Nabble - AMPL

[AMPL 2975] Re: how to turn Cplex presolve off in AMPL

2009-11-08T14:50:27Z

umm, it doesn't work. I have tried to put crossover=0, presolve=0,
prereduce=0, aggregate=0, aggfill=0 and dependency=0 in the options.
But the output
is still the same,

Here is the output:
5255 variables, all linear
2626 constraints, all linear; 15689 nonzeros
1 linear objective; 5 nonzeros.

ILOG CPLEX 10.000, licensed to "university-edinburgh", options: e m b
q use=5
CPLEX 10.0.0: baropt
crossover=0
presolve=0
prereduce=0
aggregate=0
aggfill=0
dependency=0
prestats=1
lpdisplay=1
Presolve time = 0.00 sec.
LP Presolve eliminated 72 rows and 2697 columns.
Reduced LP has 2554 rows, 2558 columns, and 7751 nonzeros.
Presolve time = 0.01 sec.
CPLEX 10.0.0: optimal solution; objective 5.494632119
7 barrier iterations; no basis.

Thank you very much indeed.

Naiyuan

On 11月8日, 下午3時55分, Paul <ru...@...> wrote:
> I'm not sure, but try adding prereduce=0 to your solver options. (You
> may also want aggregate=0 in the options.)
>
> /Paul

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

[AMPL 2974] Re: Constraint formulation

2009-11-08T11:07:17Z

Hi Bob and Paul,

thank you both for the replies. I should have expressed my question
with more clarity - sorry for that.

Thank you Paul for your answer and explanation, which answers my
question perfectly.

Is there a list available of piece-meal linearisations such as this?

Thank you so much.
Alison

On Nov 5, 11:06 pm, Paul <ru...@...> wrote:

> I agree with Bob that your question could use some refinement, but I
> think I know what you're getting at, so I'll take a shot.
>
> I don't think you can do exactly what you want, but you may come
> close. Let M be a sufficiently large positive constant and epsilon a
> sufficiently small (but strictly positive) constant, and let Y (a
> variable) be the output of your first IF. Introduce a new binary
> variable Z together with the following constraints:
>
> (1) epsilon*Z <= A <= M*Z
> (2) B - M*Z <= Y <= B + M*Z
> (3) -M*(1-Z) <= Y <= M*(1-Z).
>
> A <= 0 forces Z = 0 in the left part of (1), which forces Y = B in (2)
> and makes (3) vacuous. A >= epsilon forces Z = 1 in the right side of
> (1), which forces Y = 0 in (3) and makes (2) vacuous. Besides
> introducing a binary variable, there is one catch: 0 < A < epsilon has
> been excised from the feasible region. So this won't work if
> arbitrarily small but positive values of A are legitimate contenders
> for an optimal solution.
>
> Incidentally, I used the same M in every constraint because I'm too
> lazy to type subscripts, but you can (and probably should) use a
> distinct M (the tightest value you can find that is guaranteed not to
> cut off an optimal solution) in each constraint.
>
> /Paul
>
> On Nov 1, 12:09 pm, AlisonT <alisontho...@...> wrote:
>
>
>
> > Hi,
>
> > Can someone please advise me how I should formulate a constraint that
> > represents the following:
>
> > IF( "decision variable A" > 0, then 0, else "decision variable B")
>
> > I also need to formulate a similar constraint, where "decision
> > variable B" is 1 and is therefore a binary decision variable i.e.
> > IF( "decision variable C" > 0, then 0, else 1)
>
> > Thank you so much.
> > Alison- Hide quoted text -
>
> - Show quoted text -

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

[AMPL 2973] Re: Error Executing AMPL

2009-11-08T08:05:44Z

On Nov 7, 7:39 pm, Diana <dixig...@...> wrote:
> Hello, when I try to solve the modelo, I get a error message:
>
> file -
> line 46
> offset 1667
> cumpl_labor is not defined

You define a parameter named cumpl_lab but refer to a parameter named
cumpl_labor. Since there are no references to cumpl_lab other than in
its declaration, I suspect you meant them to be the same thing.

You also need to insert a statement
data;
before the start of the data declarations (around where you have the
comment #DATA). Otherwise AMPL thinks the data statements are actually
trying to redefine your sets and parameters.

> file -
> line 292
> offset 7619
> expected :=
> file -
> line 303
> offset 8007
> cap_horas is already defined
> Error executing "solve" command:
> error processing constraint capa_horas[5047]:
> no value for tiempo_ej[5047,409]

I don't get this far, because you did not provide data for the
parameter cap_uso.

/Paul

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

[AMPL 2972] Re: how to turn Cplex presolve off in AMPL

2009-11-08T07:55:13Z

I'm not sure, but try adding prereduce=0 to your solver options. (You
may also want aggregate=0 in the options.)

