GCG Optimizer Options for AMPL
--------------------------------------------
To set these options, assign a string specifying their values to the AMPL
option "gcg_options". For example:
ampl: option gcg_options 'mipgap=1e-6';
Options:
acc:linrange (acc:linrng)
Solver acceptance level for 'LinConRange', default 2:
0 - Not accepted natively, automatic redefinition will be attempted
1 - Accepted but automatic redefinition will be used where possible
2 - Accepted natively and preferred
alg:basis (basis)
Whether to use or return a basis:
0 - No
1 - Use incoming basis (if provided)
2 - Return final basis
3 - Both (1 + 2 = default)
alg:method (method, lpmethod)
LP algorithm for solving initial LP relaxations:
s - automatic simplex (default)
p - primal simplex
d - dual simplex
b - barrier
c - barrier with crossover
alg:relax (relax)
0*/1: Whether to relax integrality of variables.
alg:remethod (remethod, relpmethod)
LP algorithm for resolving LP relaxations if a starting basis exists:
s - automatic simplex (default)
p - primal simplex
d - dual simplex
b - barrier
c - barrier with crossover
branch:preferbinary (preferbinary)
0/1: whether branching on binary variables should be preferred
0 - Binary variables should not be preferred on branching (default)
1 - Binary variables should be preferred on branching.
cut:dircutoffdistweight (dircutoffdistweight)
Weight of directed cutoff distance in cut score calculation (default:
0.0)
cut:efficacyweight (efficacyweight)
Weight of efficacy in cut score calculation (default: 1.0)
cut:intsupportweight (intsupportweight)
Weight of integral support in cut score calculation (default: 0.1)
cut:maxcuts (maxcuts)
Maximal number of cuts separated per separation round (0: disable local
separation; default: 100)
cut:maxcutsroot (maxcutsroot)
Maximal number of separated cuts at the root node (0: disable root node
separation; default: 2000)
cut:maxrounds
Maximal number of separation rounds per node (default: -1: unlimited)
cut:maxroundsroot
Maximal number of separation rounds in the root node (default: -1:
unlimited)
cut:maxstallrounds
Maximal number of consecutive separation rounds without objective or
integrality improvement in local nodes (-1: no additional restriction;
default: 1)
cut:maxstallroundsroot
Maximal number of consecutive separation rounds without objective or
integrality improvement in the root node (-1: no additional restriction;
default: 10)
cut:minactivityquot
Minimum cut activity quotient to convert cuts into constraints during a
restart (0.0: all cuts are converted; default: 0.8)
cut:minefficacy
Minimal efficacy for a cut to enter the LP (default: 0.0001)
cut:minefficacyroot
Minimal efficacy for a cut to enter the LP in the root node (default:
0.0001)
cut:minortho (minortho)
Minimal orthogonality for a cut to enter the LP (default: 0.9)
cut:minorthoroot (minorthoroot)
Minimal orthogonality for a cut to enter the LP in the root node
(default: 0.1)
cut:objparalweight (objparalweight)
Weight of objective parallelism in cut score calculation (default: 0.1)
cut:poolfreq
Separation frequency for the global cut pool (-1: disable global cut
pool; 0: only separate pool at the root; default: 10)
cut:settings
0/1/2/3: sets cuts settings
0 - Sets cuts default (default)
1 - Sets cuts aggressive
2 - Sets cuts fast
3 - Sets cuts off.
cvt:bigM (cvt:bigm, cvt:mip:bigM, cvt:mip:bigm)
Default value of big-M for linearization of logical constraints. Not
used by default. Use with care (prefer tight bounds). Should be smaller
than (1.0 / [integrality tolerance])
cvt:expcones (expcones)
0*/1: Recognize exponential cones.
cvt:mip:eps (cvt:cmp:eps, cmp:eps)
Tolerance for strict comparison of continuous variables for MIP. Applies
to <, >, and != operators. Also applies to negation of conditional
comparisons: b==1 <==> x<=5 means that with b==0, x>=5+eps. Default:
1e-4.
