPTION_ORDER];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28198 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxSatReverseAssumptionOrder | ( | ) |
If true, adds the assumption in the reverse order of the one defined by max_sat_assumption_order.
optional bool max_sat_reverse_assumption_order = 52 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28244 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxSatStratification | ( | ) |
optional .operations_research.sat.SatParameters.MaxSatStratificationAlgorithm max_sat_stratification = 53 [default = STRATIFICATION_DESCENT];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28298 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxSizeToCreatePrecedenceLiteralsInDisjunctive | ( | ) |
Create one literal for each disjunction of two pairs of tasks. This slows down the solve time, but improves the lower bound of the objective in the makespan case. This will be triggered if the number of intervals is less or equal than the parameter and if use_strong_propagation_in_disjunctive is true.
optional int32 max_size_to_create_precedence_literals_in_disjunctive = 229 [default = 60];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28523 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxTimeInSeconds | ( | ) |
Maximum time allowed in seconds to solve a problem. The counter will starts at the beginning of the Solve() call.
optional double max_time_in_seconds = 36 [default = inf];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22115 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMaxVariableActivityValue | ( | ) |
optional double max_variable_activity_value = 16 [default = 1e+100];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21093 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMergeAtMostOneWorkLimit | ( | ) |
optional double merge_at_most_one_work_limit = 146 [default = 100000000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24970 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMergeNoOverlapWorkLimit | ( | ) |
During presolve, we use a maximum clique heuristic to merge together no-overlap constraints or at most one constraints. This code can be slow, so we have a limit in place on the number of explored nodes in the underlying graph. The internal limit is an int64, but we use double here to simplify manual input.
optional double merge_no_overlap_work_limit = 145 [default = 1000000000000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24906 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMinimizationAlgorithm | ( | ) |
optional .operations_research.sat.SatParameters.ConflictMinimizationAlgorithm minimization_algorithm = 4 [default = RECURSIVE];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20444 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMinimizeReductionDuringPbResolution | ( | ) |
A different algorithm during PB resolution. It minimizes the number of calls to ReduceCoefficients() which can be time consuming. However, the search space will be different and if the coefficients are large, this may lead to integer overflows that could otherwise be prevented.
optional bool minimize_reduction_during_pb_resolution = 48 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23265 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMinimizeSharedClauses | ( | ) |
Minimize and detect subsumption of shared clauses immediately after they are imported.
optional bool minimize_shared_clauses = 300 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27592 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMinOrthogonalityForLpConstraints | ( | ) |
While adding constraints, skip the constraints which have orthogonality less than 'min_orthogonality_for_lp_constraints' with already added constraints during current call. Orthogonality is defined as 1 - cosine(vector angle between constraints). A value of zero disable this feature.
optional double min_orthogonality_for_lp_constraints = 115 [default = 0.05];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36487 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMipAutomaticallyScaleVariables | ( | ) |
If true, some continuous variable might be automatically scaled. For now, this is only the case where we detect that a variable is actually an integer multiple of a constant. For instance, variables of the form k * 0.5 are quite frequent, and if we detect this, we will scale such variable domain by 2 to make it implied integer.
optional bool mip_automatically_scale_variables = 166 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37467 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMipCheckPrecision | ( | ) |
As explained in mip_precision and mip_max_activity_exponent, we cannot always reach the wanted precision during scaling. We use this threshold to enphasize in the logs when the precision seems bad.
optional double mip_check_precision = 128 [default = 0.0001];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37841 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMipComputeTrueObjectiveBound | ( | ) |
Even if we make big error when scaling the objective, we can always derive a correct lower bound on the original objective by using the exact lower bound on the scaled integer version of the objective. This should be fast, but if you don't care about having a precise lower bound, you can turn it off.
optional bool mip_compute_true_objective_bound = 198 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37907 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMipDropTolerance | ( | ) |
Any value in the input mip with a magnitude lower than this will be set to zero. This is to avoid some issue in LP presolving.
optional double mip_drop_tolerance = 232 [default = 1e-16];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 38124 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMipMaxActivityExponent | ( | ) |
To avoid integer overflow, we always force the maximum possible constraint activity (and objective value) according to the initial variable domain to be smaller than 2 to this given power. Because of this, we cannot always reach the "mip_wanted_precision" parameter above. This can go as high as 62, but some internal algo currently abort early if they might run into integer overflow, so it is better to keep it a bit lower than this.
optional int32 mip_max_activity_exponent = 127 [default = 53];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37762 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMipMaxBound | ( | ) |
We need to bound the maximum magnitude of the variables for CP-SAT, and that is the bound we use. If the MIP model expect larger variable value in the solution, then the converted model will likely not be relevant.
optional double mip_max_bound = 124 [default = 10000000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37273 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMipMaxValidMagnitude | ( | ) |
Any finite values in the input MIP must be below this threshold, otherwise the model will be reported invalid. This is needed to avoid floating point overflow when evaluating bounds * coeff for instance. We are a bit more defensive, but in practice, users shouldn't use super large values in a MIP.
optional double mip_max_valid_magnitude = 199 [default = 1e+20];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37979 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMipPresolveLevel | ( | ) |
When solving a MIP, we do some basic floating point presolving before scaling the problem to integer to be handled by CP-SAT. This control how much of that presolve we do. It can help to better scale floating point model, but it is not always behaving nicely.
optional int32 mip_presolve_level = 261 [default = 2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 38186 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMipScaleLargeDomain | ( | ) |
If this is false, then mip_var_scaling is only applied to variables with "small" domain. If it is true, we scale all floating point variable independenlty of their domain.
optional bool mip_scale_large_domain = 225 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37401 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMipTreatHighMagnitudeBoundsAsInfinity | ( | ) |
By default, any variable/constraint bound with a finite value and a magnitude greater than the mip_max_valid_magnitude will result with a invalid model. This flags change the behavior such that such bounds are silently transformed to +∞ or -∞. It is recommended to keep it at false, and create valid bounds.
optional bool mip_treat_high_magnitude_bounds_as_infinity = 278 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 38052 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMipVarScaling | ( | ) |
All continuous variable of the problem will be multiplied by this factor. By default, we don't do any variable scaling and rely on the MIP model to specify continuous variable domain with the wanted precision.
optional double mip_var_scaling = 125 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37337 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasMipWantedPrecision | ( | ) |
When scaling constraint with double coefficients to integer coefficients, we will multiply by a power of 2 and round the coefficients. We will choose the lowest power such that we have no potential overflow (see mip_max_activity_exponent) and the worst case constraint activity error does not exceed this threshold. Note that we also detect constraint with rational coefficients and scale them accordingly when it seems better instead of using a power of 2. We also relax all constraint bounds by this absolute value. For pure integer constraint, if this value if lower than one, this will not change anything. However it is needed when scaling MIP problems. If we manage to scale a constraint correctly, the maximum error we can make will be twice this value (once for the scaling error and once for the relaxed bounds). If we are not able to scale that well, we will display that fact but still scale as best as we can.
optional double mip_wanted_precision = 126 [default = 1e-06];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37651 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasName | ( | ) |
In some context, like in a portfolio of search, it makes sense to name a given parameters set for logging purpose.
optional string name = 171 [default = ""];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 19695 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasNewConstraintsBatchSize | ( | ) |
Add that many lazy constraints (or cuts) at once in the LP. Note that at the beginning of the solve, we do add more than this.
optional int32 new_constraints_batch_size = 122 [default = 50];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36836 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasNewLinearPropagation | ( | ) |
The new linear propagation code treat all constraints at once and use an adaptation of Bellman-Ford-Tarjan to propagate constraint in a smarter order and potentially detect propagation cycle earlier.
optional bool new_linear_propagation = 224 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35241 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasNoOverlap2DBooleanRelationsLimit | ( | ) |
If less than this number of boxes are present in a no-overlap 2d, we create 4 Booleans per pair of boxes: - Box 2 is after Box 1 on x. - Box 1 is after Box 2 on x. - Box 2 is after Box 1 on y. - Box 1 is after Box 2 on y. Note that at least one of them must be true, and at most one on x and one on y can be true. This can significantly help in closing small problem. The SAT reasoning can be a lot more powerful when we take decision on such positional relations.
optional int32 no_overlap_2d_boolean_relations_limit = 321 [default = 10];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29283 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasNumConflictsBeforeStrategyChanges | ( | ) |
After each restart, if the number of conflict since the last strategy change is greater that this, then we increment a "strategy_counter" that can be use to change the search strategy used by the following restarts.
optional int32 num_conflicts_before_strategy_changes = 68 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21992 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasNumFullSubsolvers | ( | ) |
We distinguish subsolvers that consume a full thread, and the ones that are always interleaved. If left at zero, we will fix this with a default formula that depends on num_workers. But if you start modifying what runs, you might want to fix that to a given value depending on the num_workers you use.
optional int32 num_full_subsolvers = 294 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25950 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasNumSearchWorkers | ( | ) |
optional int32 num_search_workers = 100 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25902 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasNumViolationLs | ( | ) |
This will create incomplete subsolvers (that are not LNS subsolvers) that use the feasibility jump code to find improving solution, treating the objective improvement as a hard constraint.
optional int32 num_violation_ls = 244 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32189 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasNumWorkers | ( | ) |
Specify the number of parallel workers (i.e. threads) to use during search. This should usually be lower than your number of available cpus + hyperthread in your machine. A value of 0 means the solver will try to use all cores on the machine. A number of 1 means no parallelism. Note that 'num_workers' is the preferred name, but if it is set to zero, we will still read the deprecated 'num_search_workers'. As of 2020-04-10, if you're using SAT via MPSolver (to solve integer programs) this field is overridden with a value of 8, if the field is not set *explicitly*. Thus, always set this field explicitly or via MPSolver::SetNumThreads().
optional int32 num_workers = 206 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25811 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasOnlyAddCutsAtLevelZero | ( | ) |
For the cut that can be generated at any level, this control if we only try to generate them at the root node.
optional bool only_add_cuts_at_level_zero = 92 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35700 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasOnlySolveIp | ( | ) |
If one try to solve a MIP model with CP-SAT, because we assume all variable to be integer after scaling, we will not necessarily have the correct optimal. Note however that all feasible solutions are valid since we will just solve a more restricted version of the original problem. This parameters is here to prevent user to think the solution is optimal when it might not be. One will need to manually set this to false to solve a MIP model where the optimal might be different. Note that this is tested after some MIP presolve steps, so even if not all original variable are integer, we might end up with a pure IP after presolve and after implied integer detection.
optional bool only_solve_ip = 222 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 37546 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasOptimizeWithCore | ( | ) |
The default optimization method is a simple "linear scan", each time trying to find a better solution than the previous one. If this is true, then we use a core-based approach (like in max-SAT) when we try to increase the lower bound instead.
optional bool optimize_with_core = 83 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31222 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasOptimizeWithLbTreeSearch | ( | ) |
Do a more conventional tree search (by opposition to SAT based one) where we keep all the explored node in a tree. This is meant to be used in a portfolio and focus on improving the objective lower bound. Keeping the whole tree allow us to report a better objective lower bound coming from the worst open node in the tree.
optional bool optimize_with_lb_tree_search = 188 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31291 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasOptimizeWithMaxHs | ( | ) |
This has no effect if optimize_with_core is false. If true, use a different core-based algorithm similar to the max-HS algo for max-SAT. This is a hybrid MIP/CP approach and it uses a MIP solver in addition to the CP/SAT one. This is also related to the PhD work of tobyodavies@ "Automatic Logic-Based Benders Decomposition with MiniZinc" http://aaai.org/ocs/index.php/AAAI/AAAI17/paper/view/14489
optional bool optimize_with_max_hs = 85 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31508 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPbCleanupIncrement | ( | ) |
Same as for the clauses, but for the learned pseudo-Boolean constraints.
optional int32 pb_cleanup_increment = 46 [default = 200];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20917 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPbCleanupRatio | ( | ) |
optional double pb_cleanup_ratio = 47 [default = 0.5];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20969 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPermutePresolveConstraintOrder | ( | ) |
optional bool permute_presolve_constraint_order = 179 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22762 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPermuteVariableRandomly | ( | ) |
This is mainly here to test the solver variability. Note that in tests, if not explicitly set to false, all 3 options will be set to true so that clients do not rely on the solver returning a specific solution if they are many equivalent optimal solutions.
optional bool permute_variable_randomly = 178 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22701 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPolarityExploitLsHints | ( | ) |
If true and we have first solution LS workers, tries in some phase to follow a LS solutions that violates has litle constraints as possible.
optional bool polarity_exploit_ls_hints = 309 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20046 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPolarityRephaseIncrement | ( | ) |
If non-zero, then we change the polarity heuristic after that many number of conflicts in an arithmetically increasing fashion. So x the first time, 2 * x the second time, etc...
optional int32 polarity_rephase_increment = 168 [default = 1000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 19983 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPolishLpSolution | ( | ) |
Whether we try to do a few degenerate iteration at the end of an LP solve to minimize the fractionality of the integer variable in the basis. This helps on some problems, but not so much on others. It also cost of bit of time to do such polish step.
optional bool polish_lp_solution = 175 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34726 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPreferredVariableOrder | ( | ) |
optional .operations_research.sat.SatParameters.VariableOrder preferred_variable_order = 1 [default = IN_ORDER];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 19800 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPresolveBlockedClause | ( | ) |
Whether we use an heuristic to detect some basic case of blocked clause in the SAT presolve.
optional bool presolve_blocked_clause = 88 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23735 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPresolveBvaThreshold | ( | ) |
Apply Bounded Variable Addition (BVA) if the number of clauses is reduced by stricly more than this threshold. The algorithm described in the paper uses 0, but quick experiments showed that 1 is a good value. It may not be worth it to add a new variable just to remove one clause.
optional int32 presolve_bva_threshold = 73 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23853 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPresolveBveClauseWeight | ( | ) |
During presolve, we apply BVE only if this weight times the number of clauses plus the number of clause literals is not increased.
optional int32 presolve_bve_clause_weight = 55 [default = 3];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23563 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPresolveBveThreshold | ( | ) |
During presolve, only try to perform the bounded variable elimination (BVE) of a variable x if the number of occurrences of x times the number of occurrences of not(x) is not greater than this parameter.
optional int32 presolve_bve_threshold = 54 [default = 500];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23420 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPresolveExtractIntegerEnforcement | ( | ) |
If true, we will extract from linear constraints, enforcement literals of the form "integer variable at bound => simplified constraint". This should always be beneficial except that we don't always handle them as efficiently as we could for now. This causes problem on manna81.mps (LP relaxation not as tight it seems) and on neos-3354841-apure.mps.gz (too many literals created this way).
optional bool presolve_extract_integer_enforcement = 174 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25087 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPresolveInclusionWorkLimit | ( | ) |
A few presolve operations involve detecting constraints included in other constraint. Since there can be a quadratic number of such pairs, and processing them usually involve scanning them, the complexity of these operations can be big. This enforce a local deterministic limit on the number of entries scanned. Default is 1e8. A value of zero will disable these presolve rules completely.
optional int64 presolve_inclusion_work_limit = 201 [default = 100000000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25164 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPresolveProbingDeterministicTimeLimit | ( | ) |
optional double presolve_probing_deterministic_time_limit = 57 [default = 30];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23690 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPresolveSubstitutionLevel | ( | ) |
How much substitution (also called free variable aggregation in MIP litterature) should we perform at presolve. This currently only concerns variable appearing only in linear constraints. For now the value 0 turns it off and any positive value performs substitution.
optional int32 presolve_substitution_level = 147 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25017 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPresolveUseBva | ( | ) |
Whether or not we use Bounded Variable Addition (BVA) in the presolve.
optional bool presolve_use_bva = 72 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23794 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasProbingDeterministicTimeLimit | ( | ) |
The maximum "deterministic" time limit to spend in probing. A value of zero will disable the probing. TODO(user): Clean up. The first one is used in CP-SAT, the other in pure SAT presolve.
optional double probing_deterministic_time_limit = 226 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23626 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasProbingNumCombinationsLimit | ( | ) |
How many combinations of pairs or triplets of variables we want to scan.
optional int32 probing_num_combinations_limit = 272 [default = 20000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30723 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPropagationLoopDetectionFactor | ( | ) |
Some search decisions might cause a really large number of propagations to happen when integer variables with large domains are only reduced by 1 at each step. If we propagate more than the number of variable times this parameters we try to take counter-measure. Setting this to 0.0 disable this feature. TODO(user): Setting this to something like 10 helps in most cases, but the code is currently buggy and can cause the solve to enter a bad state where no progress is made.
optional double propagation_loop_detection_factor = 221 [default = 10];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28351 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPseudoCostReliabilityThreshold | ( | ) |
The solver ignores the pseudo costs of variables with number of recordings less than this threshold.
optional int64 pseudo_cost_reliability_threshold = 123 [default = 100];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31160 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasPushAllTasksTowardStart | ( | ) |
Experimental code: specify if the objective pushes all tasks toward the start of the schedule.
optional bool push_all_tasks_toward_start = 262 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34268 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRandomBranchesRatio | ( | ) |
A number between 0 and 1 that indicates the proportion of branching variables that are selected randomly instead of choosing the first variable from the given variable_ordering strategy.
optional double random_branches_ratio = 32 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20179 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRandomizeSearch | ( | ) |
Randomize fixed search.
optional bool randomize_search = 103 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34147 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRandomPolarityRatio | ( | ) |
The proportion of polarity chosen at random. Note that this take precedence over the phase saving heuristic. This is different from initial_polarity:POLARITY_RANDOM because it will select a new random polarity each time the variable is branched upon instead of selecting one initially and then always taking this choice.