/Paul

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

[AMPL 2971] Re: Different results on Windows vs. Linux

2009-11-08T07:47:43Z

On Nov 6, 10:33 pm, Din <dinakar.g...@...> wrote:

> Hi Paul,
> Thanks for your reply.
> Yes, I am using the same data input/output format for both windows and
> Linux - I am using a model file with 3 different models, a run file,
> and a data file. I am making parameter assignments from one model's
> solve as an input to another during the iterations using AMPL's 'let'
> command. All problems are LP's.
>
> Would it be of any help if I used Callable libraries/Concert
> technology instead of making AMPL calls?
>
> Thanks again
> Din
>

Possible but doubtful. You might try writing one of the intermediate
problems where Windows and Linux results were different to an MPS
file. Read it into CPLEX interactive solver, solve it, and see what
the condition number of the basis matrix is. If the matrix is ill-
conditioned (admittedly a relative judgment), you might investigate
whether you can reformulate the problem to have a nicer condition
matrix. (Besides the condition number of the final basis, you might
get signals of ill-conditioning if CPLEX refactors the basis
frequently or issues warnings along the way).

You're using the variables values from a solution to one problem to
modify the parameters of the next problem? When you said that
intermediate problems were reported feasible under Windows but
infeasible under Linux, did you confirm that the models being tested
were identical? If any of your LPs have multiple optimal solutions,
it's possible that you get different solutions on Windows and Linux
(both optimal, but different variable values) due to different pivot
sequences. That would lead to a divergence in the subsequent problems
(different coefficient matrices) and any cross-platform comparability
would be lost.

/Paul

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

[AMPL 2968] Error Executing AMPL

2009-11-07T16:39:11Z

Hello, when I try to solve the modelo, I get a error message:

file -
line 46
offset 1667
cumpl_labor is not defined
file -
line 292
offset 7619
expected :=
file -
line 303
offset 8007
cap_horas is already defined
Error executing "solve" command:
error processing constraint capa_horas[5047]:
no value for tiempo_ej[5047,409]

Please, Why is this?

My code is the following:

#SETS
set TRAC;
set LAB;

#PARAMETERS

param cfijo{i in TRAC, j in LAB} >= 0;
param cost_var{i in TRAC, j in LAB} >= 0;
param cap_horas{i in TRAC} >= 0;
param cap_uso{i in TRAC} >= 1;
param cumpl_lab{j in LAB} = 1;
param tiempo_ej{i in TRAC,j in LAB} >= 0;

# VARIABLE
var w{i in TRAC, j in LAB} binary;
minimize costo_total: # ($/week)

#OBJECTIVE
sum{i in TRAC, j in LAB} (cfijo[i,j]*w[i,j])
+ sum{ i in TRAC, j in LAB} (cost_var[i,j]*w[i,j]);

# CONSTRAINTS

subject to capa_horas {i in TRAC}:
sum{j in LAB} (w[i,j]*tiempo_ej[i,j]) <= cap_horas[i];

subject to capa_uso {i in TRAC}:
sum{j in LAB} (w[i,j]) >= cap_uso[i];

subject to cumplm_lab {j in LAB}:
sum{i in TRAC} (w[i,j]) = cumpl_labor[j];

#DATA

# SETS

set TRAC:= 5047 5048 5049 5050 5051 5052 5053 5200

5203 5206 5221 5223 5224 5225 5226 5227

5228 6010 6011 7056 7058 7059 7061 7062

7078 8000 8007;

set LAB:= 409 411 412 419 420 421 440 442 444 445

446 491 531 131 842 140 208 130 120 142;

# PARAMETERS

param cfijo default 9999999: 409 411 412 419 420 :=

5047 . . . . .
5048 . . . . .
5049 . 737022 . . .
5050 . 737022 . . .
5051 . 737022 . . .
5052 . 737022 . . .
5053 . . . . .
5200 737022 . 737022 737022 737022
5203 737022 . 737022 737022 737022
5206 737022 . 737022 737022 737022
5221 737022 . 737022 737022 737022
5223 737022 . 737022 737022 737022
5224 737022 . 737022 737022 737022
5225 . 737022 . . .
5226 737022 . . . .
5227 737022 . 737022 737022 737022
5228 737022 . 737022 737022 737022
6010 . . . . .
6011 . . . . .
7056 737022 . 737022 737022 737022
7058 737022 . 737022 737022 737022
7059 737022 . 737022 737022 737022
7061 . . 737022 737022 737022
7062 737022 . 737022 737022 737022
7078 . . . . .
8000 . . . . .
8007 . . . . .

: 421 440 442 444 445 :=
5047 737022 737022 737022 . .
5048 737022 737022 737022 . .
5049 . 737022 737022 . .
5050 . 737022 737022 . .
5051 . 737022 737022 . .
5052 . 737022 737022 . .
5053 . . . . .
5200 . . . 737022 737022
5203 . . 737022 737022 737022
5206 . . 737022 737022 737022
5221 . . . 737022 737022
5223 . . 737022 737022 737022
5224 . . 737022 737022 737022
5225 . . 737022 . .
5226 . . 737022 . .
5227 . 737022 737022 737022 737022
5228 . . . 737022 737022
6010 . . . . .
6011 . . . . .
7056 737022 737022 737022 737022 737022
7058 737022 737022 737022 737022 737022
7059 737022 737022 737022 737022 737022
7061 737022 737022 737022 737022 737022
7062 737022 737022 737022 737022 737022
7078 737022 . . 737022 737022
8000 737022 . . . .
8007 737022 . . . .

