1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831
|
From: James Cowgill <james410@cowgill.org.uk>
To: 749555@bugs.debian.org
Subject: Bug#749555: lp-solve: please provide man page for /usr/bin/lp_solve
Date: Thu, 03 Sep 2015 23:26:56 +0100
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=749555
--- a/lp_solve.1
+++ b/lp_solve.1
@@ -0,0 +1,821 @@
+.TH lp_solve 1 "September 2015"
+.SH NAME
+lp_solve \- a mixed integer linear programming (MILP) solver
+.SH SYNOPSIS
+.B lp_solve
+.RI [ options "] [" "input file" ]
+.SH DESCRIPTION
+.B lp_solve
+solves mixed integer linear programming problems. The program accepts models of
+problems in various different formats (including custom formats through XLIs)
+and attempts to solve them. There are a large number of different options
+which can be used to alter the solving process to improve performance or get
+more accurate results.
+
+.SS "lp file format"
+The lp file format is a simple, human readable file format for linear
+programming models. It is the default format used by
+.BR lp_solve .
+
+The format consists of series of statements, each ending in a semicolon. The
+first statement is always the objective function, followed by any number of
+constraints, followed by any number of declarations.
+
+The objective function consists of an expression to be optimized, optionally
+prefixed by
+.B max:
+or
+.B min:
+to specify which direction to optimize. The default is to maximize.
+
+Each constraint consists of a relational expression, optionally prefixed by a
+name followed by a colon. The expression must contain at least one relational operator
+.RB ( < , " <=" , " =", " =>", " >" ).
+You can specify ranges two operators at once (e.g. 2 < x < 4) instead of two
+separate constraints.
+
+Declarations can be one of:
+.TP
+\fBint \fIvar1 \fR[\fB, \fIvar2\fR]... ;
+Marks the variables as integers.
+.TP
+\fBbin \fIvar1 \fR[\fB, \fIvar2\fR]... ;
+Marks the variables as binary (can only take 0 or 1).
+.TP
+\fBsec \fIvar1 \fR[\fB, \fIvar2\fR]... ;
+Marks the variables as semi-continuous. These variables can always take the
+value 0 even if it is outside the variable's range.
+.TP
+\fBfree \fIvar1 \fR[\fB, \fIvar2\fR]... ;
+Marks the variables as free. These variables have no lower bound, unlike other
+variables which have a default lower bound of 0.
+.TP
+\fBsos \fR[\fIname\fB:\fR] \fIvar1 \fR[\fB, \fIvar2\fR]... [< \fIsostype\fR];
+Marks the variables as part of a special ordered set (SOS). Out of the variables
+listed, only one can be non-zero at once. If
+.I sostype
+is specified, then up to that number of variables can be non-zero as long as
+they are all adjacent to each other (e.g. for an
+.I sostype
+of 2,
+.I var1
+and
+.I var2
+could be non-zero, but not
+.I var1
+and
+.IR var3 ).
+.RE
+.PD
+
+Both C style comments
+.RB ( "/* " "comment here" " */" )
+and C++ style comments
+.RB ( "// " "comment until end of line)"
+can be used anywhere in the file.
+.SH OPTIONS
+.SS "General options"
+.TP
+.B \-wafter
+Writes the model after solving it (normally any model conversions are processed
+before solving). This is useful if presolving is used to simplify the model.
+.TP
+.B \-parse_only
+Parse the input model, but stop before solving it. This can be used to convert
+models between formats without solving them.
+.TP
+.BI \-timeout " sec"
+Timeout if no solution has be found after
+.I sec
+seconds.
+.TP
+.B \-nonames
+Ignore variable and constraint names in the input model (this is the same as
+combining both
+.BR \-norownames " " \-nocolnames ).
+.TP
+.B \-norownames
+Ignore constraint names in the input model.
+.TP
+.B \-nocolnames
+Ignore variable names in the input model.
+.TP
+.B \-min
+Minimizes the LP problem, overriding the setting in the input model.