optional double random_polarity_ratio = 45 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20109 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRandomSeed | ( | ) |
At the beginning of each solve, the random number generator used in some part of the solver is reinitialized to this seed. If you change the random seed, the solver may make different choices during the solving process. For some problems, the running time may vary a lot depending on small change in the solving algorithm. Running the solver with different seeds enables to have more robust benchmarks when evaluating new features.
optional int32 random_seed = 31 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22624 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRelativeGapLimit | ( | ) |
optional double relative_gap_limit = 160 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22574 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRemoveFixedVariablesEarly | ( | ) |
If cp_model_presolve is true and there is a large proportion of fixed variable after the first model copy, remap all the model to a dense set of variable before the full presolve even starts. This should help for LNS on large models.
optional bool remove_fixed_variables_early = 310 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24153 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRepairHint | ( | ) |
If true, the solver tries to repair the solution given in the hint. This search terminates after the 'hint_conflict_limit' is reached and the solver switches to regular search. If false, then we do a FIXED_SEARCH using the hint until the hint_conflict_limit is reached.
optional bool repair_hint = 167 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30478 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRestartDlAverageRatio | ( | ) |
In the moving average restart algorithms, a restart is triggered if the window average times this ratio is greater that the global average.
optional double restart_dl_average_ratio = 63 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21747 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRestartLbdAverageRatio | ( | ) |
optional double restart_lbd_average_ratio = 71 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21802 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRestartPeriod | ( | ) |
Restart period for the FIXED_RESTART strategy. This is also the multiplier used by the LUBY_RESTART strategy.
optional int32 restart_period = 30 [default = 50];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21631 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRestartRunningWindowSize | ( | ) |
Size of the window for the moving average restarts.
optional int32 restart_running_window_size = 62 [default = 50];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21690 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRootLpIterations | ( | ) |
Even at the root node, we do not want to spend too much time on the LP if it is "difficult". So we solve it in "chunks" of that many iterations. The solve will be continued down in the tree or the next time we go back to the root node.
optional int32 root_lp_iterations = 227 [default = 2000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 36418 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRoutingCutDpEffort | ( | ) |
The amount of "effort" to spend in dynamic programming for computing routing cuts. This is in term of basic operations needed by the algorithm in the worst case, so a value like 1e8 should take less than a second to compute.
optional double routing_cut_dp_effort = 314 [default = 10000000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30254 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRoutingCutMaxInfeasiblePathLength | ( | ) |
If the length of an infeasible path is less than this value, a cut will be added to exclude it.
optional int32 routing_cut_max_infeasible_path_length = 317 [default = 6];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30320 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRoutingCutSubsetSizeForBinaryRelationBound | ( | ) |
If the size of a subset of nodes of a RoutesConstraint is less than this value, use linear constraints of size 1 and 2 (such as capacity and time window constraints) enforced by the arc literals to compute cuts for this subset (unless the subset size is less than routing_cut_subset_size_for_tight_binary_relation_bound, in which case the corresponding algorithm is used instead). The algorithm for these cuts has a O(n^3) complexity, where n is the subset size. Hence the value of this parameter should not be too large (e.g. 10 or 20).
optional int32 routing_cut_subset_size_for_binary_relation_bound = 312 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29978 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRoutingCutSubsetSizeForExactBinaryRelationBound | ( | ) |
Similar to above, but with an even stronger algorithm in O(n!). We try to be defensive and abort early or not run that often. Still the value of that parameter shouldn't really be much more than 10.
optional int32 routing_cut_subset_size_for_exact_binary_relation_bound = 316 [default = 8];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30121 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRoutingCutSubsetSizeForShortestPathsBound | ( | ) |
Similar to routing_cut_subset_size_for_exact_binary_relation_bound but use a bound based on shortest path distances (which respect triangular inequality). This allows to derive bounds that are valid for any superset of a given subset. This is slow, so it shouldn't really be larger than 10.
optional int32 routing_cut_subset_size_for_shortest_paths_bound = 318 [default = 8];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30186 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasRoutingCutSubsetSizeForTightBinaryRelationBound | ( | ) |
Similar to above, but with a different algorithm producing better cuts, at the price of a higher O(2^n) complexity, where n is the subset size. Hence the value of this parameter should be small (e.g. less than 10).
optional int32 routing_cut_subset_size_for_tight_binary_relation_bound = 313 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30057 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasSaveLpBasisInLbTreeSearch | ( | ) |
Experimental. Save the current LP basis at each node of the search tree so that when we jump around, we can load it and reduce the number of LP iterations needed. It currently works okay if we do not change the lp with cuts or simplification... More work is needed to make it robust in all cases.
optional bool save_lp_basis_in_lb_tree_search = 284 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31364 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasSearchBranching | ( | ) |
optional .operations_research.sat.SatParameters.SearchBranching search_branching = 82 [default = AUTOMATIC_SEARCH];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30374 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasSearchRandomVariablePoolSize | ( | ) |
Search randomization will collect the top 'search_random_variable_pool_size' valued variables, and pick one randomly. The value of the variable is specific to each strategy.
optional int64 search_random_variable_pool_size = 104 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34205 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasShareBinaryClauses | ( | ) |
Allows sharing of new learned binary clause between workers.
optional bool share_binary_clauses = 203 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27475 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasSharedTreeBalanceTolerance | ( | ) |
How much deeper compared to the ideal max depth of the tree is considered "balanced" enough to still accept a split. Without such a tolerance, sometimes the tree can only be split by a single worker, and they may not generate a split for some time. In contrast, with a tolerance of 1, at least half of all workers should be able to split the tree as soon as a split becomes required. This only has an effect on SPLIT_STRATEGY_BALANCED_TREE and SPLIT_STRATEGY_DISCREPANCY.
optional int32 shared_tree_balance_tolerance = 305 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32853 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasSharedTreeMaxNodesPerWorker | ( | ) |
In order to limit total shared memory and communication overhead, limit the total number of nodes that may be generated in the shared tree. If the shared tree runs out of unassigned leaves, workers act as portfolio workers. Note: this limit includes interior nodes, not just leaves.
optional int32 shared_tree_max_nodes_per_worker = 238 [default = 10000];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32742 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasSharedTreeNumWorkers | ( | ) |
Enables shared tree search. If positive, start this many complete worker threads to explore a shared search tree. These workers communicate objective bounds and simple decision nogoods relating to the shared prefix of the tree, and will avoid exploring the same subtrees as one another. Specifying a negative number uses a heuristic to select an appropriate number of shared tree workeres based on the total number of workers.
optional int32 shared_tree_num_workers = 235 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32373 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasSharedTreeOpenLeavesPerWorker | ( | ) |
How many open leaf nodes should the shared tree maintain per worker.
optional double shared_tree_open_leaves_per_worker = 281 [default = 2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32683 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasSharedTreeSplitStrategy | ( | ) |
optional .operations_research.sat.SatParameters.SharedTreeSplitStrategy shared_tree_split_strategy = 239 [default = SPLIT_STRATEGY_AUTO];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32802 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasSharedTreeWorkerEnablePhaseSharing | ( | ) |
If true, shared tree workers share their target phase when returning an assigned subtree for the next worker to use.
optional bool shared_tree_worker_enable_phase_sharing = 304 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32624 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasSharedTreeWorkerEnableTrailSharing | ( | ) |
If true, workers share more of the information from their local trail. Specifically, literals implied by the shared tree decisions.
optional bool shared_tree_worker_enable_trail_sharing = 295 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32564 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasSharedTreeWorkerMinRestartsPerSubtree | ( | ) |
Minimum restarts before a worker will replace a subtree that looks "bad" based on the average LBD of learned clauses.
optional int32 shared_tree_worker_min_restarts_per_subtree = 282 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32504 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasShareGlueClauses | ( | ) |
Allows sharing of short glue clauses between workers. Implicitly disabled if share_binary_clauses is false.
optional bool share_glue_clauses = 285 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27532 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasShareGlueClausesDtime | ( | ) |
The amount of dtime between each export of shared glue clauses.
optional double share_glue_clauses_dtime = 322 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27651 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasShareLevelZeroBounds | ( | ) |
Allows sharing of the bounds of modified variables at level 0.
optional bool share_level_zero_bounds = 114 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27419 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasShareObjectiveBounds | ( | ) |
Allows objective sharing between workers.
optional bool share_objective_bounds = 113 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27363 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasShavingDeterministicTimeInProbingSearch | ( | ) |
Add a shaving phase (where the solver tries to prove that the lower or upper bound of a variable are infeasible) to the probing search. (<= 0 disables it).
optional double shaving_deterministic_time_in_probing_search = 204 [default = 0.001];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30781 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasShavingSearchDeterministicTime | ( | ) |
Specifies the amount of deterministic time spent of each try at shaving a bound in the shaving search.
optional double shaving_search_deterministic_time = 205 [default = 0.1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30844 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasShavingSearchThreshold | ( | ) |
Specifies the threshold between two modes in the shaving procedure. If the range of the variable/objective is less than this threshold, then the shaving procedure will try to remove values one by one. Otherwise, it will try to remove one range at a time.
optional int64 shaving_search_threshold = 290 [default = 64];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30906 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasSolutionPoolSize | ( | ) |
Size of the top-n different solutions kept by the solver. This parameter must be > 0. Currently this only impact the "base" solution chosen for a LNS fragment.
optional int32 solution_pool_size = 193 [default = 3];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33757 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasStopAfterFirstSolution | ( | ) |
For an optimization problem, stop the solver as soon as we have a solution.
optional bool stop_after_first_solution = 98 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33391 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasStopAfterPresolve | ( | ) |
Mainly used when improving the presolver. When true, stops the solver after the presolve is complete (or after loading and root level propagation).
optional bool stop_after_presolve = 149 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33448 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasStopAfterRootPropagation | ( | ) |
optional bool stop_after_root_propagation = 252 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33503 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasStrategyChangeIncreaseRatio | ( | ) |
The parameter num_conflicts_before_strategy_changes is increased by that much after each strategy change.
optional double strategy_change_increase_ratio = 69 [default = 0];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22055 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasSubsumptionDuringConflictAnalysis | ( | ) |
At a really low cost, during the 1-UIP conflict computation, it is easy to detect if some of the involved reasons are subsumed by the current conflict. When this is true, such clauses are detached and later removed from the problem.
optional bool subsumption_during_conflict_analysis = 56 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20532 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasSymmetryDetectionDeterministicTimeLimit | ( | ) |
Deterministic time limit for symmetry detection.
optional double symmetry_detection_deterministic_time_limit = 302 [default = 1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35183 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasSymmetryLevel | ( | ) |
Whether we try to automatically detect the symmetries in a model and exploit them. Currently, at level 1 we detect them in presolve and try to fix Booleans. At level 2, we also do some form of dynamic symmetry breaking during search. At level 3, we also detect symmetries for very large models, which can be slow. At level 4, we try to break as much symmetry as possible in presolve.
optional int32 symmetry_level = 183 [default = 2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34972 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasTableCompressionLevel | ( | ) |
How much we try to "compress" a table constraint. Compressing more leads to less Booleans and faster propagation but can reduced the quality of the lp relaxation. Values goes from 0 to 3 where we always try to fully compress a table. At 2, we try to automatically decide if it is worth it.
optional int32 table_compression_level = 217 [default = 2];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 24297 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseAbslRandom | ( | ) |
optional bool use_absl_random = 180 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 22802 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseAllDifferentForCircuit | ( | ) |
Turn on extra propagation for the circuit constraint. This can be quite slow.
optional bool use_all_different_for_circuit = 311 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29912 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseAreaEnergeticReasoningInNoOverlap2D | ( | ) |
When this is true, the no_overlap_2d constraint is reinforced with an energetic reasoning that uses an area-based energy. This can be combined with the two other overlap heuristics above.
optional bool use_area_energetic_reasoning_in_no_overlap_2d = 271 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29513 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseBlockingRestart | ( | ) |
Block a moving restart algorithm if the trail size of the current conflict is greater than the multiplier times the moving average of the trail size at the previous conflicts.
optional bool use_blocking_restart = 64 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21848 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseCombinedNoOverlap | ( | ) |
This can be beneficial if there is a lot of no-overlap constraints but a relatively low number of different intervals in the problem. Like 1000 intervals, but 1M intervals in the no-overlap constraints covering them.
optional bool use_combined_no_overlap = 133 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34473 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseConservativeScaleOverloadChecker | ( | ) |
Enable a heuristic to solve cumulative constraints using a modified energy constraint. We modify the usual energy definition by applying a super-additive function (also called "conservative scale" or "dual-feasible function") to the demand and the durations of the tasks. This heuristic is fast but for most problems it does not help much to find a solution.
optional bool use_conservative_scale_overload_checker = 286 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28849 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseDisjunctiveConstraintInCumulative | ( | ) |
When this is true, the cumulative constraint is reinforced with propagators from the disjunctive constraint to improve the inference on a set of tasks that are disjunctive at the root of the problem. This additional level supplements the default level of reasoning. Propagators of the cumulative constraint will not be used at all if all the tasks are disjunctive at root node. This always result in better propagation, but it is usually slow, so depending on the problem, turning this off may lead to a faster solution.
optional bool use_disjunctive_constraint_in_cumulative = 80 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29188 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseDualSchedulingHeuristics | ( | ) |
When set, it activates a few scheduling parameters to improve the lower bound of scheduling problems. This is only effective with multiple workers as it modifies the reduced_cost, lb_tree_search, and probing workers.
optional bool use_dual_scheduling_heuristics = 214 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29849 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseDynamicPrecedenceInCumulative | ( | ) |
optional bool use_dynamic_precedence_in_cumulative = 268 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28719 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseDynamicPrecedenceInDisjunctive | ( | ) |
Whether we try to branch on decision "interval A before interval B" rather than on intervals bounds. This usually works better, but slow down a bit the time to find the first solution. These parameters are still EXPERIMENTAL, the result should be correct, but it some corner cases, they can cause some failing CHECK in the solver.
optional bool use_dynamic_precedence_in_disjunctive = 263 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28652 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseEnergeticReasoningInNoOverlap2D | ( | ) |
When this is true, the no_overlap_2d constraint is reinforced with energetic reasoning. This additional level supplements the default level of reasoning.
optional bool use_energetic_reasoning_in_no_overlap_2d = 213 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29449 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseErwaHeuristic | ( | ) |
Whether we use the ERWA (Exponential Recency Weighted Average) heuristic as described in "Learning Rate Based Branching Heuristic for SAT solvers", J.H.Liang, V. Ganesh, P. Poupart, K.Czarnecki, SAT 2016.
optional bool use_erwa_heuristic = 75 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 20243 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseExactLpReason | ( | ) |
The solver usually exploit the LP relaxation of a model. If this option is true, then whatever is infered by the LP will be used like an heuristic to compute EXACT propagation on the IP. So with this option, there is no numerical imprecision issues.
optional bool use_exact_lp_reason = 109 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34406 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseExtendedProbing | ( | ) |
Use extended probing (probe bool_or, at_most_one, exactly_one).
optional bool use_extended_probing = 269 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30667 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseFeasibilityJump | ( | ) |
Parameters for an heuristic similar to the one described in the paper: "Feasibility Jump: an LP-free Lagrangian MIP heuristic", Bjørnar Luteberget, Giorgio Sartor, 2023, Mathematical Programming Computation.
optional bool use_feasibility_jump = 265 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31581 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseFeasibilityPump | ( | ) |
Adds a feasibility pump subsolver along with lns subsolvers.
optional bool use_feasibility_pump = 164 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33875 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseHardPrecedencesInCumulative | ( | ) |
If true, detect and create constraint for integer variable that are "after" a set of intervals in the same cumulative constraint. Experimental: by default we just use "direct" precedences. If exploit_all_precedences is true, we explore the full precedence graph. This assumes we have a DAG otherwise it fails.
optional bool use_hard_precedences_in_cumulative = 215 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29068 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseImpliedBounds | ( | ) |
Stores and exploits "implied-bounds" in the solver. That is, relations of the form literal => (var >= bound). This is currently used to derive stronger cuts.
optional bool use_implied_bounds = 144 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34661 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseLbRelaxLns | ( | ) |
Turns on neighborhood generator based on local branching LP. Based on Huang et al., "Local Branching Relaxation Heuristics for Integer Linear Programs", 2023.
optional bool use_lb_relax_lns = 255 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33933 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseLinear3ForNoOverlap2DPrecedences | ( | ) |
When set, this activates a propagator for the no_overlap_2d constraint that
uses any eventual linear constraints of the model in the form
`{start interval 1} - {end interval 2} + c*w <= ub` to detect that two
intervals must overlap in one dimension for some values of `w`. This is
particularly useful for problems where the distance between two boxes is
part of the model.
optional bool use_linear3_for_no_overlap_2d_precedences = 323 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29776 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseLns | ( | ) |
Testing parameters used to disable all lns workers.
optional bool use_lns = 283 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33643 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseLnsOnly | ( | ) |
Experimental parameters to disable everything but lns.
optional bool use_lns_only = 101 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33699 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseLsOnly | ( | ) |
Disable every other type of subsolver, setting this turns CP-SAT into a pure local-search solver.
optional bool use_ls_only = 240 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31644 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseObjectiveLbSearch | ( | ) |
If true, search will search in ascending max objective value (when minimizing) starting from the lower bound of the objective.
optional bool use_objective_lb_search = 228 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30972 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseObjectiveShavingSearch | ( | ) |
This search differs from the previous search as it will not use assumptions to bound the objective, and it will recreate a full model with the hardcoded objective value.