: 446 491 531 131 842 :=

5047 . . . . 737022
5048 . . . . .
5049 . . . . 737022
5050 . . . . .
5051 . . . 737022 .
5052 . 737022 . . .
5053 . 737022 . 737022 .
5200 . . . . .
5203 737022 . . . .
5206 737022 . . . .
5221 . . . . .
5223 737022 . . . .
5224 737022 . . . .
5225 737022 . . . .
5226 737022 737022 . . .
5227 737022 . . . .
5228 . . . . .
6010 . . . 737022 .
6011 . . . 737022 .
7056 . . 737022 737022 737022
7058 . . 737022 737022 737022
7059 . . 737022 737022 737022
7061 . . . 737022 737022
7062 . . . 737022 737022
7078 . . . 737022 737022
8000 . . . . .
8007 . . . 737022 737022

:140 208 130 120 142 :=

5047 737022 737022 737022 . .
5048 . 737022 737022 737022 .
5049 737022 . . 737022 .
5050 . 737022 . . .
5051 . . . . .
5052 . . . . .
5053 . . . . 737022
5200 . . . . .
5203 . . . . .
5206 . . . . .
5221 . . . . .
5223 . . . . .
5224 . . . . .
5225 . . . . .
5226 . . . . .
5227 . . . . .
5228 . . . . .
6010 . . . 737022 .
6011 . . . 737022 .
7056 737022 . 737022 . 737022
7058 737022 . 737022 . 737022
7059 737022 . 737022 . 737022
7061 737022 . 737022 . 737022
7062 737022 . 737022 . 737022
7078 737022 . . 737022 .
8000 . . . 737022 737022
8007 . . 737022 737022 737022 ;

param cost_var default 9999999: 409 411 412 419 420 :=

5047 . . . . .
5048 . . . . .
5049 . 8782294 . . .
5050 . 12291821 . . .
5051 . 6679969 . . .
5052 . 7425955 . . .
5053 . . . . .
5200 15119759 . 23059627 0 6587407
5203 16288183 . 24841627 0 7096469
5206 19427234 . 29629094 0 8464096
5221 13218890 . 20160550 0 5759232
5223 13724626 . 20931865 0 5979572
5224 10585576 . 16144398 0 4611944
5225 . 12851311 . . .
5226 11928391 . . . .
5227 18747106 . 28591809 0 8167776
5228 16933433 . 25825718 0 7377591
6010 . . . . .
6011 . . . . .
7056 8928855 . 13617680 0 3890141
7058 7498842 . 11436724 0 3267111
7059 8457996 . 12899561 0 3684996
7061 . . 14043233 0 4011708
7062 9888009 . 15080517 0 4308027
7078 . . . . .
8000 . . . . .
8007 . . . . .

: 421 440 442 444 445 :=

5047 0 13447230 0 . .
5048 0 13828171 0 . .
5049 . 19732762 0 . .
5050 . 27618248 0 . .
5051 . 15009090 0 . .
5052 . 16685232 0 . .
5053 . . . . .
5200 . . . 0 0
5203 . . 0 0 0
5206 . . 0 0 0
5221 . . . 0 0
5223 . . 0 0 0
5224 . . 0 0 0
5225 . . 0 . .
5226 . . 0 . .
5227 . 40951195 0 0 0
5228 . . . 0 0
6010 . . . . .
6011 . . . . .
7056 0 19504197 0 0 0
7058 0 16380479 0 0 0
7059 0 18475656 0 0 0
7061 0 20113703 0 0 0
7062 0 21599374 0 0 0
7078 0 . . 0 0
8000 0 . . . .
8007 0 . . . .

: 446 491 531 131 842 :=

5047 . . . . 0
5048 . . . . .
5049 . . . . 0
5050 . . . . .
5051 . . . 0 .
5052 . 43407965 . . .
5053 . 79283953 . 0 .
5200 . . . . .
5203 7020981 . . . .
5206 8374060 . . . .
5221 . . . . .
5223 5915965 . . . .
5224 4562886 . . . .
5225 5697969 . . . .
5226 5141703 67787780 . . .
5227 8080893 . . . .
5228 . . . . .
6010 . . . 0 .
6011 . . . 0 .
7056 . . 0 0 0
7058 . . 0 0 0
7059 . . 0 0 0
7061 . . . 0 0
7062 . . . 0 0
7078 . . . 0 0
8000 . . . . .
8007 . . . 0 0