+.TP
+.B \-max
+Maximizes the LP problem, overriding the setting in the input model.
+.SS "Built-in model languages"
+.TP
+.B \-lp
+Read the model from a file in LP format. This the the default.
+.TP
+.B \-mps
+Read the model from a file in 'fixed width' MPS format.
+.TP
+.B \-fmps
+Read the model from a file in 'free' MPS format. This is the same as
+.B \-mps
+except that fields are separated by a number of blanks instead of starting at
+fixed columns.
+.TP
+.BI \-wlp " filename"
+Converts the input model to LP format and writes it to the a file.
+.TP
+.BI \-wmps " filename"
+Converts the input model to 'fixed width' MPS format and writes it to a file.
+.TP
+.BI \-wfmps " filename"
+Converts the input model to 'free' MPS format and writes it to a file.
+.SS "External language interface (XLI)"
+.TP
+.BI \-rxli " xliname " "" filename
+Read the model using an external language interface (XLI).
+.I xliname
+specifies the path to the shared library containing the XLI, and is searched for
+according to the normal library search rules in
+.BR ld.so (8)
+with the addition that the library filename is prefixed with
+.I lib
+and suffixed with
+.I .so
+if necessary.
+.TP
+.BI \-rxlidata " datafilename"
+Provides a file containing extra data used by the reading XLI library.
+.TP
+.BI \-rxliopt " options"
+Extra options to pass to the reading XLI library.
+.TP
+.BI \-wxli " xliname " "" filename
+Converts the input model to the format used by an XLI and writes it to a file.
+The format of
+.I xliname
+is described in
+.B \-rxli
+above.
+.TP
+.BI \-wxliopt " options"
+Extra options to pass to the writing XLI library.
+.TP
+.BI \-wxlisol " xliname " "" filename
+Writes the model's solution to a file using an XLI library. The format of
+.I xliname
+is described in
+.B \-rxli
+above.
+.TP
+.BI \-wxlisolopt " options"
+Extra options to pass to the solution writing XLI library.
+.SS "Printing, verbosity and debugging"
+.TP
+.B \-h
+Print a usage message and exit.
+.TP
+.BI \-S level
+Solution detail. Each
+.I level
+builds on top of the previous one and adds more detail to the solution.
+
+.PD 0
+.RS 7
+.TP 5
+.B 0
+print nothing
+.TP
+.B 1
+objective value only
+.TP
+.B 2
+variables (default level)
+.TP
+.B 3
+constraints
+.TP
+.B 4
+duals
+.TP
+.B 5
+lp model
+.TP
+.B 6
+scales
+.TP
+.B 7
+lp tableau
+.RE
+.PD
+.TP
+.B \-time
+Print CPU time to parse input and to calculate solution.
+.TP
+\fB\-v\fR[\fIlevel\fR]
+Verbosity level. Controls the level of messages printed about the operation of
+the program.
+
+.PD 0
+.RS 7
+.TP 5
+.B 0
+neutral
+.TP
+.B 1
+critical
+.TP
+.B 2
+severe
+.TP
+.B 3
+important (default when
+.B \-v
+is provided with no
+.IR level )
+.TP
+.B 4
+normal (default when no
+.B \-v
+option is given)
+.TP
+.B 5
+detailed
+.TP
+.B 6
+full
+.RE
+.PD
+.TP
+.B \-t
+Trace pivot selection.
+.TP
+.B \-d
+Debug mode. All intermediate results and branch-and-bound decisions are printed.
+.TP
+.B \-R
+Report information while solving the model.
+.TP
+.BI \-Db " filename"
+Create a dump of internal model variables before solving the model to given
+file.
+.TP
+.BI \-Da " filename"
+Create a dump of internal model variables after solving the model, to given
+file.
+.TP
+.BR \-i [ a ]
+Print all intermediate valid solutions. If
+.B a
+is used, prints only non-zero values. Can give useful solutions even if the
+total run time is too long.
