LGO - Revision history

Ango: /* Description of Inputs */

2015-06-23T08:56:19Z

Description of Inputs

← Older revision		Revision as of 03:56, 23 June 2015
Line 127:		Line 127:
	\|''PriLevOpt''\|\|Print level in MEX interface.		\|''PriLevOpt''\|\|Print level in MEX interface.
	\|-valign="top"		\|-valign="top"
	\|''optParam~~''\|\|Structure with optimization parameters~~.		\|''optParam.MaxFunc''\|\|Maximum number of function evaluations. (Prob.LGO.options.g_maxfct)
	~~\|-valign="top"~~
	~~\|''~~MaxFunc''\|\|Maximum number of function evaluations. (Prob.LGO.options.g_maxfct)
	\|-valign="top"		\|-valign="top"
	\|''LGO.PrintFile''\|\|Name of LGO Print file. Amount and type of printing determined by PriLevOpt.		\|''LGO.PrintFile''\|\|Name of LGO Print file. Amount and type of printing determined by PriLevOpt.

Ango: /* Description of Inputs */

2015-06-23T08:52:02Z

Description of Inputs

← Older revision		Revision as of 03:52, 23 June 2015
Line 129:		Line 129:
	\|''optParam''\|\|Structure with optimization parameters.		\|''optParam''\|\|Structure with optimization parameters.
	\|-valign="top"		\|-valign="top"
	\|''MaxFunc''\|\|Maximum number of function evaluations. (Prob.LGO.options.~~g maxfct~~)		\|''MaxFunc''\|\|Maximum number of function evaluations. (Prob.LGO.options.g_maxfct)
	\|-valign="top"		\|-valign="top"
	\|''LGO.PrintFile''\|\|Name of LGO Print file. Amount and type of printing determined by PriLevOpt.		\|''LGO.PrintFile''\|\|Name of LGO Print file. Amount and type of printing determined by PriLevOpt.
	\|-valign="top"		\|-valign="top"
	\|''LGO.SummFile''\|\|Name of LGO summary file.		\|''LGO.SummFile''\|\|Name of LGO summary file.

Ango: /* Setting options using the Prob.LGO structure */ Correct upper-case G_ to lower case.

2015-06-23T08:18:26Z

Setting options using the Prob.LGO structure: Correct upper-case G_ to lower case.

← Older revision		Revision as of 03:18, 23 June 2015
Line 60:		Line 60:
	<pre>		<pre>
	Prob = glcAssign(...) % Setup problem, see help glcAssign for more information		Prob = glcAssign(...) % Setup problem, see help glcAssign for more information
	Prob.LGO.options.~~G_maxfct~~ = 10000; % Setting the maximum number of global search		Prob.LGO.options.g_maxfct = 10000; % Setting the maximum number of global search
	% phase model function evaluations.		% phase model function evaluations.
	</pre>		</pre>

Ango: /* Description of Prob.LGO.options */

2015-06-23T08:09:58Z

Description of Prob.LGO.options

← Older revision		Revision as of 03:09, 23 June 2015
Line 208:		Line 208:

	{\| class="wikitable" border="1"		{\| class="wikitable" border="1"
	!Option\|\|Description\|\|		!Option\|\|Description\|\|Default
	\|-valign="top"		\|-valign="top"
	\|colspan="3"\|'''General options'''		\|colspan="3"\|'''General options'''
Line 228:		Line 228:
	\|colspan="3"\|'''Limits and Tolerances'''		\|colspan="3"\|'''Limits and Tolerances'''
	\|-valign="top"		\|-valign="top"
	\|''g_maxfct''\|\|Maximum number of merit function evaluations before termination of global -1 search phase (BB, GARS, or MS). The value of -1 uses 500(nvars+ncons), where nvars is the number of variables and ncons the number of constraints.		\|''g_maxfct''\|\|Maximum number of merit function evaluations before termination of global search phase (BB, GARS, or MS). The value of -1 uses 500(nvars+ncons), where nvars is the number of variables and ncons the number of constraints.
			The difficulty of global optimization models varies greatly: for difficult models, g_maxfct can be increased as needed.
			\|\|-1

	~~The difficulty of global optimization models varies greatly: for difficult models, g_maxfct can be increased as needed.~~
	\|-valign="top"		\|-valign="top"
	\|l_maxfct\|\|Maximal number of function evaluations in local search phase.		\|''l_maxfct'' \|\| Maximal number of function evaluations in local search phase.