: 140 208 130 120 142 :=

5047 0 0 0 . .
5048 . 0 0 0 .
5049 0 . . 0 .
5050 . 0 . . .
5051 . . . . .
5052 . . . . .
5053 . . . . 0
5200 . . . . .
5203 . . . . .
5206 . . . . .
5221 . . . . .
5223 . . . . .
5224 . . . . .
5225 . . . . .
5226 . . . . .
5227 . . . . .
5228 . . . . .
6010 . . . 0 .
6011 . . . 0 .
7056 0 . 0 . 0
7058 0 . 0 . 0
7059 0 . 0 . 0
7061 0 . 0 . 0
7062 0 . 0 . 0
7078 0 . . 0 .
8000 . . . 0 0
8007 . . 0 0 0 ;

param tiempo_ej: 409 411 412 419 420 :=

5047 0 0 0 0 0
5048 0 0 0 0 0
5049 0 129.01 0 0 0
5050 0 129.01 0 0 0
5051 0 129.01 0 0 0
5052 0 129.01 0 0 0
5053 0 0 0 0 0
5200 278.67 0 101.1925 0 75.1595
5203 278.67 0 101.1925 0 75.1595
5206 278.67 0 101.1925 0 75.1595
5221 278.67 0 101.1925 0 75.1595
5223 278.67 0 101.1925 0 75.1595
5224 278.67 0 101.1925 0 75.1595
5225 0 129.01 0 0 0
5226 278.67 0 0 0 0
5227 278.67 0 101.1925 0 75.1595
5228 278.67 0 101.1925 0 75.1595
6010 0 0 0 0 0
6011 0 0 0 0 0
7056 278.67 0 101.1925 0 75.1595
7058 278.67 0 101.1925 0 75.1595
7059 278.67 0 101.1925 0 75.1595
7061 0 0 101.1925 0 75.1595
7062 278.67 0 101.1925 0 75.1595
7078 0 0 0 0 0
8000 0 0 0 0 0
8007 0 0 0 0 0

: 421 440 442 444 445 :=

5047 0 144.935 0 0 0
5048 0 144.935 0 0 0
5049 0 144.935 0 0 0
5050 0 144.935 0 0 0
5051 0 144.935 0 0 0
5052 0 144.935 0 0 0
5053 0 0 0 0 0
5200 0 0 0 0 0
5203 0 0 0 0 0
5206 0 0 0 0 0
5221 0 0 0 0 0
5223 0 0 0 0 0
5224 0 0 0 0 0
5225 0 0 0 0 0
5226 0 0 0 0 0
5227 0 144.935 0 0 0
5228 0 0 0 0 0
6010 0 0 0 0 0
6011 0 0 0 0 0
7056 0 144.935 0 0 0
7058 0 144.935 0 0 0
7059 0 144.935 0 0 0
7061 0 144.935 0 0 0
7062 0 144.935 0 0 0
7078 0 0 0 0 0
8000 0 0 0 0 0
8007 0 0 0 0 0

: 446 491 531 131 842 :=

5047 0 0 0 0 0
5048 0 0 0 0 0
5049 0 0 0 0 0
5050 0 0 0 0 0
5051 0 0 0 0 0
5052 0 527.884 0 0 0
5053 0 527.884 0 0 0
5200 0 0 0 0 0
5203 28.6 0 0 0 0
5206 28.6 0 0 0 0
5221 0 0 0 0 0
5223 28.6 0 0 0 0
5224 28.6 0 0 0 0
5225 28.6 0 0 0 0
5226 28.6 527.884 0 0 0
5227 28.6 0 0 0 0
5228 0 0 0 0 0
6010 0 0 0 0 0
6011 0 0 0 0 0
7056 0 0 0 0 0
7058 0 0 0 0 0
7059 0 0 0 0 0
7061 0 0 0 0 0
7062 0 0 0 0 0
7078 0 0 0 0 0
8000 0 0 0 0 0
8007 0 0 0 0 0

: 140 208 130 120 142 :=

5047 0 0 0 0 0
5048 0 0 0 0 0
5049 0 0 0 0 0
5050 0 0 0 0 0
5051 0 0 0 0 0
5052 0 0 0 0 0
5053 0 0 0 0 0
5200 0 0 0 0 0
5203 0 0 0 0 0
5206 0 0 0 0 0
5221 0 0 0 0 0
5223 0 0 0 0 0
5224 0 0 0 0 0
5225 0 0 0 0 0
5226 0 0 0 0 0
5227 0 0 0 0 0
5228 0 0 0 0 0
6010 0 0 0 0 0
6011 0 0 0 0 0
7056 0 0 0 0 0
7058 0 0 0 0 0
7059 0 0 0 0 0
7061 0 0 0 0 0
7062 0 0 0 0 0
7078 0 0 0 0 0
8000 0 0 0 0 0
8007 0 0 0 0 0 ;

param cap_horas:=

5047 100
5048 100
5049 100
5050 100
5051 100
5052 100
5053 100
5200 100
5203 100
5206 100
5221 100
5223 100
5224 100
5225 100
5226 100
5227 100
5228 100
6010 100
6011 100
7056 100
7058 100
7059 100
7061 100
7062 100
7078 100
8000 100
8007 100;

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

[AMPL 2970] how to turn Cplex presolve off in AMPL

2009-11-07T15:56:34Z

Hello,

I am using AMPL v10 together with CPLEX 11.
I meet the following problem:
I want to use CPLEX barrier algorithm without presolve in AMPL. I used
the commands as below

option solver cplexamp;
option presolve 0; <--- turn off AMPL presolve.
option show_stats 1;
option cplex_options ' baropt presolve=0 crossover=0 prestats=1
lpdisplay=1 ';
solve;

Then, I get the following output:

5255 variables, all linear
2626 constraints, all linear; 15689 nonzeros
1 linear objective; 5 nonzeros.