optional bool use_objective_shaving_search = 253 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31033 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseOptimizationHints | ( | ) |
For an optimization problem, whether we follow some hints in order to find a better first solution. For a variable with hint, the solver will always try to follow the hint. It will revert to the variable_branching default otherwise.
optional bool use_optimization_hints = 35 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 27946 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseOptionalVariables | ( | ) |
If true, we automatically detect variables whose constraint are always enforced by the same literal and we mark them as optional. This allows to propagate them as if they were present in some situation. TODO(user): This is experimental and seems to lead to wrong optimal in some situation. It should however gives correct solutions. Fix.
optional bool use_optional_variables = 108 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 34332 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseOverloadCheckerInCumulative | ( | ) |
When this is true, the cumulative constraint is reinforced with overload checking, i.e., an additional level of reasoning based on energy. This additional level supplements the default level of reasoning as well as timetable edge finding. This always result in better propagation, but it is usually slow, so depending on the problem, turning this off may lead to a faster solution.
optional bool use_overload_checker_in_cumulative = 78 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28769 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUsePbResolution | ( | ) |
Whether to use pseudo-Boolean resolution to analyze a conflict. Note that this option only make sense if your problem is modelized using pseudo-Boolean constraints. If you only have clauses, this shouldn't change anything (except slow the solver down).
optional bool use_pb_resolution = 43 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 23197 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUsePhaseSaving | ( | ) |
If this is true, then the polarity of a variable will be the last value it was assigned to, or its default polarity if it was never assigned since the call to ResetDecisionHeuristic(). Actually, we use a newer version where we follow the last value in the longest non-conflicting partial assignment in the current phase. This is called 'literal phase saving'. For details see 'A Lightweight Component Caching Scheme for Satisfiability Solvers' K. Pipatsrisawat and A.Darwiche, In 10th International Conference on Theory and Applications of Satisfiability Testing, 2007.
optional bool use_phase_saving = 44 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 19895 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUsePrecedencesInDisjunctiveConstraint | ( | ) |
When this is true, then a disjunctive constraint will try to use the precedence relations between time intervals to propagate their bounds further. For instance if task A and B are both before C and task A and B are in disjunction, then we can deduce that task C must start after duration(A) + duration(B) instead of simply max(duration(A), duration(B)), provided that the start time for all task was currently zero. This always result in better propagation, but it is usually slow, so depending on the problem, turning this off may lead to a faster solution.
optional bool use_precedences_in_disjunctive_constraint = 74 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28439 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseProbingSearch | ( | ) |
If true, search will continuously probe Boolean variables, and integer variable bounds. This parameter is set to true in parallel on the probing worker.
optional bool use_probing_search = 176 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 30605 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseRinsLns | ( | ) |
Turns on relaxation induced neighborhood generator.
optional bool use_rins_lns = 129 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 33819 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseSatInprocessing | ( | ) |
Enable or disable "inprocessing" which is some SAT presolving done at each restart to the root level.
optional bool use_sat_inprocessing = 163 [default = true];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 25447 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseSharedTreeSearch | ( | ) |
Set on shared subtree workers. Users should not set this directly.
optional bool use_shared_tree_search = 236 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32447 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseStrongPropagationInDisjunctive | ( | ) |
Enable stronger and more expensive propagation on no_overlap constraint.
optional bool use_strong_propagation_in_disjunctive = 230 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28591 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseSymmetryInLp | ( | ) |
When we have symmetry, it is possible to "fold" all variables from the same orbit into a single variable, while having the same power of LP relaxation. This can help significantly on symmetric problem. However there is currently a bit of overhead as the rest of the solver need to do some translation between the folded LP and the rest of the problem.
optional bool use_symmetry_in_lp = 301 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 35047 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseTimetableEdgeFindingInCumulative | ( | ) |
When this is true, the cumulative constraint is reinforced with timetable edge finding, i.e., an additional level of reasoning based on the conjunction of energy and mandatory parts. This additional level supplements the default level of reasoning as well as overload_checker. This always result in better propagation, but it is usually slow, so depending on the problem, turning this off may lead to a faster solution.
optional bool use_timetable_edge_finding_in_cumulative = 79 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 28929 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseTimetablingInNoOverlap2D | ( | ) |
When this is true, the no_overlap_2d constraint is reinforced with propagators from the cumulative constraints. It consists of ignoring the position of rectangles in one position and projecting the no_overlap_2d on the other dimension to create a cumulative constraint. This is done on both axis. This additional level supplements the default level of reasoning.
optional bool use_timetabling_in_no_overlap_2d = 200 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29379 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasUseTryEdgeReasoningInNoOverlap2D | ( | ) |
optional bool use_try_edge_reasoning_in_no_overlap_2d = 299 [default = false];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 29571 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasVariableActivityDecay | ( | ) |
Each time a conflict is found, the activities of some variables are increased by one. Then, the activity of all variables are multiplied by variable_activity_decay. To implement this efficiently, the activity of all the variables is not decayed at each conflict. Instead, the activity increment is multiplied by 1 / decay. When an activity reach max_variable_activity_value, all the activity are multiplied by 1 / max_variable_activity_value.
optional double variable_activity_decay = 15 [default = 0.8];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 21020 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasVariablesShavingLevel | ( | ) |
This search takes all Boolean or integer variables, and maximize or minimize them in order to reduce their domain. -1 is automatic, otherwise value 0 disables it, and 1, 2, or 3 changes something.
optional int32 variables_shaving_level = 289 [default = -1];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 31097 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasViolationLsCompoundMoveProbability | ( | ) |
Probability of using compound move search each restart. TODO(user): Add reference to paper when published.
optional double violation_ls_compound_move_probability = 259 [default = 0.5];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32308 of file SatParameters.java.
| boolean com.google.ortools.sat.SatParameters.Builder.hasViolationLsPerturbationPeriod | ( | ) |
How long violation_ls should wait before perturbating a solution.
optional int32 violation_ls_perturbation_period = 249 [default = 100];
Implements com.google.ortools.sat.SatParametersOrBuilder.
Definition at line 32251 of file SatParameters.java.
|
protected |
Definition at line 15713 of file SatParameters.java.
| final boolean com.google.ortools.sat.SatParameters.Builder.isInitialized | ( | ) |
Definition at line 18152 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.mergeFrom | ( | com.google.ortools.sat.SatParameters | other | ) |
Definition at line 17254 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.mergeFrom | ( | com.google.protobuf.CodedInputStream | input, |
| com.google.protobuf.ExtensionRegistryLite | extensionRegistry ) throws java.io.IOException |
Definition at line 18157 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.mergeFrom | ( | com.google.protobuf.Message | other | ) |
Definition at line 17245 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.removeSubsolverParams | ( | int | index | ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Definition at line 27114 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setAbsoluteGapLimit | ( | double | value | ) |
Stop the search when the gap between the best feasible objective (O) and our best objective bound (B) is smaller than a limit. The exact definition is: - Absolute: abs(O - B) - Relative: abs(O - B) / max(1, abs(O)). Important: The relative gap depends on the objective offset! If you artificially shift the objective, you will get widely different value of the relative gap. Note that if the gap is reached, the search status will be OPTIMAL. But one can check the best objective bound to see the actual gap. If the objective is integer, then any absolute gap < 1 will lead to a true optimal. If the objective is floating point, a gap of zero make little sense so is is why we use a non-zero default value. At the end of the search, we will display a warning if OPTIMAL is reported yet the gap is greater than this absolute gap.
optional double absolute_gap_limit = 159 [default = 0.0001];
| value | The absoluteGapLimit to set. |
Definition at line 22529 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setAddCgCuts | ( | boolean | value | ) |
Whether we generate and add Chvatal-Gomory cuts to the LP at root node. Note that for now, this is not heavily tuned.
optional bool add_cg_cuts = 117 [default = true];
| value | The addCgCuts to set. |
Definition at line 35858 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setAddCliqueCuts | ( | boolean | value | ) |
Whether we generate clique cuts from the binary implication graph. Note that as the search goes on, this graph will contains new binary clauses learned by the SAT engine.
optional bool add_clique_cuts = 172 [default = true];
| value | The addCliqueCuts to set. |
Definition at line 36041 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setAddLinMaxCuts | ( | boolean | value | ) |
For the lin max constraints, generates the cuts described in "Strong mixed-integer programming formulations for trained neural networks" by Ross Anderson et. (https://arxiv.org/pdf/1811.01988.pdf)
optional bool add_lin_max_cuts = 152 [default = true];
| value | The addLinMaxCuts to set. |
Definition at line 36229 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setAddLpConstraintsLazily | ( | boolean | value | ) |
If true, we start by an empty LP, and only add constraints not satisfied by the current LP solution batch by batch. A constraint that is only added like this is known as a "lazy" constraint in the literature, except that we currently consider all constraints as lazy here.
optional bool add_lp_constraints_lazily = 112 [default = true];
| value | The addLpConstraintsLazily to set. |
Definition at line 36380 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setAddMirCuts | ( | boolean | value | ) |
Whether we generate MIR cuts at root node. Note that for now, this is not heavily tuned.
optional bool add_mir_cuts = 120 [default = true];
| value | The addMirCuts to set. |
Definition at line 35918 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setAddObjectiveCut | ( | boolean | value | ) |
When the LP objective is fractional, do we add the cut that forces the linear objective expression to be greater or equal to this fractional value rounded up? We can always do that since our objective is integer, and combined with MIR heuristic to reduce the coefficient of such cut, it can help.
optional bool add_objective_cut = 197 [default = false];
| value | The addObjectiveCut to set. |
Definition at line 35795 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setAddRltCuts | ( | boolean | value | ) |
Whether we generate RLT cuts. This is still experimental but can help on binary problem with a lot of clauses of size 3.
optional bool add_rlt_cuts = 279 [default = true];
| value | The addRltCuts to set. |
Definition at line 36102 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setAddZeroHalfCuts | ( | boolean | value | ) |
Whether we generate Zero-Half cuts at root node. Note that for now, this is not heavily tuned.
optional bool add_zero_half_cuts = 169 [default = true];
| value | The addZeroHalfCuts to set. |
Definition at line 35978 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setAlsoBumpVariablesInConflictReasons | ( | boolean | value | ) |
When this is true, then the variables that appear in any of the reason of the variables in a conflict have their activity bumped. This is addition to the variables in the conflict, and the one that were used during conflict resolution.
optional bool also_bump_variables_in_conflict_reasons = 77 [default = false];
| value | The alsoBumpVariablesInConflictReasons to set. |
Definition at line 20414 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setAtMostOneMaxExpansionSize | ( | int | value | ) |
All at_most_one constraints with a size <= param will be replaced by a quadratic number of binary implications.
optional int32 at_most_one_max_expansion_size = 270 [default = 3];
| value | The atMostOneMaxExpansionSize to set. |
Definition at line 34562 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setAutoDetectGreaterThanAtLeastOneOf | ( | boolean | value | ) |
If true, then the precedences propagator try to detect for each variable if it has a set of "optional incoming arc" for which at least one of them is present. This is usually useful to have but can be slow on model with a lot of precedence.
optional bool auto_detect_greater_than_at_least_one_of = 95 [default = true];
| value | The autoDetectGreaterThanAtLeastOneOf to set. |
Definition at line 33356 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setBinaryMinimizationAlgorithm | ( | com.google.ortools.sat.SatParameters.BinaryMinizationAlgorithm | value | ) |
optional .operations_research.sat.SatParameters.BinaryMinizationAlgorithm binary_minimization_algorithm = 34 [default = BINARY_MINIMIZATION_FIRST];
| value | The binaryMinimizationAlgorithm to set. |
Definition at line 20501 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setBinarySearchNumConflicts | ( | int | value | ) |
If non-negative, perform a binary search on the objective variable in order to find an [min, max] interval outside of which the solver proved unsat/sat under this amount of conflict. This can quickly reduce the objective domain on some problems.
optional int32 binary_search_num_conflicts = 99 [default = -1];
| value | The binarySearchNumConflicts to set. |
Definition at line 31468 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setBlockingRestartMultiplier | ( | double | value | ) |
optional double blocking_restart_multiplier = 66 [default = 1.4];
| value | The blockingRestartMultiplier to set. |
Definition at line 21962 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setBlockingRestartWindowSize | ( | int | value | ) |
optional int32 blocking_restart_window_size = 65 [default = 5000];
| value | The blockingRestartWindowSize to set. |
Definition at line 21922 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setBooleanEncodingLevel | ( | int | value | ) |
A non-negative level indicating how much we should try to fully encode Integer variables as Boolean.
optional int32 boolean_encoding_level = 107 [default = 1];
| value | The booleanEncodingLevel to set. |
Definition at line 35462 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setCatchSigintSignal | ( | boolean | value | ) |
Indicates if the CP-SAT layer should catch Control-C (SIGINT) signals when calling solve. If set, catching the SIGINT signal will terminate the search gracefully, as if a time limit was reached.
optional bool catch_sigint_signal = 135 [default = true];
| value | The catchSigintSignal to set. |
Definition at line 34625 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setClauseActivityDecay | ( | double | value | ) |
Clause activity parameters (same effect as the one on the variables).
optional double clause_activity_decay = 17 [default = 0.999];
| value | The clauseActivityDecay to set. |
Definition at line 21313 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setClauseCleanupLbdBound | ( | int | value | ) |
All the clauses with a LBD (literal blocks distance) lower or equal to this parameters will always be kept.
optional int32 clause_cleanup_lbd_bound = 59 [default = 5];
| value | The clauseCleanupLbdBound to set. |
Definition at line 20844 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setClauseCleanupOrdering | ( | com.google.ortools.sat.SatParameters.ClauseOrdering | value | ) |
optional .operations_research.sat.SatParameters.ClauseOrdering clause_cleanup_ordering = 60 [default = CLAUSE_ACTIVITY];
| value | The clauseCleanupOrdering to set. |
Definition at line 20889 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setClauseCleanupPeriod | ( | int | value | ) |
Trigger a cleanup when this number of "deletable" clauses is learned.
optional int32 clause_cleanup_period = 11 [default = 10000];
| value | The clauseCleanupPeriod to set. |
Definition at line 20621 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setClauseCleanupProtection | ( | com.google.ortools.sat.SatParameters.ClauseProtection | value | ) |
optional .operations_research.sat.SatParameters.ClauseProtection clause_cleanup_protection = 58 [default = PROTECTION_NONE];
| value | The clauseCleanupProtection to set. |
Definition at line 20789 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setClauseCleanupRatio | ( | double | value | ) |
During a cleanup, if clause_cleanup_target is 0, we will delete the clause_cleanup_ratio of "deletable" clauses instead of aiming for a fixed target of clauses to keep.
optional double clause_cleanup_ratio = 190 [default = 0.5];
| value | The clauseCleanupRatio to set. |
Definition at line 20743 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setClauseCleanupTarget | ( | int | value | ) |
During a cleanup, we will always keep that number of "deletable" clauses. Note that this doesn't include the "protected" clauses.
optional int32 clause_cleanup_target = 13 [default = 0];
| value | The clauseCleanupTarget to set. |
Definition at line 20680 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setConvertIntervals | ( | boolean | value | ) |
Temporary flag util the feature is more mature. This convert intervals to the newer proto format that support affine start/var/end instead of just variables.
optional bool convert_intervals = 177 [default = true];
| value | The convertIntervals to set. |
Definition at line 34933 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setCoreMinimizationLevel | ( | int | value | ) |
If positive, we spend some effort on each core: - At level 1, we use a simple heuristic to try to minimize an UNSAT core. - At level 2, we use propagation to minimize the core but also identify literal in at most one relationship in this core.
optional int32 core_minimization_level = 50 [default = 2];
| value | The coreMinimizationLevel to set. |
Definition at line 28044 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setCountAssumptionLevelsInLbd | ( | boolean | value | ) |
Whether or not the assumption levels are taken into account during the LBD computation. According to the reference below, not counting them improves the solver in some situation. Note that this only impact solves under assumptions. Gilles Audemard, Jean-Marie Lagniez, Laurent Simon, "Improving Glucose for Incremental SAT Solving with Assumptions: Application to MUS Extraction" Theory and Applications of Satisfiability Testing - SAT 2013, Lecture Notes in Computer Science Volume 7962, 2013, pp 309-317.
optional bool count_assumption_levels_in_lbd = 49 [default = true];
| value | The countAssumptionLevelsInLbd to set. |
Definition at line 23378 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setCoverOptimization | ( | boolean | value | ) |
If true, when the max-sat algo find a core, we compute the minimal number of literals in the core that needs to be true to have a feasible solution. This is also called core exhaustion in more recent max-SAT papers.
optional bool cover_optimization = 89 [default = true];
| value | The coverOptimization to set. |
Definition at line 28169 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setCpModelPresolve | ( | boolean | value | ) |
Whether we presolve the cp_model before solving it.
optional bool cp_model_presolve = 86 [default = true];
| value | The cpModelPresolve to set. |
Definition at line 24006 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setCpModelProbingLevel | ( | int | value | ) |
How much effort do we spend on probing. 0 disables it completely.
optional int32 cp_model_probing_level = 110 [default = 2];
| value | The cpModelProbingLevel to set. |
Definition at line 24062 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setCpModelUseSatPresolve | ( | boolean | value | ) |
Whether we also use the sat presolve when cp_model_presolve is true.
optional bool cp_model_use_sat_presolve = 93 [default = true];
| value | The cpModelUseSatPresolve to set. |
Definition at line 24118 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setCutActiveCountDecay | ( | double | value | ) |
optional double cut_active_count_decay = 156 [default = 0.8];
| value | The cutActiveCountDecay to set. |
Definition at line 36751 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setCutCleanupTarget | ( | int | value | ) |
Target number of constraints to remove during cleanup.