cvt:names (names, modelnames)
Whether to read or generate variable / constraint / objective names:
0 - No names
1 - (Default) Only provide names if at least one of .col / .row name
files was written by AMPL (AMPL: `option [<solver>_]auxfiles rc;`)
2 - Read names from AMPL, but create generic names if not provided
3 - Create generic names.
cvt:plapprox:domain (plapprox:domain, plapproxdomain)
For piecewise-linear approximated functions, both arguments and result
are bounded to +-[pladomain]. Default 1e6.
cvt:plapprox:reltol (plapprox:reltol, plapproxreltol)
Relative tolerance for piecewise-linear approximation. Default 0.01.
cvt:pre:all
0/1*: Set to 0 to disable most presolve in the flat converter.
cvt:pre:eqbinary
0/1*: Preprocess reified equality comparison with a binary variable.
cvt:pre:eqresult
0/1*: Preprocess reified equality comparison's boolean result bounds.
cvt:pre:unnest
0/1*: Inline nested expressions, currently Ands/Ors.
cvt:quadcon (passquadcon)
0*/1: Multiply out and pass quadratic constraint terms to the solver,
vs. linear approximation.
cvt:quadobj (passquadobj)
0*/1: Multiply out and pass quadratic objective terms to the solver, vs.
linear approximation.
cvt:socp (socpmode, socp)
Second-Order Cone recognition mode:
0 - Do not recognize SOCP forms
1 - Recognize from non-quadratic expressions only (sqrt, abs)
2 - Recognize from quadratic and non-quadratic SOCP forms
Recognized SOCP forms can be further converted to (SOCP-standardized)
quadratic constraints, see cvt:socp2qc. Default: 0.
cvt:socp2qc (socp2qcmode, socp2qc)
Mode to convert recognized SOCP forms to SOCP-standardized quadratic
constraints:
0 - Do not convert
1 - Convert if no other cone types found, and not all original
quadratics could be recognized as SOC, in particular if the
objective is quadratic
2 - Always convert
Such conversion can be necessary if the solver does not accept a mix of
conic and quadratic constraints/objectives. Default: 2.
cvt:sos (sos)
0/1*: Whether to honor declared suffixes .sosno and .ref describing SOS
sets. Each distinct nonzero .sosno value designates an SOS set, of type
1 for positive .sosno values and of type 2 for negative values. The .ref
suffix contains corresponding reference values used to order the
variables.
cvt:sos2 (sos2)
0/1*: Whether to honor SOS2 constraints for nonconvex piecewise-linear
terms, using suffixes .sos and .sosref provided by AMPL.
cvt:uenc:negctx:max (uenc:negctx:max, uenc:negctx)
If cvt:uenc:ratio applies, max number of constants in comparisons
x==const in negative context (equivalently, x!=const in positive
context) to skip UEnc(x). Default 1.
cvt:uenc:ratio (uenc:ratio)
Min ratio (ub-lb)/Nvalues to skip unary encoding for a variable x, where
Nvalues is the number of constants used in conditional comparisons
x==const. Instead, indicator constraints (or big-Ms) are used, if
uenc:negctx also applies. Default 0.
det:benders-enabled (benders-enabled)
0/1: whether benders detection should be enabled?
0 - Benders detection not enabled (default)
1 - Benders detection enabled.
det:benders-onlybinmaster (benders-onlybinmaster)
0/1: whether only decomposition with only binary variables in the master
are searched?
0 - Not only decomposition with only binary variables in the master are
searched (default)
1 - Only decomposition with only binary variables in the master are
searched.
det:benders-onlycontsubpr (benders-onlycontsubpr)
0/1: whether only decomposition with only continiuous variables in the
subproblems are searched?