+
+.SS "Parameter files"
+.TP
+.BI \-rpar " filename"
+Reads a list of parameters from a file. The parameters file can specify the
+same options as you can by passing arguments to
+.B lp_solve
+but can be more convenient if you need to change lots of settings. The format
+of the parameters file is in the INI format often used by Windows applications.
+All parameters are read from key-value pairs in a single section. The default
+section is
+.BR [Default] .
+.TP
+\fB\-rparopt\fR '\fB\-h \fIheader\fR'
+Sets options for reading the parameters file. Currently the only option is
+.B \-h
+which changes the section within the file to read the parameters from.
+.TP
+.BI \-wpar " filename"
+Writes a parameters file based on the arguments given to
+.BR lp_solve .
+.TP
+\fB\-wparopt\fR '\fB\-h \fIheader\fR'
+Sets options for writing the parameter file. Currently the only option is
+.B \-h
+which changes the section within the file to write the parameters to.
+.SS "Basis files"
+.TP
+.BI \-rbas " filename"
+Reads a basis file which is used as the starting point for solving the LP
+model. Basis files must be in MPS basis format.
+.TP
+.BI \-gbas " filename"
+Computes a basis from a list of initial guesses for each variable. Each line of
+the given file is in the format:
+.IB variable : value
+where
+.I value
+is the initial guess for
+.IR variable .
+.TP
+.BI \-wbas " filename"
+After solving the LP model, writes a basis file in MPS format. This file can be
+read back using
+.BR \-rbas .
+
+.RS 7
+.PD 0
+Available options:
+.TP 5
+.B 0
+no crash basis (default).
+.TP
+.B 1
+most feasible basis.
+.TP
+.B 2
+least degenerate basis.
+.PD
+.RE
+
+.SS "Integer and branch and bound options"
+.TP
+.B \-noint
+Ignore integer restrictions on variables.
+.TP
+.B \-f
+Stops the branch and bound algorithm immediately after finding its first
+solution.
+.TP
+.BI \-o " bound"
+Stops the branch and bound algorithm immediately after finding a solution whose
+objective function has a value greater than this bound.
+.TP
+.BI \-b " bound"
+Sets a lower bound for the objective function. The branch and bound algorithm
+will immediately reject all solutions lower than this value.
+.TP
+.BI \-depth " limit"
+Sets the depth limit for the branch and bound algorithm. A value of 0 disables
+depth limiting. A negative value sets a relative depth limit based on the number
+of variables in the input model. The default limit is \-50.
+.TP
+.BI \-e " tolerance"
+Sets the tolerance which is used to determine whether a floating point
+number is an integer. A number has to be within this value of an integer to be
+considered one. The default value is 1e-7.
+.TP
+\fB\-g \fItolerance\fR, \fB\-ga \fItolerance\fR
+Sets the absolute MIP gap used by the branch and bound algorithm. This
+tolerance is the difference between the best-found solution yet and the current
+solution. If the difference is smaller than this tolerance then the solution
+(and all the sub-solutions) is rejected. This can result in faster solving
+times, but results in a solution which is not the perfect solution. The default
+value is 1e-11.
+.TP
+.BI \-gr " tolerance"
+Sets the relative MIP gap used by the branch and bound algorithm. This is
+similar to the absolute tolerance except that the difference is scaled by the
+best found solution before being compared to this tolerance. The default value
+is 1e-11.
+.TP
+.BR \-c ", " \-cc
+During branch and bound, take the ceiling branch first.
+.TP
+.B \-cf
+During branch and bound, take the floor branch first.
+.TP
+.B \-ca
+During branch and bound, allow the algorithm to decide which branch to take.
+This is the default.
+.TP
+.BI \-n " solution"
+If the branch and bound algorithm produces multiple solutions with the same
+objective function value, prints the nth solution. Solution numbers start at 1.