optional int32 cut_cleanup_target = 157 [default = 1000];
| value | The cutCleanupTarget to set. |
Definition at line 36803 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setCutLevel | ( | int | value | ) |
Control the global cut effort. Zero will turn off all cut. For now we just have one level. Note also that most cuts are only used at linearization level >= 2.
optional int32 cut_level = 196 [default = 1];
| value | The cutLevel to set. |
Definition at line 35665 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setCutMaxActiveCountValue | ( | double | value | ) |
These parameters are similar to sat clause management activity parameters. They are effective only if the number of generated cuts exceed the storage limit. Default values are based on a few experiments on miplib instances.
optional double cut_max_active_count_value = 155 [default = 10000000000];
| value | The cutMaxActiveCountValue to set. |
Definition at line 36705 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setDebugCrashIfPresolveBreaksHint | ( | boolean | value | ) |
Crash if presolve breaks a feasible hint.
optional bool debug_crash_if_presolve_breaks_hint = 306 [default = false];
| value | The debugCrashIfPresolveBreaksHint to set. |
Definition at line 27911 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setDebugCrashOnBadHint | ( | boolean | value | ) |
Crash if we do not manage to complete the hint into a full solution.
optional bool debug_crash_on_bad_hint = 195 [default = false];
| value | The debugCrashOnBadHint to set. |
Definition at line 27855 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setDebugMaxNumPresolveOperations | ( | int | value | ) |
If positive, try to stop just after that many presolve rules have been applied. This is mainly useful for debugging presolve.
optional int32 debug_max_num_presolve_operations = 151 [default = 0];
| value | The debugMaxNumPresolveOperations to set. |
Definition at line 27798 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setDebugPostsolveWithFullSolver | ( | boolean | value | ) |
We have two different postsolve code. The default one should be better and it allows for a more powerful presolve, but it can be useful to postsolve using the full solver instead.
optional bool debug_postsolve_with_full_solver = 162 [default = false];
| value | The debugPostsolveWithFullSolver to set. |
Definition at line 27737 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setDefaultRestartAlgorithms | ( | java.lang.String | value | ) |
optional string default_restart_algorithms = 70 [default = "LUBY_RESTART,LBD_MOVING_AVERAGE_RESTART,DL_MOVING_AVERAGE_RESTART"];
| value | The defaultRestartAlgorithms to set. |
Definition at line 21588 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setDefaultRestartAlgorithmsBytes | ( | com.google.protobuf.ByteString | value | ) |
optional string default_restart_algorithms = 70 [default = "LUBY_RESTART,LBD_MOVING_AVERAGE_RESTART,DL_MOVING_AVERAGE_RESTART"];
| value | The bytes for defaultRestartAlgorithms to set. |
Definition at line 21611 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setDetectLinearizedProduct | ( | boolean | value | ) |
Infer products of Boolean or of Boolean time IntegerVariable from the linear constrainst in the problem. This can be used in some cuts, altough for now we don't really exploit it.
optional bool detect_linearized_product = 277 [default = false];
| value | The detectLinearizedProduct to set. |
Definition at line 37237 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setDetectTableWithCost | ( | boolean | value | ) |
If true, we detect variable that are unique to a table constraint and only there to encode a cost on each tuple. This is usually the case when a WCSP (weighted constraint program) is encoded into CP-SAT format. This can lead to a dramatic speed-up for such problems but is still experimental at this point.
optional bool detect_table_with_cost = 216 [default = false];
| value | The detectTableWithCost to set. |
Definition at line 24257 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setDisableConstraintExpansion | ( | boolean | value | ) |
If true, it disable all constraint expansion. This should only be used to test the presolve of expanded constraints.
optional bool disable_constraint_expansion = 181 [default = false];
| value | The disableConstraintExpansion to set. |
Definition at line 24809 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setDiversifyLnsParams | ( | boolean | value | ) |
If true, registers more lns subsolvers with different parameters.
optional bool diversify_lns_params = 137 [default = false];
| value | The diversifyLnsParams to set. |
Definition at line 34115 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setEncodeComplexLinearConstraintWithInteger | ( | boolean | value | ) |
Linear constraint with a complex right hand side (more than a single interval) need to be expanded, there is a couple of way to do that.
optional bool encode_complex_linear_constraint_with_integer = 223 [default = false];
| value | The encodeComplexLinearConstraintWithInteger to set. |
Definition at line 24869 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setEncodeCumulativeAsReservoir | ( | boolean | value | ) |
Encore cumulative with fixed demands and capacity as a reservoir constraint. The only reason you might want to do that is to test the reservoir propagation code!
optional bool encode_cumulative_as_reservoir = 287 [default = false];
| value | The encodeCumulativeAsReservoir to set. |
Definition at line 24676 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setEnumerateAllSolutions | ( | boolean | value | ) |
Whether we enumerate all solutions of a problem without objective. Note that setting this to true automatically disable some presolve reduction that can remove feasible solution. That is it has the same effect as setting keep_all_feasible_solutions_in_presolve. TODO(user): Do not do that and let the user choose what behavior is best by setting keep_all_feasible_solutions_in_presolve ?
optional bool enumerate_all_solutions = 87 [default = false];
| value | The enumerateAllSolutions to set. |
Definition at line 32969 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setExpandAlldiffConstraints | ( | boolean | value | ) |
If true, expand all_different constraints that are not permutations. Permutations (#Variables = #Values) are always expanded.
optional bool expand_alldiff_constraints = 170 [default = false];
| value | The expandAlldiffConstraints to set. |
Definition at line 24389 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setExpandReservoirConstraints | ( | boolean | value | ) |
If true, expand the reservoir constraints by creating booleans for all possible precedences between event and encoding the constraint.
optional bool expand_reservoir_constraints = 182 [default = true];
| value | The expandReservoirConstraints to set. |
Definition at line 24505 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setExpandReservoirUsingCircuit | ( | boolean | value | ) |
Mainly useful for testing. If this and expand_reservoir_constraints is true, we use a different encoding of the reservoir constraint using circuit instead of precedences. Note that this is usually slower, but can exercise different part of the solver. Note that contrary to the precedence encoding, this easily support variable demands. WARNING: with this encoding, the constraint takes a slightly different meaning. There must exist a permutation of the events occurring at the same time such that the level is within the reservoir after each of these events (in this permuted order). So we cannot have +100 and -100 at the same time if the level must be between 0 and 10 (as authorized by the reservoir constraint).
optional bool expand_reservoir_using_circuit = 288 [default = false];
| value | The expandReservoirUsingCircuit to set. |
Definition at line 24601 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setExploitAllLpSolution | ( | boolean | value | ) |
If true and the Lp relaxation of the problem has a solution, try to exploit it. This is same as above except in this case the lp solution might not be an integer solution.
optional bool exploit_all_lp_solution = 116 [default = true];
| value | The exploitAllLpSolution to set. |
Definition at line 36993 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setExploitAllPrecedences | ( | boolean | value | ) |
optional bool exploit_all_precedences = 220 [default = false];
| value | The exploitAllPrecedences to set. |
Definition at line 29151 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setExploitBestSolution | ( | boolean | value | ) |
When branching on a variable, follow the last best solution value.
optional bool exploit_best_solution = 130 [default = false];
| value | The exploitBestSolution to set. |
Definition at line 37051 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setExploitIntegerLpSolution | ( | boolean | value | ) |
If true and the Lp relaxation of the problem has an integer optimal solution, try to exploit it. Note that since the LP relaxation may not contain all the constraints, such a solution is not necessarily a solution of the full problem.
optional bool exploit_integer_lp_solution = 94 [default = true];
| value | The exploitIntegerLpSolution to set. |
Definition at line 36928 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setExploitObjective | ( | boolean | value | ) |
When branching an a variable that directly affect the objective, branch on the value that lead to the best objective first.
optional bool exploit_objective = 131 [default = true];
| value | The exploitObjective to set. |
Definition at line 37174 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setExploitRelaxationSolution | ( | boolean | value | ) |
When branching on a variable, follow the last best relaxation solution value. We use the relaxation with the tightest bound on the objective as the best relaxation solution.
optional bool exploit_relaxation_solution = 161 [default = false];
| value | The exploitRelaxationSolution to set. |
Definition at line 37113 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setExtraSubsolvers | ( | int | index, |
| java.lang.String | value ) |
A convenient way to add more workers types. These will be added at the beginning of the list.
repeated string extra_subsolvers = 219;
| index | The index to set the value at. |
| value | The extraSubsolvers to set. |
Definition at line 26442 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setFeasibilityJumpBatchDtime | ( | double | value | ) |
How much dtime for each LS batch.
optional double feasibility_jump_batch_dtime = 292 [default = 0.1];
| value | The feasibilityJumpBatchDtime to set. |
Definition at line 31903 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setFeasibilityJumpDecay | ( | double | value | ) |
On each restart, we randomly choose if we use decay (with this parameter) or no decay.
optional double feasibility_jump_decay = 242 [default = 0.95];
| value | The feasibilityJumpDecay to set. |
Definition at line 31730 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setFeasibilityJumpEnableRestarts | ( | boolean | value | ) |
When stagnating, feasibility jump will either restart from a default solution (with some possible randomization), or randomly pertubate the current solution. This parameter selects the first option.
optional bool feasibility_jump_enable_restarts = 250 [default = true];
| value | The feasibilityJumpEnableRestarts to set. |
Definition at line 32085 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setFeasibilityJumpLinearizationLevel | ( | int | value | ) |
How much do we linearize the problem in the local search code.
optional int32 feasibility_jump_linearization_level = 257 [default = 2];
| value | The feasibilityJumpLinearizationLevel to set. |
Definition at line 31787 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setFeasibilityJumpMaxExpandedConstraintSize | ( | int | value | ) |
Maximum size of no_overlap or no_overlap_2d constraint for a quadratic expansion. This might look a lot, but by expanding such constraint, we get a linear time evaluation per single variable moves instead of a slow O(n log n) one.
optional int32 feasibility_jump_max_expanded_constraint_size = 264 [default = 500];
| value | The feasibilityJumpMaxExpandedConstraintSize to set. |
Definition at line 32152 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setFeasibilityJumpRestartFactor | ( | int | value | ) |
This is a factor that directly influence the work before each restart. Increasing it leads to longer restart.
optional int32 feasibility_jump_restart_factor = 258 [default = 1];
| value | The feasibilityJumpRestartFactor to set. |
Definition at line 31846 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setFeasibilityJumpVarPerburbationRangeRatio | ( | double | value | ) |
Max distance between the default value and the pertubated value relative to the range of the domain of the variable.
optional double feasibility_jump_var_perburbation_range_ratio = 248 [default = 0.2];
| value | The feasibilityJumpVarPerburbationRangeRatio to set. |
Definition at line 32022 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setFeasibilityJumpVarRandomizationProbability | ( | double | value | ) |
Probability for a variable to have a non default value upon restarts or perturbations.
optional double feasibility_jump_var_randomization_probability = 247 [default = 0.05];
| value | The feasibilityJumpVarRandomizationProbability to set. |
Definition at line 31962 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setFillAdditionalSolutionsInResponse | ( | boolean | value | ) |
If true, the final response addition_solutions field will be filled with all solutions from our solutions pool. Note that if both this field and enumerate_all_solutions is true, we will copy to the pool all of the solution found. So if solution_pool_size is big enough, you can get all solutions this way instead of using the solution callback. Note that this only affect the "final" solution, not the one passed to the solution callbacks.
optional bool fill_additional_solutions_in_response = 194 [default = false];
| value | The fillAdditionalSolutionsInResponse to set. |
Definition at line 33214 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setFillTightenedDomainsInResponse | ( | boolean | value | ) |
If true, add information about the derived variable domains to the CpSolverResponse. It is an option because it makes the response slighly bigger and there is a bit more work involved during the postsolve to construct it, but it should still have a low overhead. See the tightened_variables field in CpSolverResponse for more details.
optional bool fill_tightened_domains_in_response = 132 [default = false];
| value | The fillTightenedDomainsInResponse to set. |
Definition at line 33127 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setFilterSatPostsolveClauses | ( | boolean | value | ) |
Internal parameter. During BVE, if we eliminate a variable x, by default we will push all clauses containing x and all clauses containing not(x) to the postsolve. However, it is possible to write the postsolve code so that only one such set is needed. The idea is that, if we push the set containing a literal l, is to set l to false except if it is needed to satisfy one of the clause in the set. This is always beneficial, but for historical reason, not all our postsolve algorithm support this.
optional bool filter_sat_postsolve_clauses = 324 [default = false];
| value | The filterSatPostsolveClauses to set. |
Definition at line 23524 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setFilterSubsolvers | ( | int | index, |
| java.lang.String | value ) |
repeated string filter_subsolvers = 293;
| index | The index to set the value at. |
| value | The filterSubsolvers to set. |
Definition at line 26790 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setFindBigLinearOverlap | ( | boolean | value | ) |
Try to find large "rectangle" in the linear constraint matrix with identical lines. If such rectangle is big enough, we can introduce a new integer variable corresponding to the common expression and greatly reduce the number of non-zero.
optional bool find_big_linear_overlap = 234 [default = true];
| value | The findBigLinearOverlap to set. |
Definition at line 25411 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setFindMultipleCores | ( | boolean | value | ) |
Whether we try to find more independent cores for a given set of assumptions in the core based max-SAT algorithms.
optional bool find_multiple_cores = 84 [default = true];
| value | The findMultipleCores to set. |
Definition at line 28106 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setFixVariablesToTheirHintedValue | ( | boolean | value | ) |
If true, variables appearing in the solution hints will be fixed to their hinted value.
optional bool fix_variables_to_their_hinted_value = 192 [default = false];
| value | The fixVariablesToTheirHintedValue to set. |
Definition at line 30570 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setFpRounding | ( | com.google.ortools.sat.SatParameters.FPRoundingMethod | value | ) |
optional .operations_research.sat.SatParameters.FPRoundingMethod fp_rounding = 165 [default = PROPAGATION_ASSISTED];
| value | The fpRounding to set. |
Definition at line 34063 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setGlucoseDecayIncrement | ( | double | value | ) |
optional double glucose_decay_increment = 23 [default = 0.01];
| value | The glucoseDecayIncrement to set. |
Definition at line 21221 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setGlucoseDecayIncrementPeriod | ( | int | value | ) |
optional int32 glucose_decay_increment_period = 24 [default = 5000];
| value | The glucoseDecayIncrementPeriod to set. |
Definition at line 21261 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setGlucoseMaxDecay | ( | double | value | ) |
The activity starts at 0.8 and increment by 0.01 every 5000 conflicts until 0.95. This "hack" seems to work well and comes from: Glucose 2.3 in the SAT 2013 Competition - SAT Competition 2013 http://edacc4.informatik.uni-ulm.de/SC13/solver-description-download/136
optional double glucose_max_decay = 22 [default = 0.95];
| value | The glucoseMaxDecay to set. |
Definition at line 21173 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setHintConflictLimit | ( | int | value | ) |
Conflict limit used in the phase that exploit the solution hint.
optional int32 hint_conflict_limit = 153 [default = 10];
| value | The hintConflictLimit to set. |
Definition at line 30443 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setIgnoreNames | ( | boolean | value | ) |
If true, we don't keep names in our internal copy of the user given model.
optional bool ignore_names = 202 [default = true];
| value | The ignoreNames to set. |
Definition at line 25262 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setIgnoreSubsolvers | ( | int | index, |
| java.lang.String | value ) |
Rather than fully specifying subsolvers, it is often convenient to just remove the ones that are not useful on a given problem or only keep specific ones for testing. Each string is interpreted as a "glob", so we support '*' and '?'. The way this work is that we will only accept a name that match a filter pattern (if non-empty) and do not match an ignore pattern. Note also that these fields work on LNS or LS names even if these are currently not specified via the subsolvers field.
repeated string ignore_subsolvers = 209;
| index | The index to set the value at. |
| value | The ignoreSubsolvers to set. |
Definition at line 26632 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setInferAllDiffs | ( | boolean | value | ) |
Run a max-clique code amongst all the x != y we can find and try to infer set of variables that are all different. This allows to close neos16.mps for instance. Note that we only run this code if there is no all_diff already in the model so that if a user want to add some all_diff, we assume it is well done and do not try to add more. This will also detect and add no_overlap constraints, if all the relations x != y have "offsets" between them. I.e. x > y + offset.
optional bool infer_all_diffs = 233 [default = true];
| value | The inferAllDiffs to set. |
Definition at line 25339 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setInitialPolarity | ( | com.google.ortools.sat.SatParameters.Polarity | value | ) |
optional .operations_research.sat.SatParameters.Polarity initial_polarity = 2 [default = POLARITY_FALSE];
| value | The initialPolarity to set. |
Definition at line 19857 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setInitialVariablesActivity | ( | double | value | ) |
The initial value of the variables activity. A non-zero value only make sense when use_erwa_heuristic is true. Experiments with a value of 1e-2 together with the ERWA heuristic showed slighthly better result than simply using zero. The idea is that when the "learning rate" of a variable becomes lower than this value, then we prefer to branch on never explored before variables. This is not in the ERWA paper.
optional double initial_variables_activity = 76 [default = 0];
| value | The initialVariablesActivity to set. |
Definition at line 20344 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setInprocessingDtimeRatio | ( | double | value | ) |
Proportion of deterministic time we should spend on inprocessing. At each "restart", if the proportion is below this ratio, we will do some inprocessing, otherwise, we skip it for this restart.