0 - Not only decomposition with only continiuous variables in the
subproblems are searched (default)
1 - Only decomposition with only continiuous variables in the
subproblems are searched.
det:classification-enabled (classification-enabled)
0/1: whether classification should be enabled?
0 - Classification not enabled
1 - Classification enabled (default).
det:enabled
0/1: whether detection should be enabled?
0 - Detection not enabled
1 - Detection enabled (default).
det:maxnclassesperpartition (maxnclassesperpartition)
Maximum number of classes per partition (default: 9)
det:maxnclassesperpartitionforlargeprobs (maxnclassesperpartitionforlargeprobs)
Maximum number of classes per partition for large problems (nconss +
nvars >= 50000) (default: 5)
det:maxrounds
Maximum number of detection loop rounds (default: 1)
det:maxtime
Maximum detection time in seconds (default: 600)
det:origprob-classificationenabled (origprob-classificationenabled)
0/1: whether classification for the original problem should be enabled?
0 - Classification for the original problem not enabled
1 - Classification for the original problem enabled (default).
det:origprob-enabled (origprob-enabled)
0/1: whether detection for the original problem should be enabled?
0 - Detection for the original problem not enabled
1 - Detection for the original problem enabled (default).
det:postprocess (postprocess)
0/1: whether postprocessing of complete decompositions should be
enabled?
0 - Postprocessing of complete decompositions not enabled
1 - Postprocessing of complete decompositions enabled (default).
det:scoretype (scoretype)
Score calculation for comparing (partial) decompositions:
maxwhi - max white
border - border area
classi - classic
forswh - max foreseeing white
spfwh - ppc-max-white (default)
fawh - max foreseeing white with aggregation info
spfawh - ppc-max-white with aggregation info
bender - experimental benders score
strong - strong decomposition score
gcg:aggregation
0/1: whether identical blocks should be aggregated (only for
discretization approach)?
0 - Identical blocks should not be aggregated (only for discretization
approach)
1 - Identical blocks should be aggregated (only for discretization
approach) (default).
gcg:bliss-enabled (bliss-enabled)
0/1: whether bliss should be used to check for identical blocks?
0 - Bliss should not be used to check for identical blocks
1 - Bliss should be used to check for identical blocks (default).
gcg:discretization (discretization)
0/1: whether discretization (TRUE) or convexification (FALSE) approach
should be used?
0 - convexification approach should be used
1 - discretization approach should be used (default).
gcg:mipdiscretization (mipdiscretization)
0/1: whether discretization (TRUE) or convexification (FALSE) approach
should be used in mixed-integer programs?
0 - convexification approach should be used in mixed-integer programs
1 - discretization approach should be used in mixed-integer programs
(default).
gcg:mode (mode)
The decomposition mode that GCG will use:
0 - Dantzig-Wolfe (default)
1 - Benders' decomposition
2 - No decomposition
heu:settings
0/1/2/3: sets heuristics settings
0 - Sets heuristics default (default)
1 - Sets heuristics aggressive
2 - Sets heuristics fast
3 - Sets heuristics off.
lim:absgap (absgap)
Solving stops, if the absolute gap = |primalbound - dualbound| is below
the given value (default: 0.0)
lim:autorestartnodes
If solve exceeds this number of nodes for the first time, an automatic
restart is triggered (default: -1: no automatic restart)
lim:bestsol
Solving stops, if the given number of solution improvements were found
(default: -1: no limit)
lim:gap (gap)
Solving stops, if the relative gap = |primal -
dual|/MIN(|dual|,|primal|) is below the given value, the gap is
'Infinity', if primal and dual bound have opposite signs (default: 0.0)
lim:maxorigsol
Maximal number of solutions candidates to store in the solution storage
of the original problem (default: 10)
lim:maxsol
Maximal number of solutions to store in the solution storage (default:
100)
lim:memory (memory)
#maximal memory usage in MB; reported memory usage is lower than real
memory usage! (default: 8796093022207.0)
lim:nodes
Maximal number of nodes to process (default: -1: no limit)
lim:restarts
Solving stops, if the given number of restarts was triggered (default:
-1: no limit)
lim:solutions
Solving stops, if the given number of solutions were found (default: -1:
no limit)
lim:stallnodes
Solving stops, if the given number of nodes was processed since the last
improvement of the primal solution value (default: -1: no limit)
lim:time (timelim, timelimit, time_limit)
Limit on solve time (in seconds; default: 1e+20).