+.TP
+.BI \-B rule
+Sets a branch and bound rule to use. The default rules are:
+.B \-B5 \-Bg \-Bd \-Bc
+
+.RS 7
+.PD 0
+These rules are mutually exclusive:
+.TP 5
+.B 0
+select lowest indexed non-integer column.
+.TP
+.B 1
+selection based on distance from the current bounds.
+.TP
+.B 2
+selection based on the largest current bound.
+.TP
+.B 3
+selection based on largest fractional value.
+.TP
+.B 4
+simple, unweighted pseudo-cost of a variable.
+.TP
+.B 5
+extended pseudo-costing strategy based on minimizing the number of integer
+infeasibilities.
+.TP
+.B 6
+extended pseudo-costing strategy based on maximizing the normal pseudo-cost
+divided by the number of infeasibilities.
+.PD
+.PP
+.PD 0
+The above rules can be combined with any of the following:
+.TP 5
+.B w
+WeightReverse - select by criterion minimum (worst), rather than criterion
+maximum (best).
+.TP
+.B b
+BranchReverse - when
+.B \-ca
+is selected, choose the opposite direction.
+.TP
+.B g
+Greedy
+.TP
+.B p
+PseudoCost - toggles between weighting based on pseudocost or objective
+function value.
+.TP
+.B f
+DepthFirst - select the node that has already been selected before the most
+number of times.
+.TP
+.B r
+Randomize - ddds a randomization factor to the score for any node candidate.
+.TP
+.B G
+GubMode
+.TP
+.B d
+Dynamic - when
+.B \-Bf
+is selected, switch off this mode when a first solution is found.
+.TP
+.B s
+RestartMode - enables regular restarts of pseudocost value calculations.
+.TP
+.B B
+BreadthFirst - select the node that has been selected before the fewest number
+of times or not at all.
+.TP
+.B o
+AutoOrder - order variables to improve branch-and-bound performance.
+.TP
+.B c
+ReducedCostFixing - do bound tightening during B&B based of reduced cost info.
+.TP
+.B i
+StringInit - initialize pseudo-costs by strong branching.
+.PD
+.RE
+.SS "Simplex algorithm options"
+.TP
+.BR \-prim ", " \-simplexpp
+Prefer the primal simplex method for both phase 1 and phase 2.
+.TP
+.BR \-dual ", " \-simplexdd
+Prefer the dual simplex method for both phase 1 and phase 2.
+.TP
+.BR \-simplexdp
+Prefer the dual simplex method for phase 1 and primal method for phase 2. This
+is the default.
+.TP
+.BR \-simplexpd
+Prefer the primal simplex method for phase 1 and dual method for phase 2.
+.TP
+.BI \-bfp " filename"
+Sets the basis factorization package to use.
+.I filename
+refers to a shared library loaded in the same way
+.B \-rxli
+loads libraries.
+.TP
+.BI \-o value
+Sets whether the objective function is stored in the top row of the constraint
+matrix or in separate storage.
+.B \-o0
+places the matrix in separate storage, and
+.B \-o1
+(the default) places it in the constraint matrix.
+.TP
+.BI \-C mode
+Sets the basis crash mode. When base crash is enabled, a heuristic 'crash
+procedure' is executed before the first simplex iteration to quickly choose
+a basis matrix that has fewer artificial variables. This procedure tends to
+reduce the number of iterations to optimality since a number of iterations are
+skipped.
+.TP
+.BI \-r " value"
+Sets the maximum number of pivots between a re-inversion of the matrix. For
+stability reasons, lp_solve re-inverts the matrix on regular times. The default
+is 250 for the LUSOL bfp and 42 for the other BFPs.
+.TP
+.BI \-trej " value"
+Sets the value that is used as a tolerance pivot element to determine whether a
+value should be considered as 0. The default is 2e-7.
+.TP
+.BI \-epsd " value"
+Sets the value that is used as a tolerance for reduced costs to determine
+whether a value should be considered as 0. The default is 1e-9.