optional double inprocessing_dtime_ratio = 273 [default = 0.2];
| value | The inprocessingDtimeRatio to set. |
Definition at line 25536 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setInprocessingMinimizationDtime | ( | double | value | ) |
Parameters for an heuristic similar to the one described in "An effective learnt clause minimization approach for CDCL Sat Solvers", https://www.ijcai.org/proceedings/2017/0098.pdf This is the amount of dtime we should spend on this technique during each inprocessing phase. The minimization technique is the same as the one used to minimize core in max-sat. We also minimize problem clauses and not just the learned clause that we keep forever like in the paper.
optional double inprocessing_minimization_dtime = 275 [default = 1];
| value | The inprocessingMinimizationDtime to set. |
Definition at line 25677 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setInprocessingMinimizationUseAllOrderings | ( | boolean | value | ) |
optional bool inprocessing_minimization_use_all_orderings = 298 [default = false];
| value | The inprocessingMinimizationUseAllOrderings to set. |
Definition at line 25770 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setInprocessingMinimizationUseConflictAnalysis | ( | boolean | value | ) |
optional bool inprocessing_minimization_use_conflict_analysis = 297 [default = true];
| value | The inprocessingMinimizationUseConflictAnalysis to set. |
Definition at line 25730 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setInprocessingProbingDtime | ( | double | value | ) |
The amount of dtime we should spend on probing for each inprocessing round.
optional double inprocessing_probing_dtime = 274 [default = 1];
| value | The inprocessingProbingDtime to set. |
Definition at line 25594 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setInstantiateAllVariables | ( | boolean | value | ) |
If true, the solver will add a default integer branching strategy to the already defined search strategy. If not, some variable might still not be fixed at the end of the search. For now we assume these variable can just be set to their lower bound.
optional bool instantiate_all_variables = 106 [default = true];
| value | The instantiateAllVariables to set. |
Definition at line 33288 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setInterleaveBatchSize | ( | int | value | ) |
optional int32 interleave_batch_size = 134 [default = 0];
| value | The interleaveBatchSize to set. |
Definition at line 27335 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setInterleaveSearch | ( | boolean | value | ) |
Experimental. If this is true, then we interleave all our major search strategy and distribute the work amongst num_workers. The search is deterministic (independently of num_workers!), and we schedule and wait for interleave_batch_size task to be completed before synchronizing and scheduling the next batch of tasks.
optional bool interleave_search = 136 [default = false];
| value | The interleaveSearch to set. |
Definition at line 27286 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setKeepAllFeasibleSolutionsInPresolve | ( | boolean | value | ) |
If true, we disable the presolve reductions that remove feasible solutions from the search space. Such solution are usually dominated by a "better" solution that is kept, but depending on the situation, we might want to keep all solutions. A trivial example is when a variable is unused. If this is true, then the presolve will not fix it to an arbitrary value and it will stay in the search space.
optional bool keep_all_feasible_solutions_in_presolve = 173 [default = false];
| value | The keepAllFeasibleSolutionsInPresolve to set. |
Definition at line 33052 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setKeepSymmetryInPresolve | ( | boolean | value | ) |
Experimental. This will compute the symmetry of the problem once and for all. All presolve operations we do should keep the symmetry group intact or modify it properly. For now we have really little support for this. We will disable a bunch of presolve operations that could be supported.
optional bool keep_symmetry_in_presolve = 303 [default = false];
| value | The keepSymmetryInPresolve to set. |
Definition at line 35148 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setLbRelaxNumWorkersThreshold | ( | int | value | ) |
Only use lb-relax if we have at least that many workers.
optional int32 lb_relax_num_workers_threshold = 296 [default = 16];
| value | The lbRelaxNumWorkersThreshold to set. |
Definition at line 34019 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setLinearizationLevel | ( | int | value | ) |
A non-negative level indicating the type of constraints we consider in the LP relaxation. At level zero, no LP relaxation is used. At level 1, only the linear constraint and full encoding are added. At level 2, we also add all the Boolean constraints.
optional int32 linearization_level = 90 [default = 1];
| value | The linearizationLevel to set. |
Definition at line 35400 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setLinearSplitSize | ( | int | value | ) |
Linear constraints that are not pseudo-Boolean and that are longer than this size will be split into sqrt(size) intermediate sums in order to have faster propation in the CP engine.
optional int32 linear_split_size = 256 [default = 100];
| value | The linearSplitSize to set. |
Definition at line 35333 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setLnsInitialDeterministicLimit | ( | double | value | ) |
optional double lns_initial_deterministic_limit = 308 [default = 0.1];
| value | The lnsInitialDeterministicLimit to set. |
Definition at line 33615 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setLnsInitialDifficulty | ( | double | value | ) |
Initial parameters for neighborhood generation.
optional double lns_initial_difficulty = 307 [default = 0.5];
| value | The lnsInitialDifficulty to set. |
Definition at line 33571 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setLogPrefix | ( | java.lang.String | value | ) |
Add a prefix to all logs.
optional string log_prefix = 185 [default = ""];
| value | The logPrefix to set. |
Definition at line 23032 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setLogPrefixBytes | ( | com.google.protobuf.ByteString | value | ) |
Add a prefix to all logs.
optional string log_prefix = 185 [default = ""];
| value | The bytes for logPrefix to set. |
Definition at line 23063 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setLogSearchProgress | ( | boolean | value | ) |
Whether the solver should log the search progress. This is the maing logging parameter and if this is false, none of the logging (callbacks, log_to_stdout, log_to_response, ...) will do anything.
optional bool log_search_progress = 41 [default = false];
| value | The logSearchProgress to set. |
Definition at line 22876 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setLogSubsolverStatistics | ( | boolean | value | ) |
Whether the solver should display per sub-solver search statistics. This is only useful is log_search_progress is set to true, and if the number of search workers is > 1. Note that in all case we display a bit of stats with one line per subsolver.
optional bool log_subsolver_statistics = 189 [default = false];
| value | The logSubsolverStatistics to set. |
Definition at line 22943 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setLogToResponse | ( | boolean | value | ) |
Log to response proto.
optional bool log_to_response = 187 [default = false];
| value | The logToResponse to set. |
Definition at line 23162 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setLogToStdout | ( | boolean | value | ) |
Log to stdout.
optional bool log_to_stdout = 186 [default = true];
| value | The logToStdout to set. |
Definition at line 23106 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setLpDualTolerance | ( | double | value | ) |
optional double lp_dual_tolerance = 267 [default = 1e-07];
| value | The lpDualTolerance to set. |
Definition at line 34875 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setLpPrimalTolerance | ( | double | value | ) |
The internal LP tolerances used by CP-SAT. These applies to the internal and scaled problem. If the domains of your variables are large it might be good to use lower tolerances. If your problem is binary with low coefficients, it might be good to use higher ones to speed-up the lp solves.
optional double lp_primal_tolerance = 266 [default = 1e-07];
| value | The lpPrimalTolerance to set. |
Definition at line 34827 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxAllDiffCutSize | ( | int | value | ) |
Cut generator for all diffs can add too many cuts for large all_diff constraints. This parameter restricts the large all_diff constraints to have a cut generator.
optional int32 max_all_diff_cut_size = 148 [default = 64];
| value | The maxAllDiffCutSize to set. |
Definition at line 36165 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxAlldiffDomainSize | ( | int | value | ) |
Max domain size for all_different constraints to be expanded.
optional int32 max_alldiff_domain_size = 320 [default = 256];
| value | The maxAlldiffDomainSize to set. |
Definition at line 24446 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxClauseActivityValue | ( | double | value | ) |
optional double max_clause_activity_value = 18 [default = 1e+20];
| value | The maxClauseActivityValue to set. |
Definition at line 21357 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxConsecutiveInactiveCount | ( | int | value | ) |
If a constraint/cut in LP is not active for that many consecutive OPTIMAL solves, remove it from the LP. Note that it might be added again later if it become violated by the current LP solution.
optional int32 max_consecutive_inactive_count = 121 [default = 100];
| value | The maxConsecutiveInactiveCount to set. |
Definition at line 36641 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxCutRoundsAtLevelZero | ( | int | value | ) |
Max number of time we perform cut generation and resolve the LP at level 0.
optional int32 max_cut_rounds_at_level_zero = 154 [default = 1];
| value | The maxCutRoundsAtLevelZero to set. |
Definition at line 36579 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxDeterministicTime | ( | double | value | ) |
Maximum time allowed in deterministic time to solve a problem. The deterministic time should be correlated with the real time used by the solver, the time unit being as close as possible to a second.
optional double max_deterministic_time = 67 [default = inf];
| value | The maxDeterministicTime to set. |
Definition at line 22204 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxDomainSizeWhenEncodingEqNeqConstraints | ( | int | value | ) |
When loading a*x + b*y ==/!= c when x and y are both fully encoded. The solver may decide to replace the linear equation by a set of clauses. This is triggered if the sizes of the domains of x and y are below the threshold.
optional int32 max_domain_size_when_encoding_eq_neq_constraints = 191 [default = 16];
| value | The maxDomainSizeWhenEncodingEqNeqConstraints to set. |
Definition at line 35528 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaximumRegionsToSplitInDisconnectedNoOverlap2D | ( | int | value | ) |
Detects when the space where items of a no_overlap_2d constraint can placed is disjoint (ie., fixed boxes split the domain). When it is the case, we can introduce a boolean for each pair <item, component> encoding whether the item is in the component or not. Then we replace the original no_overlap_2d constraint by one no_overlap_2d constraint for each component, with the new booleans as the enforcement_literal of the intervals. This is equivalent to expanding the original no_overlap_2d constraint into a bin packing problem with each connected component being a bin. This heuristic is only done when the number of regions to split is less than this parameter and <= 1 disables it.
optional int32 maximum_regions_to_split_in_disconnected_no_overlap_2d = 315 [default = 0];
| value | The maximumRegionsToSplitInDisconnectedNoOverlap2d to set. |
Definition at line 29730 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxIntegerRoundingScaling | ( | int | value | ) |
In the integer rounding procedure used for MIR and Gomory cut, the maximum "scaling" we use (must be positive). The lower this is, the lower the integer coefficients of the cut will be. Note that cut generated by lower values are not necessarily worse than cut generated by larger value. There is no strict dominance relationship. Setting this to 2 result in the "strong fractional rouding" of Letchford and Lodi.
optional int32 max_integer_rounding_scaling = 119 [default = 600];
| value | The maxIntegerRoundingScaling to set. |
Definition at line 36308 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxLinMaxSizeForExpansion | ( | int | value | ) |
If the number of expressions in the lin_max is less that the max size parameter, model expansion replaces target = max(xi) by linear constraint with the introduction of new booleans bi such that bi => target == xi. This is mainly for experimenting compared to a custom lin_max propagator.
optional int32 max_lin_max_size_for_expansion = 280 [default = 0];
| value | The maxLinMaxSizeForExpansion to set. |
Definition at line 24746 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxMemoryInMb | ( | long | value | ) |
Maximum memory allowed for the whole thread containing the solver. The solver will abort as soon as it detects that this limit is crossed. As a result, this limit is approximative, but usually the solver will not go too much over. TODO(user): This is only used by the pure SAT solver, generalize to CP-SAT.
optional int64 max_memory_in_mb = 40 [default = 10000];
| value | The maxMemoryInMb to set. |
Definition at line 22417 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxNumberOfConflicts | ( | long | value | ) |
Maximum number of conflicts allowed to solve a problem. TODO(user): Maybe change the way the conflict limit is enforced? currently it is enforced on each independent internal SAT solve, rather than on the overall number of conflicts across all solves. So in the context of an optimization problem, this is not really usable directly by a client.
optional int64 max_number_of_conflicts = 37 [default = 9223372036854775807];
| value | The maxNumberOfConflicts to set. |
Definition at line 22340 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxNumCuts | ( | int | value | ) |
The limit on the number of cuts in our cut pool. When this is reached we do not generate cuts anymore. TODO(user): We should probably remove this parameters, and just always generate cuts but only keep the best n or something.
optional int32 max_num_cuts = 91 [default = 10000];
| value | The maxNumCuts to set. |
Definition at line 35599 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxNumDeterministicBatches | ( | int | value | ) |
Stops after that number of batches has been scheduled. This only make sense when interleave_search is true.
optional int32 max_num_deterministic_batches = 291 [default = 0];
| value | The maxNumDeterministicBatches to set. |
Definition at line 22265 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxNumIntervalsForTimetableEdgeFinding | ( | int | value | ) |
Max number of intervals for the timetable_edge_finding algorithm to propagate. A value of 0 disables the constraint.
optional int32 max_num_intervals_for_timetable_edge_finding = 260 [default = 100];
| value | The maxNumIntervalsForTimetableEdgeFinding to set. |
Definition at line 29030 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxPairsPairwiseReasoningInNoOverlap2D | ( | int | value | ) |
If the number of pairs to look is below this threshold, do an extra step of propagation in the no_overlap_2d constraint by looking at all pairs of intervals.
optional int32 max_pairs_pairwise_reasoning_in_no_overlap_2d = 276 [default = 1250];
| value | The maxPairsPairwiseReasoningInNoOverlap2d to set. |
Definition at line 29645 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxPresolveIterations | ( | int | value | ) |
In case of large reduction in a presolve iteration, we perform multiple presolve iterations. This parameter controls the maximum number of such presolve iterations.
optional int32 max_presolve_iterations = 138 [default = 3];
| value | The maxPresolveIterations to set. |
Definition at line 23948 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxSatAssumptionOrder | ( | com.google.ortools.sat.SatParameters.MaxSatAssumptionOrder | value | ) |
optional .operations_research.sat.SatParameters.MaxSatAssumptionOrder max_sat_assumption_order = 51 [default = DEFAULT_ASSUMPTION_ORDER];
| value | The maxSatAssumptionOrder to set. |
Definition at line 28215 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxSatReverseAssumptionOrder | ( | boolean | value | ) |
If true, adds the assumption in the reverse order of the one defined by max_sat_assumption_order.
optional bool max_sat_reverse_assumption_order = 52 [default = false];
| value | The maxSatReverseAssumptionOrder to set. |
Definition at line 28270 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxSatStratification | ( | com.google.ortools.sat.SatParameters.MaxSatStratificationAlgorithm | value | ) |
optional .operations_research.sat.SatParameters.MaxSatStratificationAlgorithm max_sat_stratification = 53 [default = STRATIFICATION_DESCENT];
| value | The maxSatStratification to set. |
Definition at line 28315 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxSizeToCreatePrecedenceLiteralsInDisjunctive | ( | int | value | ) |
Create one literal for each disjunction of two pairs of tasks. This slows down the solve time, but improves the lower bound of the objective in the makespan case. This will be triggered if the number of intervals is less or equal than the parameter and if use_strong_propagation_in_disjunctive is true.
optional int32 max_size_to_create_precedence_literals_in_disjunctive = 229 [default = 60];
| value | The maxSizeToCreatePrecedenceLiteralsInDisjunctive to set. |
Definition at line 28555 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxTimeInSeconds | ( | double | value | ) |
Maximum time allowed in seconds to solve a problem. The counter will starts at the beginning of the Solve() call.
optional double max_time_in_seconds = 36 [default = inf];
| value | The maxTimeInSeconds to set. |
Definition at line 22141 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMaxVariableActivityValue | ( | double | value | ) |
optional double max_variable_activity_value = 16 [default = 1e+100];
| value | The maxVariableActivityValue to set. |
Definition at line 21109 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMergeAtMostOneWorkLimit | ( | double | value | ) |
optional double merge_at_most_one_work_limit = 146 [default = 100000000];
| value | The mergeAtMostOneWorkLimit to set. |
Definition at line 24986 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMergeNoOverlapWorkLimit | ( | double | value | ) |
During presolve, we use a maximum clique heuristic to merge together no-overlap constraints or at most one constraints. This code can be slow, so we have a limit in place on the number of explored nodes in the underlying graph. The internal limit is an int64, but we use double here to simplify manual input.
optional double merge_no_overlap_work_limit = 145 [default = 1000000000000];
| value | The mergeNoOverlapWorkLimit to set. |
Definition at line 24938 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMinimizationAlgorithm | ( | com.google.ortools.sat.SatParameters.ConflictMinimizationAlgorithm | value | ) |
optional .operations_research.sat.SatParameters.ConflictMinimizationAlgorithm minimization_algorithm = 4 [default = RECURSIVE];
| value | The minimizationAlgorithm to set. |
Definition at line 20461 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMinimizeReductionDuringPbResolution | ( | boolean | value | ) |
A different algorithm during PB resolution. It minimizes the number of calls to ReduceCoefficients() which can be time consuming. However, the search space will be different and if the coefficients are large, this may lead to integer overflows that could otherwise be prevented.
optional bool minimize_reduction_during_pb_resolution = 48 [default = false];
| value | The minimizeReductionDuringPbResolution to set. |
Definition at line 23295 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMinimizeSharedClauses | ( | boolean | value | ) |
Minimize and detect subsumption of shared clauses immediately after they are imported.
optional bool minimize_shared_clauses = 300 [default = true];
| value | The minimizeSharedClauses to set. |
Definition at line 27618 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMinOrthogonalityForLpConstraints | ( | double | value | ) |
While adding constraints, skip the constraints which have orthogonality less than 'min_orthogonality_for_lp_constraints' with already added constraints during current call. Orthogonality is defined as 1 - cosine(vector angle between constraints). A value of zero disable this feature.
optional double min_orthogonality_for_lp_constraints = 115 [default = 0.05];
| value | The minOrthogonalityForLpConstraints to set. |
Definition at line 36519 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMipAutomaticallyScaleVariables | ( | boolean | value | ) |
If true, some continuous variable might be automatically scaled. For now, this is only the case where we detect that a variable is actually an integer multiple of a constant. For instance, variables of the form k * 0.5 are quite frequent, and if we detect this, we will scale such variable domain by 2 to make it implied integer.