lim:totalnodes
Maximal number of total nodes (incl. restarts) to process (default: -1:
no limit)
lp:alwaysgetduals (alwaysgetfarkasduals, alwaysgetduals)
0/1: whether the Farkas duals should always be collected when an LP is
found to be infeasible
0 - The Farkas duals should not always be collected when an LP is found
to be infeasible (default)
1 - The Farkas duals should always be collected when an LP is found to
be infeasible.
lp:presolving
0/1: whether presolving of LP solver should be used
0 - Presolving of LP solver should not be used
1 - Presolving of LP solver should be used (default).
lp:pricing (pricing)
Pricing strategy:
l - lpi default (default)
a - auto
f - full pricing
p - partial
s - steepest edge pricing
q - quickstart steepest edge pricing
d - devex pricing
lp:solvedepth
Maximal depth for solving LP at the nodes (default: -1: no depth limit)
lp:solvefreq
Frequency for solving LP at the nodes (-1: never; 0: only root LP;
default: 0)
lp:threads
Number of threads used for solving the LP (default: 0: automatic)
mip:bestbound (bestbound, return_bound)
Whether to return suffix .bestbound for the best known MIP dual bound on
the objective value:
0 - No (default)
1 - Yes.
The suffix is on the objective and problem and is -Infinity for
minimization problems and +Infinity for maximization problems if there
are no integer variables or if a dual bound is not available.
mip:return_gap (return_mipgap)
Whether to return mipgap suffixes or include mipgap values (|objectve -
.bestbound|) in the solve_message: sum of
1 - Return .relmipgap suffix (relative to |obj|)
2 - Return .absmipgap suffix (absolute mipgap)
4 - Suppress mipgap values in solve_message.
Default = 0. The suffixes are on the objective and problem. Returned
suffix values are +Infinity if no integer-feasible solution has been
found, in which case no mipgap values are reported in the solve_message.
mip:round (round)
Whether to round integer variables to integral values before returning
the solution, and whether to report that the solver returned noninteger
values for integer values: sum of
1 ==> Round nonintegral integer variables
2 ==> Modify solve_result
4 ==> Modify solve_message
Default = 0. Modifications that were or would be made are reported in
solve_result and solve_message only if the maximum deviation from
integrality exceeded mip:round_reptol.
mip:round_reptol (round_reptol)
Tolerance for reporting rounding of integer variables to integer values;
see "mip:round". Default = 1e-9.
misc:allowstrongdualreds (allowstrongdualreds)
0/1: whether strong dual reductions should be allowed in propagation and
presolving
0 - Strong dual reductions should not be allowed in propagation and
presolving
1 - Strong dual reductions should be allowed in propagation and
presolving (default).
misc:allowweakdualreds (allowweakdualreds)
0/1: whether weak dual reductions should be allowed in propagation and
presolving
0 - Weak dual reductions should not be allowed in propagation and
presolving
1 - Weak dual reductions should be allowed in propagation and presolving
(default).
misc:scaleobj (scaleobj)
0/1: whether the objective function should be scaled so that it is
always integer
0 - The objective function should not be scaled so that it is always
integer
1 - The objective function should be scaled so that it is always integer
(default).