+.TP
+.BI \-epsb " value"
+Sets the value that is used as a tolerance for the Right Hand Side (RHS) to
+determine whether a value should be considered as 0.
+The default is 1e-10.
+.TP
+.BI \-epsel " value"
+Sets the value that is used as a tolerance for rounding values to zero. The
+default is 1e-12.
+.TP
+.BI \-epsp " value"
+Sets the value that is used as perturbation scalar for degenerative problems.
+The default is 1e-5.
+.TP
+.BI \-improve level
+Sets the iterative improvement level.
+.I level
+is a number which can be created by combining the different levels below. The
+default is
+.BR \-improve6 .
+
+.RS 7
+.PD 0
+.TP 5
+.B 0
+none.
+.TP
+.B 1
+running accuracy measurement of solved equations on Bx=r.
+.TP
+.B 2
+improve initial dual feasibility by bound flips.
+.TP
+.B 4
+low-cost accuracy monitoring in the dual.
+.TP
+.B 8
+check for primal/dual feasibility at the node level.
+.PD
+.RE
+.TP
+.BI \-piv rule
+Sets a simplex pivot rule or mode to use. The default rules are:
+.B \-piv2 \-piva
+
+.RS 7
+.PD 0
+These rules are mutually exclusive:
+.TP 5
+.B 0
+select first.
+.TP
+.B 1
+select according to Dantzig.
+.TP
+.B 2
+select Devex pricing from Paula Harris.
+.TP
+.B 3
+select steepest edge.
+.PD
+.PP
+.PD 0
+The above rules can be combined with any of the following modes:
+.TP 5
+.B f
+PrimalFallback - when steepest edge
+.RB ( \-priv3 )
+is selected, fallback to Devex in primal.
+.TP
+.B m
+Multiple - preliminary implementation of the multiple pricing scheme. This
+means that attractive candidate entering columns from one iteration may be used
+in the subsequent iteration, avoiding full updating of reduced costs. In the
+current implementation, lp_solve only reuses the 2nd best entering column
+alternative
+.TP
+.B a
+Adaptive - temporarily use alternative strategy if cycling is detected.
+.TP
+.B r
+Randomize - adds a small randomization effect to the selected pricer.
+.TP
+.B ll
+LoopLeft - scan entering/leaving columns left rather than right.
+.TP
+.B la
+LoopAlternate - scan entering/leaving columns alternating left/right.
+.TP
+.B h
+HarrisTwoPass - use Harris' primal pivot logic rather than the default.
+.TP
+.B t
+TrueNormInit - use true norms for Devex and Steepest Edge initializations.
+.PD
+.RE
+.TP
+.BI \-degen option
+.\"Strictly speaking 'option' is optional, but using -degen appears to be a
+.\"no-op since the modes it enables are already enabled automatically.
+Specifies if special handling must be done to reduce degeneracy/cycling while
+solving. The default options are:
+.BR \-degenf " " \-degens .
+
+.RS 7
+.PD 0
+Available options:
+.TP 5
+.B c
+ColumnCheck
+.TP
+.B d
+Dynamic
+.TP
+.B f
+FixedVars
+.TP
+.B s
+Stalling
+.TP
+.B n
+NumFailure
+.TP
+.B l
+Lostfeas
+.TP
+.B i
+Infeasible
+.TP
+.B b
+DuringBB
+.TP
+.B r
+RHSPerturb
+.TP
+.B p
+BoundFlip
+.PD
+.RE
+.SS "Other solving options"
+.TP
+\fB\-presolve\fR[\fIoption\fR]
+Enables a presolve option. These options attempt to speed up calculations by
+simplifying the model before solving it. Any of these options can be combined
+together. If
+.I option
+is not specified, row and column presolving is enabled (as if
+.B \-presolverow \-presolvecol
+had been used). The default not to presolve.
+
+.RS 7
+.PD 0
+Available options:
+.TP 5
+.B row
+Rows - presolve rows.
+.TP
+.B col
+Cols - presolve columns.