optional bool mip_automatically_scale_variables = 166 [default = true];
| value | The mipAutomaticallyScaleVariables to set. |
Definition at line 37499 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMipCheckPrecision | ( | double | value | ) |
As explained in mip_precision and mip_max_activity_exponent, we cannot always reach the wanted precision during scaling. We use this threshold to enphasize in the logs when the precision seems bad.
optional double mip_check_precision = 128 [default = 0.0001];
| value | The mipCheckPrecision to set. |
Definition at line 37869 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMipComputeTrueObjectiveBound | ( | boolean | value | ) |
Even if we make big error when scaling the objective, we can always derive a correct lower bound on the original objective by using the exact lower bound on the scaled integer version of the objective. This should be fast, but if you don't care about having a precise lower bound, you can turn it off.
optional bool mip_compute_true_objective_bound = 198 [default = true];
| value | The mipComputeTrueObjectiveBound to set. |
Definition at line 37939 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMipDropTolerance | ( | double | value | ) |
Any value in the input mip with a magnitude lower than this will be set to zero. This is to avoid some issue in LP presolving.
optional double mip_drop_tolerance = 232 [default = 1e-16];
| value | The mipDropTolerance to set. |
Definition at line 38150 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMipMaxActivityExponent | ( | int | value | ) |
To avoid integer overflow, we always force the maximum possible constraint activity (and objective value) according to the initial variable domain to be smaller than 2 to this given power. Because of this, we cannot always reach the "mip_wanted_precision" parameter above. This can go as high as 62, but some internal algo currently abort early if they might run into integer overflow, so it is better to keep it a bit lower than this.
optional int32 mip_max_activity_exponent = 127 [default = 53];
| value | The mipMaxActivityExponent to set. |
Definition at line 37800 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMipMaxBound | ( | double | value | ) |
We need to bound the maximum magnitude of the variables for CP-SAT, and that is the bound we use. If the MIP model expect larger variable value in the solution, then the converted model will likely not be relevant.
optional double mip_max_bound = 124 [default = 10000000];
| value | The mipMaxBound to set. |
Definition at line 37301 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMipMaxValidMagnitude | ( | double | value | ) |
Any finite values in the input MIP must be below this threshold, otherwise the model will be reported invalid. This is needed to avoid floating point overflow when evaluating bounds * coeff for instance. We are a bit more defensive, but in practice, users shouldn't use super large values in a MIP.
optional double mip_max_valid_magnitude = 199 [default = 1e+20];
| value | The mipMaxValidMagnitude to set. |
Definition at line 38011 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMipPresolveLevel | ( | int | value | ) |
When solving a MIP, we do some basic floating point presolving before scaling the problem to integer to be handled by CP-SAT. This control how much of that presolve we do. It can help to better scale floating point model, but it is not always behaving nicely.
optional int32 mip_presolve_level = 261 [default = 2];
| value | The mipPresolveLevel to set. |
Definition at line 38216 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMipScaleLargeDomain | ( | boolean | value | ) |
If this is false, then mip_var_scaling is only applied to variables with "small" domain. If it is true, we scale all floating point variable independenlty of their domain.
optional bool mip_scale_large_domain = 225 [default = false];
| value | The mipScaleLargeDomain to set. |
Definition at line 37429 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMipTreatHighMagnitudeBoundsAsInfinity | ( | boolean | value | ) |
By default, any variable/constraint bound with a finite value and a magnitude greater than the mip_max_valid_magnitude will result with a invalid model. This flags change the behavior such that such bounds are silently transformed to +∞ or -∞. It is recommended to keep it at false, and create valid bounds.
optional bool mip_treat_high_magnitude_bounds_as_infinity = 278 [default = false];
| value | The mipTreatHighMagnitudeBoundsAsInfinity to set. |
Definition at line 38086 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMipVarScaling | ( | double | value | ) |
All continuous variable of the problem will be multiplied by this factor. By default, we don't do any variable scaling and rely on the MIP model to specify continuous variable domain with the wanted precision.
optional double mip_var_scaling = 125 [default = 1];
| value | The mipVarScaling to set. |
Definition at line 37365 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setMipWantedPrecision | ( | double | value | ) |
When scaling constraint with double coefficients to integer coefficients, we will multiply by a power of 2 and round the coefficients. We will choose the lowest power such that we have no potential overflow (see mip_max_activity_exponent) and the worst case constraint activity error does not exceed this threshold. Note that we also detect constraint with rational coefficients and scale them accordingly when it seems better instead of using a power of 2. We also relax all constraint bounds by this absolute value. For pure integer constraint, if this value if lower than one, this will not change anything. However it is needed when scaling MIP problems. If we manage to scale a constraint correctly, the maximum error we can make will be twice this value (once for the scaling error and once for the relaxed bounds). If we are not able to scale that well, we will display that fact but still scale as best as we can.
optional double mip_wanted_precision = 126 [default = 1e-06];
| value | The mipWantedPrecision to set. |
Definition at line 37707 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setName | ( | java.lang.String | value | ) |
In some context, like in a portfolio of search, it makes sense to name a given parameters set for logging purpose.
optional string name = 171 [default = ""];
| value | The name to set. |
Definition at line 19753 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setNameBytes | ( | com.google.protobuf.ByteString | value | ) |
In some context, like in a portfolio of search, it makes sense to name a given parameters set for logging purpose.
optional string name = 171 [default = ""];
| value | The bytes for name to set. |
Definition at line 19786 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setNewConstraintsBatchSize | ( | int | value | ) |
Add that many lazy constraints (or cuts) at once in the LP. Note that at the beginning of the solve, we do add more than this.
optional int32 new_constraints_batch_size = 122 [default = 50];
| value | The newConstraintsBatchSize to set. |
Definition at line 36862 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setNewLinearPropagation | ( | boolean | value | ) |
The new linear propagation code treat all constraints at once and use an adaptation of Bellman-Ford-Tarjan to propagate constraint in a smarter order and potentially detect propagation cycle earlier.
optional bool new_linear_propagation = 224 [default = true];
| value | The newLinearPropagation to set. |
Definition at line 35269 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setNoOverlap2DBooleanRelationsLimit | ( | int | value | ) |
If less than this number of boxes are present in a no-overlap 2d, we create 4 Booleans per pair of boxes: - Box 2 is after Box 1 on x. - Box 1 is after Box 2 on x. - Box 2 is after Box 1 on y. - Box 1 is after Box 2 on y. Note that at least one of them must be true, and at most one on x and one on y can be true. This can significantly help in closing small problem. The SAT reasoning can be a lot more powerful when we take decision on such positional relations.
optional int32 no_overlap_2d_boolean_relations_limit = 321 [default = 10];
| value | The noOverlap2dBooleanRelationsLimit to set. |
Definition at line 29331 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setNumConflictsBeforeStrategyChanges | ( | int | value | ) |
After each restart, if the number of conflict since the last strategy change is greater that this, then we increment a "strategy_counter" that can be use to change the search strategy used by the following restarts.
optional int32 num_conflicts_before_strategy_changes = 68 [default = 0];
| value | The numConflictsBeforeStrategyChanges to set. |
Definition at line 22020 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setNumFullSubsolvers | ( | int | value | ) |
We distinguish subsolvers that consume a full thread, and the ones that are always interleaved. If left at zero, we will fix this with a default formula that depends on num_workers. But if you start modifying what runs, you might want to fix that to a given value depending on the num_workers you use.
optional int32 num_full_subsolvers = 294 [default = 0];
| value | The numFullSubsolvers to set. |
Definition at line 25982 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setNumSearchWorkers | ( | int | value | ) |
optional int32 num_search_workers = 100 [default = 0];
| value | The numSearchWorkers to set. |
Definition at line 25918 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setNumViolationLs | ( | int | value | ) |
This will create incomplete subsolvers (that are not LNS subsolvers) that use the feasibility jump code to find improving solution, treating the objective improvement as a hard constraint.
optional int32 num_violation_ls = 244 [default = 0];
| value | The numViolationLs to set. |
Definition at line 32217 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setNumWorkers | ( | int | value | ) |
Specify the number of parallel workers (i.e. threads) to use during search. This should usually be lower than your number of available cpus + hyperthread in your machine. A value of 0 means the solver will try to use all cores on the machine. A number of 1 means no parallelism. Note that 'num_workers' is the preferred name, but if it is set to zero, we will still read the deprecated 'num_search_workers'. As of 2020-04-10, if you're using SAT via MPSolver (to solve integer programs) this field is overridden with a value of 8, if the field is not set *explicitly*. Thus, always set this field explicitly or via MPSolver::SetNumThreads().
optional int32 num_workers = 206 [default = 0];
| value | The numWorkers to set. |
Definition at line 25861 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setOnlyAddCutsAtLevelZero | ( | boolean | value | ) |
For the cut that can be generated at any level, this control if we only try to generate them at the root node.
optional bool only_add_cuts_at_level_zero = 92 [default = false];
| value | The onlyAddCutsAtLevelZero to set. |
Definition at line 35726 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setOnlySolveIp | ( | boolean | value | ) |
If one try to solve a MIP model with CP-SAT, because we assume all variable to be integer after scaling, we will not necessarily have the correct optimal. Note however that all feasible solutions are valid since we will just solve a more restricted version of the original problem. This parameters is here to prevent user to think the solution is optimal when it might not be. One will need to manually set this to false to solve a MIP model where the optimal might be different. Note that this is tested after some MIP presolve steps, so even if not all original variable are integer, we might end up with a pure IP after presolve and after implied integer detection.
optional bool only_solve_ip = 222 [default = false];
| value | The onlySolveIp to set. |
Definition at line 37592 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setOptimizeWithCore | ( | boolean | value | ) |
The default optimization method is a simple "linear scan", each time trying to find a better solution than the previous one. If this is true, then we use a core-based approach (like in max-SAT) when we try to increase the lower bound instead.
optional bool optimize_with_core = 83 [default = false];
| value | The optimizeWithCore to set. |
Definition at line 31252 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setOptimizeWithLbTreeSearch | ( | boolean | value | ) |
Do a more conventional tree search (by opposition to SAT based one) where we keep all the explored node in a tree. This is meant to be used in a portfolio and focus on improving the objective lower bound. Keeping the whole tree allow us to report a better objective lower bound coming from the worst open node in the tree.
optional bool optimize_with_lb_tree_search = 188 [default = false];
| value | The optimizeWithLbTreeSearch to set. |
Definition at line 31323 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setOptimizeWithMaxHs | ( | boolean | value | ) |
This has no effect if optimize_with_core is false. If true, use a different core-based algorithm similar to the max-HS algo for max-SAT. This is a hybrid MIP/CP approach and it uses a MIP solver in addition to the CP/SAT one. This is also related to the PhD work of tobyodavies@ "Automatic Logic-Based Benders Decomposition with MiniZinc" http://aaai.org/ocs/index.php/AAAI/AAAI17/paper/view/14489
optional bool optimize_with_max_hs = 85 [default = false];
| value | The optimizeWithMaxHs to set. |
Definition at line 31542 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPbCleanupIncrement | ( | int | value | ) |
Same as for the clauses, but for the learned pseudo-Boolean constraints.
optional int32 pb_cleanup_increment = 46 [default = 200];
| value | The pbCleanupIncrement to set. |
Definition at line 20941 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPbCleanupRatio | ( | double | value | ) |
optional double pb_cleanup_ratio = 47 [default = 0.5];
| value | The pbCleanupRatio to set. |
Definition at line 20985 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPermutePresolveConstraintOrder | ( | boolean | value | ) |
optional bool permute_presolve_constraint_order = 179 [default = false];
| value | The permutePresolveConstraintOrder to set. |
Definition at line 22778 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPermuteVariableRandomly | ( | boolean | value | ) |
This is mainly here to test the solver variability. Note that in tests, if not explicitly set to false, all 3 options will be set to true so that clients do not rely on the solver returning a specific solution if they are many equivalent optimal solutions.
optional bool permute_variable_randomly = 178 [default = false];
| value | The permuteVariableRandomly to set. |
Definition at line 22731 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPolarityExploitLsHints | ( | boolean | value | ) |
If true and we have first solution LS workers, tries in some phase to follow a LS solutions that violates has litle constraints as possible.
optional bool polarity_exploit_ls_hints = 309 [default = false];
| value | The polarityExploitLsHints to set. |
Definition at line 20072 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPolarityRephaseIncrement | ( | int | value | ) |
If non-zero, then we change the polarity heuristic after that many number of conflicts in an arithmetically increasing fashion. So x the first time, 2 * x the second time, etc...
optional int32 polarity_rephase_increment = 168 [default = 1000];
| value | The polarityRephaseIncrement to set. |
Definition at line 20011 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPolishLpSolution | ( | boolean | value | ) |
Whether we try to do a few degenerate iteration at the end of an LP solve to minimize the fractionality of the integer variable in the basis. This helps on some problems, but not so much on others. It also cost of bit of time to do such polish step.
optional bool polish_lp_solution = 175 [default = false];
| value | The polishLpSolution to set. |
Definition at line 34756 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPreferredVariableOrder | ( | com.google.ortools.sat.SatParameters.VariableOrder | value | ) |
optional .operations_research.sat.SatParameters.VariableOrder preferred_variable_order = 1 [default = IN_ORDER];
| value | The preferredVariableOrder to set. |
Definition at line 19817 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPresolveBlockedClause | ( | boolean | value | ) |
Whether we use an heuristic to detect some basic case of blocked clause in the SAT presolve.
optional bool presolve_blocked_clause = 88 [default = true];
| value | The presolveBlockedClause to set. |
Definition at line 23761 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPresolveBvaThreshold | ( | int | value | ) |
Apply Bounded Variable Addition (BVA) if the number of clauses is reduced by stricly more than this threshold. The algorithm described in the paper uses 0, but quick experiments showed that 1 is a good value. It may not be worth it to add a new variable just to remove one clause.
optional int32 presolve_bva_threshold = 73 [default = 1];
| value | The presolveBvaThreshold to set. |
Definition at line 23883 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPresolveBveClauseWeight | ( | int | value | ) |
During presolve, we apply BVE only if this weight times the number of clauses plus the number of clause literals is not increased.
optional int32 presolve_bve_clause_weight = 55 [default = 3];
| value | The presolveBveClauseWeight to set. |
Definition at line 23589 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPresolveBveThreshold | ( | int | value | ) |
During presolve, only try to perform the bounded variable elimination (BVE) of a variable x if the number of occurrences of x times the number of occurrences of not(x) is not greater than this parameter.
optional int32 presolve_bve_threshold = 54 [default = 500];
| value | The presolveBveThreshold to set. |
Definition at line 23448 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPresolveExtractIntegerEnforcement | ( | boolean | value | ) |
If true, we will extract from linear constraints, enforcement literals of the form "integer variable at bound => simplified constraint". This should always be beneficial except that we don't always handle them as efficiently as we could for now. This causes problem on manna81.mps (LP relaxation not as tight it seems) and on neos-3354841-apure.mps.gz (too many literals created this way).
optional bool presolve_extract_integer_enforcement = 174 [default = false];
| value | The presolveExtractIntegerEnforcement to set. |
Definition at line 25121 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPresolveInclusionWorkLimit | ( | long | value | ) |
A few presolve operations involve detecting constraints included in other constraint. Since there can be a quadratic number of such pairs, and processing them usually involve scanning them, the complexity of these operations can be big. This enforce a local deterministic limit on the number of entries scanned. Default is 1e8. A value of zero will disable these presolve rules completely.
optional int64 presolve_inclusion_work_limit = 201 [default = 100000000];
| value | The presolveInclusionWorkLimit to set. |
Definition at line 25200 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPresolveProbingDeterministicTimeLimit | ( | double | value | ) |
optional double presolve_probing_deterministic_time_limit = 57 [default = 30];
| value | The presolveProbingDeterministicTimeLimit to set. |
Definition at line 23706 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPresolveSubstitutionLevel | ( | int | value | ) |
How much substitution (also called free variable aggregation in MIP litterature) should we perform at presolve. This currently only concerns variable appearing only in linear constraints. For now the value 0 turns it off and any positive value performs substitution.
optional int32 presolve_substitution_level = 147 [default = 1];
| value | The presolveSubstitutionLevel to set. |
Definition at line 25047 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPresolveUseBva | ( | boolean | value | ) |
Whether or not we use Bounded Variable Addition (BVA) in the presolve.
optional bool presolve_use_bva = 72 [default = true];
| value | The presolveUseBva to set. |
Definition at line 23818 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setProbingDeterministicTimeLimit | ( | double | value | ) |
The maximum "deterministic" time limit to spend in probing. A value of zero will disable the probing. TODO(user): Clean up. The first one is used in CP-SAT, the other in pure SAT presolve.
optional double probing_deterministic_time_limit = 226 [default = 1];
| value | The probingDeterministicTimeLimit to set. |
Definition at line 23658 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setProbingNumCombinationsLimit | ( | int | value | ) |
How many combinations of pairs or triplets of variables we want to scan.
optional int32 probing_num_combinations_limit = 272 [default = 20000];
| value | The probingNumCombinationsLimit to set. |
Definition at line 30747 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPropagationLoopDetectionFactor | ( | double | value | ) |
Some search decisions might cause a really large number of propagations to happen when integer variables with large domains are only reduced by 1 at each step. If we propagate more than the number of variable times this parameters we try to take counter-measure. Setting this to 0.0 disable this feature. TODO(user): Setting this to something like 10 helps in most cases, but the code is currently buggy and can cause the solve to enter a bad state where no progress is made.