ILOG CPLEX 10.000, licensed to "university-edinburgh", options: e m b
q use=5
CPLEX 10.0.0: baropt
presolve=0
crossover=0
prestats=1
lpdisplay=1
Reduced presolve eliminated 72 rows and 2697 columns.
Reduced LP has 2554 rows, 2558 columns, and 7751 nonzeros.
Presolve time = 0.01 sec.
CPLEX 10.0.0: optimal solution; objective 5.494632119
7 barrier iterations; no basis.

Hence, it seems presolve still exists in CPLEX. I wonder how to turn
off it

Thank you very much in advance.

best regards,

Naiyuan

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

[AMPL 2969] Re: Different results on Windows vs. Linux

2009-11-06T19:33:22Z

Hi Paul,
Thanks for your reply.
Yes, I am using the same data input/output format for both windows and
Linux - I am using a model file with 3 different models, a run file,
and a data file. I am making parameter assignments from one model's
solve as an input to another during the iterations using AMPL's 'let'
command. All problems are LP's.

Would it be of any help if I used Callable libraries/Concert
technology instead of making AMPL calls?

Thanks again
Din

On Nov 6, 6:21 pm, Paul <ru...@...> wrote:

> Din wrote:
> > Hi,
> > I am running ILOG AMPL 10.000 on both my local machine (Windows) and a
> > Linux Server. I am using a commands file to test a cutting plane
> > scheme. When I run the same command file with the same model and data
> > I am getting different results on Windows and Linux: In windows my
> > run finishes in 19 iterations, while in Linux, the run finishes in 35.
> > I am also testing the feasibility of an intermediate problem during
> > the iterative process. AMPL-CPLEX on Windows reports feasibility on
> > all instances, where as on Linux there are some cases when the problem
> > is reported infeasible.
>
> > I am using the following options in my commands file.
> > option omit_zero_rows 1;
> > option cplex_options 'feasibility 1.0e-9 presolve 0';
> > option display_eps 0.000001;
> > #option presolve_eps 1e-8;
> > option presolve 0;
>
> > Could anyone shed some light?
>
> Different compilers on different hardware platforms introduce some
> "minor" differences in floating-point arithmetic, enough to alter the
> branching sequence in a MIP and the pivot sequence in an LP in some
> cases. I think that Emilie Danna at CPLEX has documented this and
> presented about it at conferences. That might account for some
> differences in iteration count.
>
> As far as feasibility v. infeasibility, if an LP/MIP has an ill-
> conditioned constraint matrix, funny things can happen when it comes
> to feasibility. Combine that with the platform issue above and it
> might explain why you're seeing different conclusions about
> feasibility on the two platforms (although I can't be sure that's it).
>
> I take it the problem data is entered the same way on both platforms
> (calculated from the same script, read from the same file, ...)?
> Solving a problem using data generated within the AMPL model/script
> (or by a calling program) and solving the same problem after writing
> out the data/model (say to an MPS file) and reading it into a solver
> can produce entirely different results if the matrix is ill-
> conditioned, because write to file/read back causes a loss of
> precision (at least if the file being written is text rather than
> binary).
>
> /Paul

RE: [AMPL 2967] Impossible deduced bounds ...

2009-11-06T16:53:15Z

Notice that the TxRng2 constraints are like

subject to TxRng2[2,6,1,3,1,2]:

57*S[2,6,1,3] >= 16.165;

Since the S-variables are binary, these force variables S[2,6,1,3], S[3,6,1,3], ..., S[7,6,1,3] to be 1. Setting them all to 1 in the constraint

subject to VisFlsA[6]:

0.002*S[1,6,1,3] + 0.001*S[2,6,1,3] + 0.001*S[3,6,1,3] +

0.003*S[4,6,1,3] + 0.002*S[5,6,1,3] + 0.002*S[6,6,1,3] +

0.002*S[7,6,1,3] <= 0.005;

and simplifying gives S[1,6,1,3] <= -3, hence the message about the impossible deduced bound "upper = -3". There is not feasible solution possible since also as a binary variable S[1,6,1,3] >= 0.

Bob Fourer

4er@...

From: Daniel Fokum [mailto:fokumdt@...]
Sent: Wednesday, November 04, 2009 5:15 PM
To: ampl@...
Subject: [AMPL 2962] Impossible deduced bounds ...