+.TP
+.B l
+LinDep - eliminate linearly dependent rows.
+.TP
+.B s
+Sos - convert constraints to SOSes (only SOS type 1 handled).
+.TP
+.B r
+ReduceMip - if the phase 1 solution process finds that a constraint is redundant
+then this constraint is deleted.
+.TP
+.B k
+Knapsack - simplification of knapsack-type constraints through addition of an
+extra variable, which also helps bound the objective function.
+.TP
+.B q
+ElimEQ2 - direct substitution of one variable in 2-element equality constraints;
+this requires changes to the constraint matrix.
+.TP
+.B m
+MergeRows - merges neighboring >= or <= constraints when the vectors are
+otherwise relatively identical into a single ranged constraint.
+.TP
+.B fd
+ColFixDual - variable fixing and removal based on considering signs of the
+associated dual constraint.
+.TP
+.B bnd
+Bounds - does bound tightening based on full-row constraint information.
+.TP
+.B d
+Duals - presolve duals.
+.TP
+.B f
+ImpliedFree - identify implied free variables (releasing their explicit bounds).
+.TP
+.B slk
+ImpliedSlk- converts qualifying equalities to inequalities by converting a
+column singleton variable to slack.
+.TP
+.B g
+ReduceGCD - reduce (tighten) coefficients in integer models based on GCD
+argument.
+.TP
+.B b
+ProbeFix - attempt to fix binary variables at one of their bounds.
+.TP
+.B c
+ProbeReduce - attempt to reduce coefficients in binary models.
+.TP
+.B rowd
+RowDominate - identify and delete qualifying constraints that are dominated by
+others, also fixes variables at a bound.
+.TP
+.B cold
+ColDominate - deletes variables (mainly binary), that are dominated by others
+(only one can be non-zero).
+.PD
+.RE
+.TP
+\fB\-s\fR[\fImode\fR] [\fIscalelimit\fR]
+Sets the scaling algorithm and/or scaling limit to use.
+.I scalelimit
+can only be specified when using the mutually exclusive integer modes. Omitting
+.I mode
+is the equivalent of using
+.BR \-s4 .
+The default algorithm and limit is:
+.B \-s1 5 \-si \-se
+
+.RS 7
+.PD 0
+These rules are mutually exclusive:
+.TP 5
+.B 0
+no scaling.
+.TP
+.B 1
+geometric scaling.
+.TP
+.B 2
+Curtis-Reid scaling.
+.TP
+.B 3
+scale to convergence using largest absolute value.
+.TP
+.B 4
+numerical range-based scaling.
+.TP
+.B 5
+scale to convergence using logarithmic mean of all values.
+.TP
+.B 6
+scale based on the simple numerical range.
+.TP
+.B 7
+scale quadratic.
+.PD
+.PP
+.PD 0
+The above rules can be combined with any of the following:
+.TP 5
+.B p
+also do power scaling.
+.TP
+.B i
+also scale integer variables.
+.TP
+.B e
+ensure no scaled number is outside the range -1..1.
+.PD
+.RE
+.SH EXAMPLE
+.SS "Model solving"
+The following example shows a model being solved by
+.BR lp_solve .
+The first line contains the objective function to be maximized and the last 3
+lines contain the constraints.
+
+.nf
+$ \fBcat example.lp\fR
+max: 143 x + 60 y;
+
+120 x + 210 y <= 15000;
+110 x + 30 y <= 4000;
+x + y <= 75;
+
+$ \fBlp_solve -S3 example.lp\fR
+Value of objective function: 6315.63
+
+Actual values of the variables:
+x 21.875
+y 53.125
+
+Actual values of the constraints:
+R1 13781.2
+R2 4000
+R3 75
+.fi
+.SS "Format conversion"
+This example converts a file in lp format to mps format.
+
+$
+.B lp_solve -parse_only -lp example.lp -wmps example.mps
+.SH "SEE ALSO"
+.BR ld.so (8)
|