optional double propagation_loop_detection_factor = 221 [default = 10];
| value | The propagationLoopDetectionFactor to set. |
Definition at line 28391 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPseudoCostReliabilityThreshold | ( | long | value | ) |
The solver ignores the pseudo costs of variables with number of recordings less than this threshold.
optional int64 pseudo_cost_reliability_threshold = 123 [default = 100];
| value | The pseudoCostReliabilityThreshold to set. |
Definition at line 31186 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setPushAllTasksTowardStart | ( | boolean | value | ) |
Experimental code: specify if the objective pushes all tasks toward the start of the schedule.
optional bool push_all_tasks_toward_start = 262 [default = false];
| value | The pushAllTasksTowardStart to set. |
Definition at line 34294 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRandomBranchesRatio | ( | double | value | ) |
A number between 0 and 1 that indicates the proportion of branching variables that are selected randomly instead of choosing the first variable from the given variable_ordering strategy.
optional double random_branches_ratio = 32 [default = 0];
| value | The randomBranchesRatio to set. |
Definition at line 20207 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRandomizeSearch | ( | boolean | value | ) |
Randomize fixed search.
optional bool randomize_search = 103 [default = false];
| value | The randomizeSearch to set. |
Definition at line 34171 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRandomPolarityRatio | ( | double | value | ) |
The proportion of polarity chosen at random. Note that this take precedence over the phase saving heuristic. This is different from initial_polarity:POLARITY_RANDOM because it will select a new random polarity each time the variable is branched upon instead of selecting one initially and then always taking this choice.
optional double random_polarity_ratio = 45 [default = 0];
| value | The randomPolarityRatio to set. |
Definition at line 20141 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRandomSeed | ( | int | value | ) |
At the beginning of each solve, the random number generator used in some part of the solver is reinitialized to this seed. If you change the random seed, the solver may make different choices during the solving process. For some problems, the running time may vary a lot depending on small change in the solving algorithm. Running the solver with different seeds enables to have more robust benchmarks when evaluating new features.
optional int32 random_seed = 31 [default = 1];
| value | The randomSeed to set. |
Definition at line 22660 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRelativeGapLimit | ( | double | value | ) |
optional double relative_gap_limit = 160 [default = 0];
| value | The relativeGapLimit to set. |
Definition at line 22590 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRemoveFixedVariablesEarly | ( | boolean | value | ) |
If cp_model_presolve is true and there is a large proportion of fixed variable after the first model copy, remap all the model to a dense set of variable before the full presolve even starts. This should help for LNS on large models.
optional bool remove_fixed_variables_early = 310 [default = true];
| value | The removeFixedVariablesEarly to set. |
Definition at line 24183 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRepairHint | ( | boolean | value | ) |
If true, the solver tries to repair the solution given in the hint. This search terminates after the 'hint_conflict_limit' is reached and the solver switches to regular search. If false, then we do a FIXED_SEARCH using the hint until the hint_conflict_limit is reached.
optional bool repair_hint = 167 [default = false];
| value | The repairHint to set. |
Definition at line 30508 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRestartAlgorithms | ( | int | index, |
| com.google.ortools.sat.SatParameters.RestartAlgorithm | value ) |
The restart strategies will change each time the strategy_counter is increased. The current strategy will simply be the one at index strategy_counter modulo the number of strategy. Note that if this list includes a NO_RESTART, nothing will change when it is reached because the strategy_counter will only increment after a restart. The idea of switching of search strategy tailored for SAT/UNSAT comes from Chanseok Oh with his COMiniSatPS solver, see http://cs.nyu.edu/~chanseok/. But more generally, it seems REALLY beneficial to try different strategy.
repeated .operations_research.sat.SatParameters.RestartAlgorithm restart_algorithms = 61;
| index | The index to set the value at. |
| value | The restartAlgorithms to set. |
Definition at line 21459 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRestartDlAverageRatio | ( | double | value | ) |
In the moving average restart algorithms, a restart is triggered if the window average times this ratio is greater that the global average.
optional double restart_dl_average_ratio = 63 [default = 1];
| value | The restartDlAverageRatio to set. |
Definition at line 21773 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRestartLbdAverageRatio | ( | double | value | ) |
optional double restart_lbd_average_ratio = 71 [default = 1];
| value | The restartLbdAverageRatio to set. |
Definition at line 21818 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRestartPeriod | ( | int | value | ) |
Restart period for the FIXED_RESTART strategy. This is also the multiplier used by the LUBY_RESTART strategy.
optional int32 restart_period = 30 [default = 50];
| value | The restartPeriod to set. |
Definition at line 21657 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRestartRunningWindowSize | ( | int | value | ) |
Size of the window for the moving average restarts.
optional int32 restart_running_window_size = 62 [default = 50];
| value | The restartRunningWindowSize to set. |
Definition at line 21714 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRootLpIterations | ( | int | value | ) |
Even at the root node, we do not want to spend too much time on the LP if it is "difficult". So we solve it in "chunks" of that many iterations. The solve will be continued down in the tree or the next time we go back to the root node.
optional int32 root_lp_iterations = 227 [default = 2000];
| value | The rootLpIterations to set. |
Definition at line 36448 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRoutingCutDpEffort | ( | double | value | ) |
The amount of "effort" to spend in dynamic programming for computing routing cuts. This is in term of basic operations needed by the algorithm in the worst case, so a value like 1e8 should take less than a second to compute.
optional double routing_cut_dp_effort = 314 [default = 10000000];
| value | The routingCutDpEffort to set. |
Definition at line 30284 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRoutingCutMaxInfeasiblePathLength | ( | int | value | ) |
If the length of an infeasible path is less than this value, a cut will be added to exclude it.
optional int32 routing_cut_max_infeasible_path_length = 317 [default = 6];
| value | The routingCutMaxInfeasiblePathLength to set. |
Definition at line 30346 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRoutingCutSubsetSizeForBinaryRelationBound | ( | int | value | ) |
If the size of a subset of nodes of a RoutesConstraint is less than this value, use linear constraints of size 1 and 2 (such as capacity and time window constraints) enforced by the arc literals to compute cuts for this subset (unless the subset size is less than routing_cut_subset_size_for_tight_binary_relation_bound, in which case the corresponding algorithm is used instead). The algorithm for these cuts has a O(n^3) complexity, where n is the subset size. Hence the value of this parameter should not be too large (e.g. 10 or 20).
optional int32 routing_cut_subset_size_for_binary_relation_bound = 312 [default = 0];
| value | The routingCutSubsetSizeForBinaryRelationBound to set. |
Definition at line 30016 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRoutingCutSubsetSizeForExactBinaryRelationBound | ( | int | value | ) |
Similar to above, but with an even stronger algorithm in O(n!). We try to be defensive and abort early or not run that often. Still the value of that parameter shouldn't really be much more than 10.
optional int32 routing_cut_subset_size_for_exact_binary_relation_bound = 316 [default = 8];
| value | The routingCutSubsetSizeForExactBinaryRelationBound to set. |
Definition at line 30149 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRoutingCutSubsetSizeForShortestPathsBound | ( | int | value | ) |
Similar to routing_cut_subset_size_for_exact_binary_relation_bound but use a bound based on shortest path distances (which respect triangular inequality). This allows to derive bounds that are valid for any superset of a given subset. This is slow, so it shouldn't really be larger than 10.
optional int32 routing_cut_subset_size_for_shortest_paths_bound = 318 [default = 8];
| value | The routingCutSubsetSizeForShortestPathsBound to set. |
Definition at line 30216 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setRoutingCutSubsetSizeForTightBinaryRelationBound | ( | int | value | ) |
Similar to above, but with a different algorithm producing better cuts, at the price of a higher O(2^n) complexity, where n is the subset size. Hence the value of this parameter should be small (e.g. less than 10).
optional int32 routing_cut_subset_size_for_tight_binary_relation_bound = 313 [default = 0];
| value | The routingCutSubsetSizeForTightBinaryRelationBound to set. |
Definition at line 30085 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSaveLpBasisInLbTreeSearch | ( | boolean | value | ) |
Experimental. Save the current LP basis at each node of the search tree so that when we jump around, we can load it and reduce the number of LP iterations needed. It currently works okay if we do not change the lp with cuts or simplification... More work is needed to make it robust in all cases.
optional bool save_lp_basis_in_lb_tree_search = 284 [default = false];
| value | The saveLpBasisInLbTreeSearch to set. |
Definition at line 31398 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSearchBranching | ( | com.google.ortools.sat.SatParameters.SearchBranching | value | ) |
optional .operations_research.sat.SatParameters.SearchBranching search_branching = 82 [default = AUTOMATIC_SEARCH];
| value | The searchBranching to set. |
Definition at line 30391 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSearchRandomVariablePoolSize | ( | long | value | ) |
Search randomization will collect the top 'search_random_variable_pool_size' valued variables, and pick one randomly. The value of the variable is specific to each strategy.
optional int64 search_random_variable_pool_size = 104 [default = 0];
| value | The searchRandomVariablePoolSize to set. |
Definition at line 34233 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setShareBinaryClauses | ( | boolean | value | ) |
Allows sharing of new learned binary clause between workers.
optional bool share_binary_clauses = 203 [default = true];
| value | The shareBinaryClauses to set. |
Definition at line 27499 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSharedTreeBalanceTolerance | ( | int | value | ) |
How much deeper compared to the ideal max depth of the tree is considered "balanced" enough to still accept a split. Without such a tolerance, sometimes the tree can only be split by a single worker, and they may not generate a split for some time. In contrast, with a tolerance of 1, at least half of all workers should be able to split the tree as soon as a split becomes required. This only has an effect on SPLIT_STRATEGY_BALANCED_TREE and SPLIT_STRATEGY_DISCREPANCY.
optional int32 shared_tree_balance_tolerance = 305 [default = 1];
| value | The sharedTreeBalanceTolerance to set. |
Definition at line 32889 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSharedTreeMaxNodesPerWorker | ( | int | value | ) |
In order to limit total shared memory and communication overhead, limit the total number of nodes that may be generated in the shared tree. If the shared tree runs out of unassigned leaves, workers act as portfolio workers. Note: this limit includes interior nodes, not just leaves.
optional int32 shared_tree_max_nodes_per_worker = 238 [default = 10000];
| value | The sharedTreeMaxNodesPerWorker to set. |
Definition at line 32772 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSharedTreeNumWorkers | ( | int | value | ) |
Enables shared tree search. If positive, start this many complete worker threads to explore a shared search tree. These workers communicate objective bounds and simple decision nogoods relating to the shared prefix of the tree, and will avoid exploring the same subtrees as one another. Specifying a negative number uses a heuristic to select an appropriate number of shared tree workeres based on the total number of workers.
optional int32 shared_tree_num_workers = 235 [default = 0];
| value | The sharedTreeNumWorkers to set. |
Definition at line 32409 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSharedTreeOpenLeavesPerWorker | ( | double | value | ) |
How many open leaf nodes should the shared tree maintain per worker.
optional double shared_tree_open_leaves_per_worker = 281 [default = 2];
| value | The sharedTreeOpenLeavesPerWorker to set. |
Definition at line 32707 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSharedTreeSplitStrategy | ( | com.google.ortools.sat.SatParameters.SharedTreeSplitStrategy | value | ) |
optional .operations_research.sat.SatParameters.SharedTreeSplitStrategy shared_tree_split_strategy = 239 [default = SPLIT_STRATEGY_AUTO];
| value | The sharedTreeSplitStrategy to set. |
Definition at line 32819 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSharedTreeWorkerEnablePhaseSharing | ( | boolean | value | ) |
If true, shared tree workers share their target phase when returning an assigned subtree for the next worker to use.
optional bool shared_tree_worker_enable_phase_sharing = 304 [default = true];
| value | The sharedTreeWorkerEnablePhaseSharing to set. |
Definition at line 32650 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSharedTreeWorkerEnableTrailSharing | ( | boolean | value | ) |
If true, workers share more of the information from their local trail. Specifically, literals implied by the shared tree decisions.
optional bool shared_tree_worker_enable_trail_sharing = 295 [default = true];
| value | The sharedTreeWorkerEnableTrailSharing to set. |
Definition at line 32590 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSharedTreeWorkerMinRestartsPerSubtree | ( | int | value | ) |
Minimum restarts before a worker will replace a subtree that looks "bad" based on the average LBD of learned clauses.
optional int32 shared_tree_worker_min_restarts_per_subtree = 282 [default = 1];
| value | The sharedTreeWorkerMinRestartsPerSubtree to set. |
Definition at line 32530 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setShareGlueClauses | ( | boolean | value | ) |
Allows sharing of short glue clauses between workers. Implicitly disabled if share_binary_clauses is false.
optional bool share_glue_clauses = 285 [default = false];
| value | The shareGlueClauses to set. |
Definition at line 27558 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setShareGlueClausesDtime | ( | double | value | ) |
The amount of dtime between each export of shared glue clauses.
optional double share_glue_clauses_dtime = 322 [default = 1];
| value | The shareGlueClausesDtime to set. |
Definition at line 27675 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setShareLevelZeroBounds | ( | boolean | value | ) |
Allows sharing of the bounds of modified variables at level 0.
optional bool share_level_zero_bounds = 114 [default = true];
| value | The shareLevelZeroBounds to set. |
Definition at line 27443 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setShareObjectiveBounds | ( | boolean | value | ) |
Allows objective sharing between workers.
optional bool share_objective_bounds = 113 [default = true];
| value | The shareObjectiveBounds to set. |
Definition at line 27387 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setShavingDeterministicTimeInProbingSearch | ( | double | value | ) |
Add a shaving phase (where the solver tries to prove that the lower or upper bound of a variable are infeasible) to the probing search. (<= 0 disables it).
optional double shaving_deterministic_time_in_probing_search = 204 [default = 0.001];
| value | The shavingDeterministicTimeInProbingSearch to set. |
Definition at line 30809 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setShavingSearchDeterministicTime | ( | double | value | ) |
Specifies the amount of deterministic time spent of each try at shaving a bound in the shaving search.
optional double shaving_search_deterministic_time = 205 [default = 0.1];
| value | The shavingSearchDeterministicTime to set. |
Definition at line 30870 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setShavingSearchThreshold | ( | long | value | ) |
Specifies the threshold between two modes in the shaving procedure. If the range of the variable/objective is less than this threshold, then the shaving procedure will try to remove values one by one. Otherwise, it will try to remove one range at a time.
optional int64 shaving_search_threshold = 290 [default = 64];
| value | The shavingSearchThreshold to set. |
Definition at line 30936 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSolutionPoolSize | ( | int | value | ) |
Size of the top-n different solutions kept by the solver. This parameter must be > 0. Currently this only impact the "base" solution chosen for a LNS fragment.
optional int32 solution_pool_size = 193 [default = 3];
| value | The solutionPoolSize to set. |
Definition at line 33785 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setStopAfterFirstSolution | ( | boolean | value | ) |
For an optimization problem, stop the solver as soon as we have a solution.
optional bool stop_after_first_solution = 98 [default = false];
| value | The stopAfterFirstSolution to set. |
Definition at line 33415 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setStopAfterPresolve | ( | boolean | value | ) |
Mainly used when improving the presolver. When true, stops the solver after the presolve is complete (or after loading and root level propagation).
optional bool stop_after_presolve = 149 [default = false];
| value | The stopAfterPresolve to set. |
Definition at line 33474 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setStopAfterRootPropagation | ( | boolean | value | ) |
optional bool stop_after_root_propagation = 252 [default = false];
| value | The stopAfterRootPropagation to set. |
Definition at line 33519 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setStrategyChangeIncreaseRatio | ( | double | value | ) |
The parameter num_conflicts_before_strategy_changes is increased by that much after each strategy change.
optional double strategy_change_increase_ratio = 69 [default = 0];
| value | The strategyChangeIncreaseRatio to set. |
Definition at line 22081 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSubsolverParams | ( | int | index, |
| com.google.ortools.sat.SatParameters | value ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Definition at line 26930 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSubsolverParams | ( | int | index, |
| com.google.ortools.sat.SatParameters.Builder | builderForValue ) |
It is possible to specify additional subsolver configuration. These can be referred by their params.name() in the fields above. Note that only the specified field will "overwrite" the ones of the base parameter. If a subsolver_params has the name of an existing subsolver configuration, the named parameters will be merged into the subsolver configuration.
repeated .operations_research.sat.SatParameters subsolver_params = 210;
Definition at line 26955 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSubsolvers | ( | int | index, |
| java.lang.String | value ) |
In multi-thread, the solver can be mainly seen as a portfolio of solvers with different parameters. This field indicates the names of the parameters that are used in multithread. This only applies to "full" subsolvers. See cp_model_search.cc to see a list of the names and the default value (if left empty) that looks like: - default_lp (linearization_level:1) - fixed (only if fixed search specified or scheduling) - no_lp (linearization_level:0) - max_lp (linearization_level:2) - pseudo_costs (only if objective, change search heuristic) - reduced_costs (only if objective, change search heuristic) - quick_restart (kind of probing) - quick_restart_no_lp (kind of probing with linearization_level:0) - lb_tree_search (to improve lower bound, MIP like tree search) - probing (continuous probing and shaving) Also, note that some set of parameters will be ignored if they do not make sense. For instance if there is no objective, pseudo_cost or reduced_cost search will be ignored. Core based search will only work if the objective has many terms. If there is no fixed strategy fixed will be ignored. And so on. The order is important, as only the first num_full_subsolvers will be scheduled. You can see in the log which one are selected for a given run.