Hello,

When I try to solve the attached model I get a message that states:

presolve, variable S[1,6,1,3]:
    impossible deduced bounds: lower = 0, upper = -3

However, S is a binary variable. I have executed a sequence of expand commands as shown below to determine why the bounds are impossible, unfortunately I am not able to resolve this error.

ampl: expand S[1,6,1,3];
Coefficients of S[1,6,1,3]:
    VisDtct[6]               0.9
    VisFlsA[6]               0.002
    VisTimely[6]             0.96
    TxRng2[1,6,1,3,1,2]     52
    costMetric           12586.8

ampl: expand VisDtct[6];
subject to VisDtct[6]:
    0.9*S[1,6,1,3] + 0.93*S[2,6,1,3] + 0.97*S[3,6,1,3] + 0.905*S[4,6,1,3]
    + 0.915*S[5,6,1,3] + 0.925*S[6,6,1,3] + 0.895*S[7,6,1,3] >= 0.89;

ampl: expand VisFlsA[6];
subject to VisFlsA[6]:
    0.002*S[1,6,1,3] + 0.001*S[2,6,1,3] + 0.001*S[3,6,1,3] +
    0.003*S[4,6,1,3] + 0.002*S[5,6,1,3] + 0.002*S[6,6,1,3] +
    0.002*S[7,6,1,3] <= 0.005;

ampl: expand VisTimely[6];
subject to VisTimely[6]:
    0.96*S[1,6,1,3] + 0.98*S[2,6,1,3] + 0.97*S[3,6,1,3] + 0.955*S[4,6,1,3]
    + 0.965*S[5,6,1,3] + 0.95*S[6,6,1,3] + 0.96*S[7,6,1,3] >= 0.91;

ampl: expand TxRng2[1,6,1,3,1,2];
subject to TxRng2[1,6,1,3,1,2]:
    52*S[1,6,1,3] >= 16.165;

Thanks in advance for any pointers.

Daniel

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

[AMPL 2966] Re: Impossible deduced bounds ...

2009-11-06T15:49:56Z

On Nov 4, 6:14 pm, Daniel Fokum <foku...@...> wrote:

> Hello,
>
> When I try to solve the attached model I get a message that states:
>
> presolve, variable S[1,6,1,3]:
> impossible deduced bounds: lower = 0, upper = -3
>
> However, S is a binary variable. I have executed a sequence of expand
> commands as shown below to determine why the bounds are impossible,
> unfortunately I am not able to resolve this error.
>
> ampl: expand S[1,6,1,3];
> Coefficients of S[1,6,1,3]:
> VisDtct[6] 0.9
> VisFlsA[6] 0.002
> VisTimely[6] 0.96
> TxRng2[1,6,1,3,1,2] 52
> costMetric 12586.8
>
> ampl: expand VisDtct[6];
> subject to VisDtct[6]:
> 0.9*S[1,6,1,3] + 0.93*S[2,6,1,3] + 0.97*S[3,6,1,3] + 0.905*S[4,6,1,3]
> + 0.915*S[5,6,1,3] + 0.925*S[6,6,1,3] + 0.895*S[7,6,1,3] >= 0.89;
>
> ampl: expand VisFlsA[6];
> subject to VisFlsA[6]:
> 0.002*S[1,6,1,3] + 0.001*S[2,6,1,3] + 0.001*S[3,6,1,3] +
> 0.003*S[4,6,1,3] + 0.002*S[5,6,1,3] + 0.002*S[6,6,1,3] +
> 0.002*S[7,6,1,3] <= 0.005;
>
> ampl: expand VisTimely[6];
> subject to VisTimely[6]:
> 0.96*S[1,6,1,3] + 0.98*S[2,6,1,3] + 0.97*S[3,6,1,3] + 0.955*S[4,6,1,3]
> + 0.965*S[5,6,1,3] + 0.95*S[6,6,1,3] + 0.96*S[7,6,1,3] >= 0.91;
>
> ampl: expand TxRng2[1,6,1,3,1,2];
> subject to TxRng2[1,6,1,3,1,2]:
> 52*S[1,6,1,3] >= 16.165;
>
> Thanks in advance for any pointers.
>
> Daniel
>
> SnsrAssignRMtd.mod
> 8KViewDownload
>
> SnsrAssignRMtd7.dat
> 3KViewDownload

The LP relaxation is feasible, but with the binary restrictions the
model is infeasible. If you are convinced that the model should have
a feasible binary solution, about the only suggestion I can make is to
concoct one (it does not need a nice objective value, it just has to
be feasible), plug the values in for S, and see which constraints are
violated.

Another possibility, assuming you have the CPLEX interactive
optimizer, would be to write the model out from AMPL to an MPS file,
read it into CPLEX, and see if the CPLEX conflict refiner can point
you to a set of constraints that collectively make the problem
infeasible when S is binary.

