Search found 1037 matches

by bussieck
8 months ago
Forum: Solvers
Topic: BARON .how to use .EquaClass option?
Replies: 1
Views: 4992

Re: BARON .how to use .EquaClass option?

Since you haven't gotten any answer, I suggest to share the model to make this reproducible. -Michael
by bussieck
8 months ago
Forum: Bugs
Topic: Studio runs when first opened but then gives an error
Replies: 2
Views: 4570

Re: Studio runs when first opened but then gives an error

Rob, Studio adds a couple of parameters to the GAMS command line. Obviously, the last one pf="." is wrong. pf points to a GAMS parameter file with more command line options. Studio recently introduced the "GAMS Parameter File" feature, see https://www.gams.com/latest/docs/T_STUDI...
by bussieck
8 months ago
Forum: Modeling
Topic: How to accelerate variable multiplication calculation.
Replies: 1
Views: 3334

Re: How to accelerate variable multiplication calculation.

The steps that takes long is the extraction of the Q matrix from the non-linear instructions. There are basically two algorithms to do this and we default to the one (ThreePass) that is is fast in most cases. You might benefit from using the other algorithm (DoubleForward). We even have a "conc...
by bussieck
9 months ago
Forum: Announcements
Topic: GAMS Distribution 44.0.0 (Beta Release)
Replies: 0
Views: 18567

GAMS Distribution 44.0.0 (Beta Release)

Hi GAMS Users, We would like to announce the availability of the GAMS Beta Release 44.0.0. Please consult the beta release notes for details about new features and updated solvers: https://www.gams.com/beta/docs/RN_44.html If you are interested in experimenting with this beta release, please downloa...
by bussieck
9 months ago
Forum: Syntax
Topic: Table Error 225
Replies: 1
Views: 4827

Re: Table Error 225

I guess some mix with TAB and spaces to get the table aligned. I would remove TABs whereever possible and go with spaces only. If that's not the case you should upload (not copy paste because too much can change) the model file so one can reproduce.

-Michael
by bussieck
9 months ago
Forum: Syntax
Topic: dimension error
Replies: 5
Views: 5745

Re: dimension error

GAMS has domain checking for a good reason. If you circumvent this by declaring your equations without a domain, then the system can't help you with your own mistakes. Your equation res_dem is defined over (s,t) but you access this (to assign to price) as res_dem.M(t,s). If you would have declared w...
by bussieck
9 months ago
Forum: Solvers
Topic: Saving Time of when best solution was found & log and lst files with unique names
Replies: 3
Views: 5433

Re: Saving Time of when best solution was found & log and lst files with unique names

I don't think there is a general "time saving" option in BARON. Even though your best incumbent is found quickly BARON still needs to move the dual bound to reach the optimality tolerance. That can sometimes take a long time, even if you already know and hold the optimal solution. If you c...
by bussieck
9 months ago
Forum: Solvers
Topic: BARON numLoc & local search preprocessing
Replies: 2
Views: 5023

Re: BARON numLoc & local search preprocessing

In GAMS, solver option are passed via a solver option file. You can create this inside a GAMS model via $on/offEcho > baron.opt (at compile time) or via put files at execution time. The use of solver option files is explained here: https://www.gams.com/latest/docs/UG_SolverUsage.html#BASIC_USAGE_SOL...
by bussieck
9 months ago
Forum: Solvers
Topic: BARON: terminates solver when solution does not improve after x amount of time
Replies: 1
Views: 4804

Re: BARON: terminates solver when solution does not improve after x amount of time

Not with BARON. Some solvers allow you "user defined" termination via callbacks and this is made available to GAMS users, see e.g. https://www.gams.com/latest/docs/S_CPLE ... ipstopexpr.

-Michael
by bussieck
9 months ago
Forum: Modeling
Topic: Problem infeasible or unbounded
Replies: 1
Views: 2646

Re: Problem infeasible or unbounded

Nothing. You should use code boxes for your model and output though. You model in unbounded. Just limit the objective variable (e.g. z.lo = -1e6;) and see what "run away" variables you get. Check your application why this should not happen, turn your insight into constraints/model logic. -...