	\|-valign="top"		\|-valign="top"
Line 239:		Line 240:
	\|''penmult''\|\|Constraint penalty multiplier. Global search phase merit function is defined as objective + the violated constraints weighted by penmult.\|\|100		\|''penmult''\|\|Constraint penalty multiplier. Global search phase merit function is defined as objective + the violated constraints weighted by penmult.\|\|100
	\|-valign="top"		\|-valign="top"
	\|''g_target''\|\|Global search termination criterion parameter (acceptability threshold). The global search phase (BB, GARS, or MS) ends, if an overall merit function value found in the global search phase is less than ~~acc tr~~. If a good estimate is known, then its usage may results in a considerably faster search.\|\|-1.0E+10		\|''g_target''\|\|Global search termination criterion parameter (acceptability threshold). The global search phase (BB, GARS, or MS) ends, if an overall merit function value found in the global search phase is less than g_target. If a good estimate is known, then its usage may results in a considerably faster search.\|\|-1.0E+10
	\|-valign="top"		\|-valign="top"
	\|''l_target''\|\|Target merit function value in local search phase.\|\|-1.0E+10		\|''l_target''\|\|Target merit function value in local search phase.\|\|-1.0E+10
Line 251:		Line 252:
	\|''irngs''\|\|Random number seed. An arbitrary integer value can be used instead of the 0 default.		\|''irngs''\|\|Random number seed. An arbitrary integer value can be used instead of the 0 default.

	Note that the local search operational mode (~~Opmode~~ 0) is the fastest, and that it will work for convex, as well as for many mildly non-convex models. If the model has a highly non-convex (multiextremal) structure, then at least one of the global search modes should be used. In difficult or complex models, it may be a good idea to apply all three global search modes, to verify the global solution, or perhaps to find alternative good solutions. Usually, ~~Opmode~~ 3 is the safest (and slowest), since it applies several local searches; ~~Opmodes~~ 1 and 2 launch only a single local search from the best point found in the global search phase.		Note that the local search operational mode (opmode 0) is the fastest, and that it will work for convex, as well as for many mildly non-convex models. If the model has a highly non-convex (multiextremal) structure, then at least one of the global search modes should be used. In difficult or complex models, it may be a good idea to apply all three global search modes, to verify the global solution, or perhaps to find alternative good solutions. Usually, opmode 3 is the safest (and slowest), since it applies several local searches; opmodes 1 and 2 launch only a single local search from the best point found in the global search phase.

	Note that if model-specific information is known (more sensible target objective/merit function value, tolerances, tighter variable bounds), then such information should always be used, since it may help to solve the model far more efficiently than by using 'blind' defaults		Note that if model-specific information is known (more sensible target objective/merit function value, tolerances, tighter variable bounds), then such information should always be used, since it may help to solve the model far more efficiently than by using 'blind' defaults
	\|}		\|}

Ango: /* Overview */

2015-04-08T10:14:13Z

Overview

← Older revision		Revision as of 05:14, 8 April 2015
Line 3:		Line 3:
	===Overview===		===Overview===

	Welcome to the TOMLAB /LGO User's Guide. TOMLAB /LGO includes the LGO solver from ~~Pint´er~~ Consulting Services and an interface to MATLAB, by MathWorks'.		Welcome to the TOMLAB /LGO User's Guide. TOMLAB /LGO includes the LGO solver from Pintér Consulting Services and an interface to MATLAB, by MathWorks.

	The ''Lipschitz-Continuous Global Optimizer ''(LGO) solver suite serves for the analysis and global solution of general nonlinear programming (NLP) models. The LGO solver system has been developed and gradually extended for more than a decade and it now incorporates a suite of robust and efficient global and local nonlinear solvers. It can also handle smaller LP models. LGO is documented elsewhere in detail: see for example, ~~Pint´er~~ (1996, 2001, 2003). \[1, 2, 3\] TOMLAB /LGO integrates the following global scope algorithms:		The ''Lipschitz-Continuous Global Optimizer ''(LGO) solver suite serves for the analysis and global solution of general nonlinear programming (NLP) models. The LGO solver system has been developed and gradually extended for more than a decade and it now incorporates a suite of robust and efficient global and local nonlinear solvers. It can also handle smaller LP models. LGO is documented elsewhere in detail: see for example, Pintér (1996, 2001, 2003). \[1, 2, 3\] TOMLAB /LGO integrates the following global scope algorithms:

	*Branch-and-bound (adaptive partition and sampling) based global search (BB)		*Branch-and-bound (adaptive partition and sampling) based global search (BB)