/Paul

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

[AMPL 2965] Re: Different results on Windows vs. Linux

2009-11-06T15:21:27Z

Din wrote:

> Hi,
> I am running ILOG AMPL 10.000 on both my local machine (Windows) and a
> Linux Server. I am using a commands file to test a cutting plane
> scheme. When I run the same command file with the same model and data
> I am getting different results on Windows and Linux: In windows my
> run finishes in 19 iterations, while in Linux, the run finishes in 35.
> I am also testing the feasibility of an intermediate problem during
> the iterative process. AMPL-CPLEX on Windows reports feasibility on
> all instances, where as on Linux there are some cases when the problem
> is reported infeasible.
>
> I am using the following options in my commands file.
> option omit_zero_rows 1;
> option cplex_options 'feasibility 1.0e-9 presolve 0';
> option display_eps 0.000001;
> #option presolve_eps 1e-8;
> option presolve 0;
>
> Could anyone shed some light?
>

Different compilers on different hardware platforms introduce some
"minor" differences in floating-point arithmetic, enough to alter the
branching sequence in a MIP and the pivot sequence in an LP in some
cases. I think that Emilie Danna at CPLEX has documented this and
presented about it at conferences. That might account for some
differences in iteration count.

As far as feasibility v. infeasibility, if an LP/MIP has an ill-
conditioned constraint matrix, funny things can happen when it comes
to feasibility. Combine that with the platform issue above and it
might explain why you're seeing different conclusions about
feasibility on the two platforms (although I can't be sure that's it).

I take it the problem data is entered the same way on both platforms
(calculated from the same script, read from the same file, ...)?
Solving a problem using data generated within the AMPL model/script
(or by a calling program) and solving the same problem after writing
out the data/model (say to an MPS file) and reading it into a solver
can produce entirely different results if the matrix is ill-
conditioned, because write to file/read back causes a loss of
precision (at least if the file being written is text rather than
binary).

/Paul

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

[AMPL 2964] Different results on Windows vs. Linux

2009-11-06T13:09:45Z

Hi,
I am running ILOG AMPL 10.000 on both my local machine (Windows) and a
Linux Server. I am using a commands file to test a cutting plane
scheme. When I run the same command file with the same model and data
I am getting different results on Windows and Linux: In windows my
run finishes in 19 iterations, while in Linux, the run finishes in 35.
I am also testing the feasibility of an intermediate problem during
the iterative process. AMPL-CPLEX on Windows reports feasibility on
all instances, where as on Linux there are some cases when the problem
is reported infeasible.

I am using the following options in my commands file.
option omit_zero_rows 1;
option cplex_options 'feasibility 1.0e-9 presolve 0';
option display_eps 0.000001;
#option presolve_eps 1e-8;
option presolve 0;

Could anyone shed some light?

Thanks!
Din

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

Re: [AMPL 2963] expanding defined variables

2009-11-06T09:59:19Z

Thank you for the response.

Unfortunately it is still not what I have been looking for. They still
do not appear in the .fix file.

Are you suggesting that I should also expand the "defined variable"
explicit constraints and then copy the corresponding part of the
constraint to the .fix file with an appropriate script?

Is there another way to avoid writing a script for that?

Any help is appreciated.

Ali

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

RE: [AMPL 2961] expanding defined variables

2009-11-05T20:15:34Z

This is true, the definitions of defined variables are not expanded. To
export the problem using "expand" or "solexpand" you would have to replace
these definitions by explicit constraints of the form <var> = <defining
expression>.

Bob Fourer
4er@...

> -----Original Message-----
> From: Ali Baharev [mailto:ali.baharev@...]
> Sent: Thursday, November 05, 2009 10:53 AM
> To: ampl@...
> Subject: [AMPL 2957] expanding defined variables
>
> I would like to transform my benchmarks into a form that can be used
> by other researchers not using AMPL. (In my highly biased opinion they
> should be using AMPL :) )
>
> I would like to expand the model sent to the solver after presolve and
> i would like to achieve that in an automated manner.
>
> I am almost happy with "solexpand _scon;" and "option auxfiles cfr;"
> with one notable exception. The expressions are not shown for those
> variables in the .fix file that are not fixed to a constant value. How
> can i "expand" those defined variables in an automated manner?
>
> Any help is appreciated.
>
> Ali
>

[AMPL 2959] Re: Constraint formulation

2009-11-05T15:06:18Z

I agree with Bob that your question could use some refinement, but I
think I know what you're getting at, so I'll take a shot.

I don't think you can do exactly what you want, but you may come
close. Let M be a sufficiently large positive constant and epsilon a
sufficiently small (but strictly positive) constant, and let Y (a
variable) be the output of your first IF. Introduce a new binary
variable Z together with the following constraints:

(1) epsilon*Z <= A <= M*Z
(2) B - M*Z <= Y <= B + M*Z
(3) -M*(1-Z) <= Y <= M*(1-Z).

A <= 0 forces Z = 0 in the left part of (1), which forces Y = B in (2)
and makes (3) vacuous. A >= epsilon forces Z = 1 in the right side of
(1), which forces Y = 0 in (3) and makes (2) vacuous. Besides
introducing a binary variable, there is one catch: 0 < A < epsilon has
been excised from the feasible region. So this won't work if
arbitrarily small but positive values of A are legitimate contenders
for an optimal solution.