nod:childsel
Child selection rule:
d - down
u - up
p - pseudo costs
i - inference
l - lp value
r - root LP value difference
h - hybrid inference/root LP value difference (default)
num:checkfeastolfac (checkfeastolfac)
Feasibility tolerance factor; for checking the feasibility of the best
solution (default: 1.0)
num:dualfeastol (dualfeastol)
Feasibility tolerance for reduced costs in LP solution (default: 1e-07)
num:epsilon (epsilon)
Absolute values smaller than this are considered zero (default: 1e-09)
num:feastol (feastol)
Feasibility tolerance for constraints (default: 1e-06)
num:infinity (infinity)
Values larger than this are considered infinity (default: 1e+20)
num:lpfeastolfactor (lpfeastolfactor)
Factor w.r.t. primal feasibility tolerance that determines default (and
maximal) primal feasibility tolerance of LP solver (default: 1.0)
num:sumepsilon (sumepsilon)
Absolute values of sums smaller than this are considered (default:
1e-06)
obj:no (objno)
Objective to optimize:
0 - None
1 - First (default, if available)
2 - Second (if available), etc.
par:maxnthreads (maxnthreads)
Maximum number of threads used during parallel solve (default: 8)
par:minnthreads (minnthreads)
Minimum number of threads used during parallel solve (default: 1)
par:mode (mode)
0/1: Parallel optimisation mode
0 - Opportunistic
1 - Deterministic (default)
pre:abortfac (abortfac)
Abort presolve, if at most this fraction of the problem was changed in
last presolve round (default: 0.0008)
pre:clqtablefac (clqtablefac)
Limit on number of entries in clique table relative to number of problem
nonzeros (default: 2.0)
pre:donotaggr (donotaggr)
0/1: whether aggregation of variables should be forbidden
0 - Aggregation of variables should not be forbidden (default)
1 - Aggregation of variables should be forbidden.
pre:donotmultaggr (donotmultaggr)
0/1: whether multi-aggregation of variables should be forbidden
0 - Multi-aggregation of variables should not be forbidden (default)
1 - Multi-aggregation of variables should be forbidden.
pre:immrestartfac (immrestartfac)
Fraction of integer variables that were fixed in the root node
triggering an immediate restart with preprocessing (default: 0.1)
pre:maxrestarts
Maximal number of restarts (default: -1: unlimited)
pre:maxrounds
Maximal number of presolving rounds (default: -1: unlimited; 0: off)
pre:restartfac (restartfac)
Fraction of integer variables that were fixed in the root node
triggering a restart with preprocessing after root node evaluation
(default: 0.025)
pre:restartminred (restartminred)
Minimal fraction of integer variables removed after restart to allow for
an additional restart (default: 0.1)
pre:settings
0/1/2/3: sets presolvings settings
0 - Sets presolvings default (default)
1 - Sets presolvings aggressive
2 - Sets presolvings fast
3 - Sets presolvings off.
pre:subrestartfac (subrestartfac)
Fraction of integer variables that were globally fixed during the
solving process triggering a restart with preprocessing (default: 1.0)
pro:abortoncutoff
0/1: whether propagation should be aborted immediately (setting this to
0 could help conflict analysis to produce more conflict constraints)
0 - Propagation should not be aborted immediately
1 - Propagation should be aborted immediately (default).
pro:maxrounds
Maximal number of propagation rounds per node (-1: unlimited; 0: off;
default: 100)
pro:maxroundsroot
Maximal number of propagation rounds in the root node (-1: unlimited; 0:
off; default: 1000)
ran:lpseed (lpseed)
Random seed for LP solver, e.g. for perturbations in the simplex
(default: 0: LP default)
ran:permutationseed (permutationseed)
Seed value for permuting the problem after reading/transformation
(default: 0: no permutation)
ran:permuteconss (permuteconss)
0/1: whether the order of constraints should be permuted (depends on
permutationseed)?
0 - Order of constraints should not be permuted
1 - Order of constraints should be permuted (default).
ran:permutevars (permutevars)
0/1: whether the order of variables should be permuted (depends on
permutationseed)?