Ango: /* Description of Prob.LGO.options */

2015-04-02T09:45:03Z

Description of Prob.LGO.options

← Older revision		Revision as of 04:45, 2 April 2015
Line 239:		Line 239:
	\|''penmult''\|\|Constraint penalty multiplier. Global search phase merit function is defined as objective + the violated constraints weighted by penmult.\|\|100		\|''penmult''\|\|Constraint penalty multiplier. Global search phase merit function is defined as objective + the violated constraints weighted by penmult.\|\|100
	\|-valign="top"		\|-valign="top"
	\|''~~acc_tr~~''\|\|Global search termination criterion parameter (acceptability threshold). The global search phase (BB, GARS, or MS) ends, if an overall merit function value found in the global search phase is less than acc tr. If a good estimate is known, then its usage may results in a considerably faster search.\|\|-1.0E+10		\|''g_target''\|\|Global search termination criterion parameter (acceptability threshold). The global search phase (BB, GARS, or MS) ends, if an overall merit function value found in the global search phase is less than acc tr. If a good estimate is known, then its usage may results in a considerably faster search.\|\|-1.0E+10
	\|-valign="top"		\|-valign="top"
	\|''~~fct_trg~~''\|\|Target merit function value in local search phase.\|\|-1.0E+10		\|''l_target''\|\|Target merit function value in local search phase.\|\|-1.0E+10
	\|-valign="top"		\|-valign="top"
	\|''fi_tol''\|\|Local search merit function improvement tolerance.\|\|1.0E-06		\|''fi_tol''\|\|Local search merit function improvement tolerance.\|\|1.0E-06

Ango: /* Description of Prob.LGO.options */ l_maxfct

2015-03-25T16:10:04Z

Description of Prob.LGO.options: l_maxfct

← Older revision		Revision as of 11:10, 25 March 2015
Line 207:		Line 207:
	User options for the TOMLAB /LGO solver. The following fields are used:		User options for the TOMLAB /LGO solver. The following fields are used:

	{\|class="wikitable"		{\| class="wikitable" border="1"
	!Option\|\|Description\|\|		!Option\|\|Description\|\|
	\|-valign="top"		\|-valign="top"
Line 230:		Line 230:
	\|''g_maxfct''\|\|Maximum number of merit function evaluations before termination of global -1 search phase (BB, GARS, or MS). The value of -1 uses 500(nvars+ncons), where nvars is the number of variables and ncons the number of constraints.		\|''g_maxfct''\|\|Maximum number of merit function evaluations before termination of global -1 search phase (BB, GARS, or MS). The value of -1 uses 500(nvars+ncons), where nvars is the number of variables and ncons the number of constraints.

	The difficulty of global optimization models varies greatly: for difficult models, g_maxfct can be increased to as needed.		The difficulty of global optimization models varies greatly: for difficult models, g_maxfct can be increased as needed.
			\|-valign="top"
			\|l_maxfct\|\|Maximal number of function evaluations in local search phase.

	\|-valign="top"		\|-valign="top"
	\|''max_nosuc''\|\|Maximum number of merit function evaluations in global search phase (BB, GARS, or MS) where no improvement is made. Global search terminates upon reaching this limit. The value of -1 uses 100(nvars+ncons), where nvars is the number of variables and ncons the number of constraints. For difficult models, this value can also be increased as deemed necessary.\|\|-1		\|''max_nosuc''\|\|Maximum number of merit function evaluations in global search phase (BB, GARS, or MS) where no improvement is made. Global search terminates upon reaching this limit. The value of -1 uses 100(nvars+ncons), where nvars is the number of variables and ncons the number of constraints. For difficult models, this value can also be increased as deemed necessary.\|\|-1

Elias at 15:23, 17 December 2011

2011-12-17T15:23:03Z

← Older revision		Revision as of 10:23, 17 December 2011
Line 91:		Line 91:

	<math>\begin{array}{ccccccl}		<math>\begin{array}{ccccccl}
	\min\limits_{x} & ~~\multicolumn{5}{l}{~~f(x)} & \\		\min\limits_{x} & f(x) & \\
	\\		\\
	s/t & x_L & \leq & x & \leq & x_U & \\		s/t & x_L & \leq & x & \leq & x_U & \\

Elias: /* Description of Inputs */

2011-10-19T10:28:19Z

Description of Inputs

← Older revision		Revision as of 05:28, 19 October 2011
Line 135:		Line 135:
	\|''LGO.SummFile''\|\|Name of LGO summary file.		\|''LGO.SummFile''\|\|Name of LGO summary file.
	\|-valign="top"		\|-valign="top"
	\|''LGO.options''\|\|Structure with special fields for LGO optimization parameters. See ~~Table 4~~.		\|''LGO.options''\|\|Structure with special fields for LGO optimization parameters. See [[#Description of Prob.LGO.options]]
	\|}		\|}

Elias: /* Setting options using the Prob.LGO structure */

2011-10-19T10:26:50Z

Setting options using the Prob.LGO structure

← Older revision		Revision as of 05:26, 19 October 2011
Line 66:		Line 66:
	In the cases where a solver specific parameter has a corresponding TOMLAB general parameter, the latter is used only if the user has not given the solver specific parameter.		In the cases where a solver specific parameter has a corresponding TOMLAB general parameter, the latter is used only if the user has not given the solver specific parameter.

	A complete description of the available LGO parameters can be found in ~~Section 5.1~~.		A complete description of the available LGO parameters can be found in [[#lgoTL]].

	==LGO Test Examples==		==LGO Test Examples==