repeated string subsolvers = 207;
| index | The index to set the value at. |
| value | The subsolvers to set. |
Definition at line 26195 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSubsumptionDuringConflictAnalysis | ( | boolean | value | ) |
At a really low cost, during the 1-UIP conflict computation, it is easy to detect if some of the involved reasons are subsumed by the current conflict. When this is true, such clauses are detached and later removed from the problem.
optional bool subsumption_during_conflict_analysis = 56 [default = true];
| value | The subsumptionDuringConflictAnalysis to set. |
Definition at line 20562 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSymmetryDetectionDeterministicTimeLimit | ( | double | value | ) |
Deterministic time limit for symmetry detection.
optional double symmetry_detection_deterministic_time_limit = 302 [default = 1];
| value | The symmetryDetectionDeterministicTimeLimit to set. |
Definition at line 35207 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setSymmetryLevel | ( | int | value | ) |
Whether we try to automatically detect the symmetries in a model and exploit them. Currently, at level 1 we detect them in presolve and try to fix Booleans. At level 2, we also do some form of dynamic symmetry breaking during search. At level 3, we also detect symmetries for very large models, which can be slow. At level 4, we try to break as much symmetry as possible in presolve.
optional int32 symmetry_level = 183 [default = 2];
| value | The symmetryLevel to set. |
Definition at line 35006 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setTableCompressionLevel | ( | int | value | ) |
How much we try to "compress" a table constraint. Compressing more leads to less Booleans and faster propagation but can reduced the quality of the lp relaxation. Values goes from 0 to 3 where we always try to fully compress a table. At 2, we try to automatically decide if it is worth it.
optional int32 table_compression_level = 217 [default = 2];
| value | The tableCompressionLevel to set. |
Definition at line 24327 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseAbslRandom | ( | boolean | value | ) |
optional bool use_absl_random = 180 [default = false];
| value | The useAbslRandom to set. |
Definition at line 22818 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseAllDifferentForCircuit | ( | boolean | value | ) |
Turn on extra propagation for the circuit constraint. This can be quite slow.
optional bool use_all_different_for_circuit = 311 [default = false];
| value | The useAllDifferentForCircuit to set. |
Definition at line 29938 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseAreaEnergeticReasoningInNoOverlap2D | ( | boolean | value | ) |
When this is true, the no_overlap_2d constraint is reinforced with an energetic reasoning that uses an area-based energy. This can be combined with the two other overlap heuristics above.
optional bool use_area_energetic_reasoning_in_no_overlap_2d = 271 [default = false];
| value | The useAreaEnergeticReasoningInNoOverlap2d to set. |
Definition at line 29541 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseBlockingRestart | ( | boolean | value | ) |
Block a moving restart algorithm if the trail size of the current conflict is greater than the multiplier times the moving average of the trail size at the previous conflicts.
optional bool use_blocking_restart = 64 [default = false];
| value | The useBlockingRestart to set. |
Definition at line 21876 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseCombinedNoOverlap | ( | boolean | value | ) |
This can be beneficial if there is a lot of no-overlap constraints but a relatively low number of different intervals in the problem. Like 1000 intervals, but 1M intervals in the no-overlap constraints covering them.
optional bool use_combined_no_overlap = 133 [default = false];
| value | The useCombinedNoOverlap to set. |
Definition at line 34501 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseConservativeScaleOverloadChecker | ( | boolean | value | ) |
Enable a heuristic to solve cumulative constraints using a modified energy constraint. We modify the usual energy definition by applying a super-additive function (also called "conservative scale" or "dual-feasible function") to the demand and the durations of the tasks. This heuristic is fast but for most problems it does not help much to find a solution.
optional bool use_conservative_scale_overload_checker = 286 [default = false];
| value | The useConservativeScaleOverloadChecker to set. |
Definition at line 28885 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseDisjunctiveConstraintInCumulative | ( | boolean | value | ) |
When this is true, the cumulative constraint is reinforced with propagators from the disjunctive constraint to improve the inference on a set of tasks that are disjunctive at the root of the problem. This additional level supplements the default level of reasoning. Propagators of the cumulative constraint will not be used at all if all the tasks are disjunctive at root node. This always result in better propagation, but it is usually slow, so depending on the problem, turning this off may lead to a faster solution.
optional bool use_disjunctive_constraint_in_cumulative = 80 [default = true];
| value | The useDisjunctiveConstraintInCumulative to set. |
Definition at line 29230 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseDualSchedulingHeuristics | ( | boolean | value | ) |
When set, it activates a few scheduling parameters to improve the lower bound of scheduling problems. This is only effective with multiple workers as it modifies the reduced_cost, lb_tree_search, and probing workers.
optional bool use_dual_scheduling_heuristics = 214 [default = true];
| value | The useDualSchedulingHeuristics to set. |
Definition at line 29877 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseDynamicPrecedenceInCumulative | ( | boolean | value | ) |
optional bool use_dynamic_precedence_in_cumulative = 268 [default = false];
| value | The useDynamicPrecedenceInCumulative to set. |
Definition at line 28735 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseDynamicPrecedenceInDisjunctive | ( | boolean | value | ) |
Whether we try to branch on decision "interval A before interval B" rather than on intervals bounds. This usually works better, but slow down a bit the time to find the first solution. These parameters are still EXPERIMENTAL, the result should be correct, but it some corner cases, they can cause some failing CHECK in the solver.
optional bool use_dynamic_precedence_in_disjunctive = 263 [default = false];
| value | The useDynamicPrecedenceInDisjunctive to set. |
Definition at line 28686 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseEnergeticReasoningInNoOverlap2D | ( | boolean | value | ) |
When this is true, the no_overlap_2d constraint is reinforced with energetic reasoning. This additional level supplements the default level of reasoning.
optional bool use_energetic_reasoning_in_no_overlap_2d = 213 [default = false];
| value | The useEnergeticReasoningInNoOverlap2d to set. |
Definition at line 29477 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseErwaHeuristic | ( | boolean | value | ) |
Whether we use the ERWA (Exponential Recency Weighted Average) heuristic as described in "Learning Rate Based Branching Heuristic for SAT solvers", J.H.Liang, V. Ganesh, P. Poupart, K.Czarnecki, SAT 2016.
optional bool use_erwa_heuristic = 75 [default = false];
| value | The useErwaHeuristic to set. |
Definition at line 20271 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseExactLpReason | ( | boolean | value | ) |
The solver usually exploit the LP relaxation of a model. If this option is true, then whatever is infered by the LP will be used like an heuristic to compute EXACT propagation on the IP. So with this option, there is no numerical imprecision issues.
optional bool use_exact_lp_reason = 109 [default = true];
| value | The useExactLpReason to set. |
Definition at line 34436 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseExtendedProbing | ( | boolean | value | ) |
Use extended probing (probe bool_or, at_most_one, exactly_one).
optional bool use_extended_probing = 269 [default = true];
| value | The useExtendedProbing to set. |
Definition at line 30691 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseFeasibilityJump | ( | boolean | value | ) |
Parameters for an heuristic similar to the one described in the paper: "Feasibility Jump: an LP-free Lagrangian MIP heuristic", Bjørnar Luteberget, Giorgio Sartor, 2023, Mathematical Programming Computation.
optional bool use_feasibility_jump = 265 [default = true];
| value | The useFeasibilityJump to set. |
Definition at line 31609 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseFeasibilityPump | ( | boolean | value | ) |
Adds a feasibility pump subsolver along with lns subsolvers.
optional bool use_feasibility_pump = 164 [default = true];
| value | The useFeasibilityPump to set. |
Definition at line 33899 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseHardPrecedencesInCumulative | ( | boolean | value | ) |
If true, detect and create constraint for integer variable that are "after" a set of intervals in the same cumulative constraint. Experimental: by default we just use "direct" precedences. If exploit_all_precedences is true, we explore the full precedence graph. This assumes we have a DAG otherwise it fails.
optional bool use_hard_precedences_in_cumulative = 215 [default = false];
| value | The useHardPrecedencesInCumulative to set. |
Definition at line 29102 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseImpliedBounds | ( | boolean | value | ) |
Stores and exploits "implied-bounds" in the solver. That is, relations of the form literal => (var >= bound). This is currently used to derive stronger cuts.
optional bool use_implied_bounds = 144 [default = true];
| value | The useImpliedBounds to set. |
Definition at line 34689 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseLbRelaxLns | ( | boolean | value | ) |
Turns on neighborhood generator based on local branching LP. Based on Huang et al., "Local Branching Relaxation Heuristics for Integer Linear Programs", 2023.
optional bool use_lb_relax_lns = 255 [default = true];
| value | The useLbRelaxLns to set. |
Definition at line 33961 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseLinear3ForNoOverlap2DPrecedences | ( | boolean | value | ) |
When set, this activates a propagator for the no_overlap_2d constraint that
uses any eventual linear constraints of the model in the form
`{start interval 1} - {end interval 2} + c*w <= ub` to detect that two
intervals must overlap in one dimension for some values of `w`. This is
particularly useful for problems where the distance between two boxes is
part of the model.
optional bool use_linear3_for_no_overlap_2d_precedences = 323 [default = true];
| value | The useLinear3ForNoOverlap2dPrecedences to set. |
Definition at line 29810 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseLns | ( | boolean | value | ) |
Testing parameters used to disable all lns workers.
optional bool use_lns = 283 [default = true];
| value | The useLns to set. |
Definition at line 33667 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseLnsOnly | ( | boolean | value | ) |
Experimental parameters to disable everything but lns.
optional bool use_lns_only = 101 [default = false];
| value | The useLnsOnly to set. |
Definition at line 33723 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseLsOnly | ( | boolean | value | ) |
Disable every other type of subsolver, setting this turns CP-SAT into a pure local-search solver.
optional bool use_ls_only = 240 [default = false];
| value | The useLsOnly to set. |
Definition at line 31670 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseObjectiveLbSearch | ( | boolean | value | ) |
If true, search will search in ascending max objective value (when minimizing) starting from the lower bound of the objective.
optional bool use_objective_lb_search = 228 [default = false];
| value | The useObjectiveLbSearch to set. |
Definition at line 30998 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseObjectiveShavingSearch | ( | boolean | value | ) |
This search differs from the previous search as it will not use assumptions to bound the objective, and it will recreate a full model with the hardcoded objective value.
optional bool use_objective_shaving_search = 253 [default = false];
| value | The useObjectiveShavingSearch to set. |
Definition at line 31061 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseOptimizationHints | ( | boolean | value | ) |
For an optimization problem, whether we follow some hints in order to find a better first solution. For a variable with hint, the solver will always try to follow the hint. It will revert to the variable_branching default otherwise.
optional bool use_optimization_hints = 35 [default = true];
| value | The useOptimizationHints to set. |
Definition at line 27976 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseOptionalVariables | ( | boolean | value | ) |
If true, we automatically detect variables whose constraint are always enforced by the same literal and we mark them as optional. This allows to propagate them as if they were present in some situation. TODO(user): This is experimental and seems to lead to wrong optimal in some situation. It should however gives correct solutions. Fix.
optional bool use_optional_variables = 108 [default = false];
| value | The useOptionalVariables to set. |
Definition at line 34366 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseOverloadCheckerInCumulative | ( | boolean | value | ) |
When this is true, the cumulative constraint is reinforced with overload checking, i.e., an additional level of reasoning based on energy. This additional level supplements the default level of reasoning as well as timetable edge finding. This always result in better propagation, but it is usually slow, so depending on the problem, turning this off may lead to a faster solution.
optional bool use_overload_checker_in_cumulative = 78 [default = false];
| value | The useOverloadCheckerInCumulative to set. |
Definition at line 28805 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUsePbResolution | ( | boolean | value | ) |
Whether to use pseudo-Boolean resolution to analyze a conflict. Note that this option only make sense if your problem is modelized using pseudo-Boolean constraints. If you only have clauses, this shouldn't change anything (except slow the solver down).
optional bool use_pb_resolution = 43 [default = false];
| value | The usePbResolution to set. |
Definition at line 23227 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUsePhaseSaving | ( | boolean | value | ) |
If this is true, then the polarity of a variable will be the last value it was assigned to, or its default polarity if it was never assigned since the call to ResetDecisionHeuristic(). Actually, we use a newer version where we follow the last value in the longest non-conflicting partial assignment in the current phase. This is called 'literal phase saving'. For details see 'A Lightweight Component Caching Scheme for Satisfiability Solvers' K. Pipatsrisawat and A.Darwiche, In 10th International Conference on Theory and Applications of Satisfiability Testing, 2007.
optional bool use_phase_saving = 44 [default = true];
| value | The usePhaseSaving to set. |
Definition at line 19939 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUsePrecedencesInDisjunctiveConstraint | ( | boolean | value | ) |
When this is true, then a disjunctive constraint will try to use the precedence relations between time intervals to propagate their bounds further. For instance if task A and B are both before C and task A and B are in disjunction, then we can deduce that task C must start after duration(A) + duration(B) instead of simply max(duration(A), duration(B)), provided that the start time for all task was currently zero. This always result in better propagation, but it is usually slow, so depending on the problem, turning this off may lead to a faster solution.
optional bool use_precedences_in_disjunctive_constraint = 74 [default = true];
| value | The usePrecedencesInDisjunctiveConstraint to set. |
Definition at line 28479 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseProbingSearch | ( | boolean | value | ) |
If true, search will continuously probe Boolean variables, and integer variable bounds. This parameter is set to true in parallel on the probing worker.
optional bool use_probing_search = 176 [default = false];
| value | The useProbingSearch to set. |
Definition at line 30633 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseRinsLns | ( | boolean | value | ) |
Turns on relaxation induced neighborhood generator.
optional bool use_rins_lns = 129 [default = true];
| value | The useRinsLns to set. |
Definition at line 33843 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseSatInprocessing | ( | boolean | value | ) |
Enable or disable "inprocessing" which is some SAT presolving done at each restart to the root level.
optional bool use_sat_inprocessing = 163 [default = true];
| value | The useSatInprocessing to set. |
Definition at line 25473 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseSharedTreeSearch | ( | boolean | value | ) |
Set on shared subtree workers. Users should not set this directly.
optional bool use_shared_tree_search = 236 [default = false];
| value | The useSharedTreeSearch to set. |
Definition at line 32471 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseStrongPropagationInDisjunctive | ( | boolean | value | ) |
Enable stronger and more expensive propagation on no_overlap constraint.
optional bool use_strong_propagation_in_disjunctive = 230 [default = false];
| value | The useStrongPropagationInDisjunctive to set. |
Definition at line 28615 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseSymmetryInLp | ( | boolean | value | ) |
When we have symmetry, it is possible to "fold" all variables from the same orbit into a single variable, while having the same power of LP relaxation. This can help significantly on symmetric problem. However there is currently a bit of overhead as the rest of the solver need to do some translation between the folded LP and the rest of the problem.
optional bool use_symmetry_in_lp = 301 [default = false];
| value | The useSymmetryInLp to set. |
Definition at line 35079 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseTimetableEdgeFindingInCumulative | ( | boolean | value | ) |
When this is true, the cumulative constraint is reinforced with timetable edge finding, i.e., an additional level of reasoning based on the conjunction of energy and mandatory parts. This additional level supplements the default level of reasoning as well as overload_checker. This always result in better propagation, but it is usually slow, so depending on the problem, turning this off may lead to a faster solution.
optional bool use_timetable_edge_finding_in_cumulative = 79 [default = false];
| value | The useTimetableEdgeFindingInCumulative to set. |
Definition at line 28965 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseTimetablingInNoOverlap2D | ( | boolean | value | ) |
When this is true, the no_overlap_2d constraint is reinforced with propagators from the cumulative constraints. It consists of ignoring the position of rectangles in one position and projecting the no_overlap_2d on the other dimension to create a cumulative constraint. This is done on both axis. This additional level supplements the default level of reasoning.
optional bool use_timetabling_in_no_overlap_2d = 200 [default = false];
| value | The useTimetablingInNoOverlap2d to set. |
Definition at line 29411 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setUseTryEdgeReasoningInNoOverlap2D | ( | boolean | value | ) |
optional bool use_try_edge_reasoning_in_no_overlap_2d = 299 [default = false];
| value | The useTryEdgeReasoningInNoOverlap2d to set. |
Definition at line 29587 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setVariableActivityDecay | ( | double | value | ) |
Each time a conflict is found, the activities of some variables are increased by one. Then, the activity of all variables are multiplied by variable_activity_decay. To implement this efficiently, the activity of all the variables is not decayed at each conflict. Instead, the activity increment is multiplied by 1 / decay. When an activity reach max_variable_activity_value, all the activity are multiplied by 1 / max_variable_activity_value.
optional double variable_activity_decay = 15 [default = 0.8];
| value | The variableActivityDecay to set. |
Definition at line 21058 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setVariablesShavingLevel | ( | int | value | ) |
This search takes all Boolean or integer variables, and maximize or minimize them in order to reduce their domain. -1 is automatic, otherwise value 0 disables it, and 1, 2, or 3 changes something.
optional int32 variables_shaving_level = 289 [default = -1];
| value | The variablesShavingLevel to set. |
Definition at line 31125 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setViolationLsCompoundMoveProbability | ( | double | value | ) |
Probability of using compound move search each restart. TODO(user): Add reference to paper when published.
optional double violation_ls_compound_move_probability = 259 [default = 0.5];
| value | The violationLsCompoundMoveProbability to set. |
Definition at line 32334 of file SatParameters.java.
| Builder com.google.ortools.sat.SatParameters.Builder.setViolationLsPerturbationPeriod | ( | int | value | ) |
How long violation_ls should wait before perturbating a solution.
optional int32 violation_ls_perturbation_period = 249 [default = 100];
| value | The violationLsPerturbationPeriod to set. |
Definition at line 32275 of file SatParameters.java.
1.14.0