0 - Order of variables should not be permuted (default)
1 - Order of variables should be permuted.
ran:randomseedshift (randomseedshift)
Global shift of all random seeds in the plugins and the LP random seed
(default: 0)
sol:chk:fail (chk:fail, checkfail)
Fail on MP solution check violations, with solve result 150.
sol:chk:feastol (sol:chk:eps, chk:eps, chk:feastol)
Absolute tolerance to check objective values, variable and constraint
bounds. Default 1e-6.
sol:chk:feastolrel (sol:chk:epsrel, chk:epsrel, chk:feastolrel)
Relative tolerance to check objective values, variable and constraint
bounds. Default 1e-6.
sol:chk:infeas (chk:infeas, checkinfeas)
Check even infeasible solution condidates, whenever solver reports them.
sol:chk:inttol (sol:chk:inteps, sol:inteps, chk:inttol)
Solution checking tolerance for variables' integrality. Default 1e-5.
sol:chk:mode (solcheck, checkmode, chk:mode)
Solution checking mode. Sum of a subset of the following bits:
1 - Check variable bounds and integrality.
2 - Check original model constraints, as well as any non-linear
expression values reported by the solver.
4 - Check intermediate auxiliary constraints (i.e., those which were
reformulated further).
8 - Check final auxiliary constraints sent to solver.
16 - Check objective values.
32, 64, 128, 256, 512 - similar, but non-linear expressions are
recomputed (vs using their values reported by the solver.)
*Experimental.* This is an idealistic check, because it does not
consider possible tolerances applied by the solver when computing
expression values.
Default: 1+2+512.
sol:chk:prec (chk:prec, chk:precision)
AMPL solution_precision option when checking: number of significant
digits.
sol:chk:round (chk:round, chk:rnd)
AMPL solution_round option when checking: round to this number of
decimals after comma (before comma if negative.)
sol:count (countsolutions)
0*/1: Whether to count the number of solutions and return it in the
".nsol" problem suffix.
sol:stub (solstub, solutionstub)
Stub for solution files. If "solutionstub" is specified, found solutions
are written to files ("solutionstub & '1' & '.sol'") ... ("solutionstub
& Current.nsol & '.sol'"), where "Current.nsol" holds the number of
returned solutions. That is, file names are obtained by appending 1, 2,
... "Current.nsol" to "solutionstub".
tech:debug (debug)
0*/1: whether to assist testing & debugging, e.g., by outputting
auxiliary information.
tech:exportfile (writeprob, writemodel)
Specifies the name of a file where to export the model before solving
it. This file name can have extension ".lp", ".mps", etc. Default = ""
(don't export the model).
tech:logfile (logfile)
Log file name.
tech:optionfile (optionfile, option:file)
Name of an AMPL solver option file to read (surrounded by 'single' or
"double" quotes if the name contains blanks). Lines that start with #
are ignored. Otherwise, each nonempty line should contain "name=value",
e.g., "lim:iter=500".
tech:optionnativeread (optionnativeread, tech:param:read, param:read)
Filename of GCG parameter file (as path).The suffix on a parameter file
should be .set.
tech:outlev (outlev)
0*/1: Whether to write GCG log lines (chatter) to stdout and to file.
tech:outlev-native (outlev-native)
0*/1/2/3/4/5: Whether to write GCG log lines (chatter) to stdout and to
file (native output level of SCIP).
tech:timing (timing)
0*/1: Whether to display timings for the run.
tech:version (version)
Single-word phrase: report version details before solving the problem.
tech:wantsol (wantsol)
In a stand-alone invocation (no "-AMPL" on the command line), what
solution information to write. Sum of
1 - Write ".sol" file
2 - Primal variables to stdout
4 - Dual variables to stdout
8 - Suppress solution message.
tech:writegraph (writegraph, exportgraph)
File to export conversion graph. Format: JSON Lines.