Incidentally, I used the same M in every constraint because I'm too
lazy to type subscripts, but you can (and probably should) use a
distinct M (the tightest value you can find that is guaranteed not to
cut off an optimal solution) in each constraint.

/Paul

On Nov 1, 12:09 pm, AlisonT <alisontho...@...> wrote:

> Hi,
>
> Can someone please advise me how I should formulate a constraint that
> represents the following:
>
> IF( "decision variable A" > 0, then 0, else "decision variable B")
>
> I also need to formulate a similar constraint, where "decision
> variable B" is 1 and is therefore a binary decision variable i.e.
> IF( "decision variable C" > 0, then 0, else 1)
>
> Thank you so much.
> Alison

[AMPL 2957] expanding defined variables

2009-11-05T08:53:22Z

I would like to transform my benchmarks into a form that can be used
by other researchers not using AMPL. (In my highly biased opinion they
should be using AMPL :) )

I would like to expand the model sent to the solver after presolve and
i would like to achieve that in an automated manner.

I am almost happy with "solexpand _scon;" and "option auxfiles cfr;"
with one notable exception. The expressions are not shown for those
variables in the .fix file that are not fixed to a constant value. How
can i "expand" those defined variables in an automated manner?

Any help is appreciated.

Ali

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

[AMPL 2958] bug in the Linux student version

2009-11-05T08:52:43Z

I have obtained the following weird output below. The true number of
constraints after presolve is 280. The benchmark giving this result is
here:

http://reliablecomputing.eu/homoazeo40.mod

Output:

AMPL Student Version 20090926 (Linux 2.6.25.20-0.5-pae)

Presolve eliminates 120 constraints and 120 variables.
Substitution eliminates 760 variables.
Adjusted problem:
280 variables, all nonlinear
280 constraints; 1194 nonzeros
240 nonlinear constraints
40 linear constraints
0 objectives.

Sorry, the student edition of AMPL is limited to 300 variables
and 300 constraints and objectives (after presolve) for nonlinear
problems. You have 280 variables, 1040 constraints, and 0 objectives.

Ali

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

RE: [AMPL 2956] Constraint formulation

2009-11-04T16:30:24Z

Alison,

Your "IF ..." looks like an expression that you would like to have in your
constraint, but it is not the whole constraint -- there is no "=" or ">=" or
"<=" in it. You would need to state the whole constraint in order for
someone to say how it could be reformulated for a solver. Also you would
need to specify for every decision variable whether it is integer or
continuous, and what its lower and upper bounds are.

Bob Fourer
4er@...

> -----Original Message-----
> From: AlisonT [mailto:alisonthomas@...]
> Sent: Sunday, November 01, 2009 11:09 AM
> To: AMPL Modeling Language
> Subject: [AMPL 2939] Constraint formulation
>
> Hi,
>
> Can someone please advise me how I should formulate a constraint that
> represents the following:
>
> IF( "decision variable A" > 0, then 0, else "decision variable B")
>
> I also need to formulate a similar constraint, where "decision
> variable B" is 1 and is therefore a binary decision variable i.e.
> IF( "decision variable C" > 0, then 0, else 1)
>
> Thank you so much.
> Alison
>

RE: [AMPL 2955] Re: checking intersection of two sets

2009-11-04T15:32:16Z

Raffaele,

You can write something like

set A;
set B;
check: card(A inter B) = 0;

In fact you can write any statement of the form

check: <expr1> = <expr2>;

where <expr1> and <expr2> are expressions that evaluate to numbers. This
can be further generalized to allow an indexing expression between "check"
and ":" and to use any other numerical comparison operator, such as ">=" or
"<=".

Bob Fourer
4er@...

> -----Original Message-----
> From: ruf@... [mailto:rbo@...]
> Sent: Monday, November 02, 2009 2:42 AM
> To: AMPL Modeling Language
> Subject: [AMPL 2945] Re: checking intersection of two sets
>
> Thank you Paul and Bob for your replies.
>
> In fact, I don't need the intersection to do something, but I must
> ensure that the intersection is empty to keep my model coherent. So, I
> think the "check" command is more useful in this case, since it throws
> an error if the condition is not satisfied.
>
> My problem is that I didn't find any example on the usage of the
> "check" command applied to sets. Is that possible? Or can I throw an
> error using the "if" statement as you proposed?
>
> Thank you
>
> Raffaele
>
> On Nov 2, 1:45 am, "Robert Fourer" <4...@...> wrote:
> > In fact you can write the expression for the intersection of A and B
> > directly inside the cardinality function:
> >
> > if card(A inter B) = 0 then {
> > #...
> > }
> >
> > That way you don't have to define another set just in order to assign it
> the
> > value of A inter B.
> >
> > Bob Fourer
> > 4...@...
> >
> > > -----Original Message-----
> > > From: ampl@... [mailto:ampl@...]
> > > On Behalf Of Paul
> > > Sent: Friday, October 30, 2009 10:21 AM
> > > To: AMPL Modeling Language
> > > Subject: [AMPL 2937] Re: checking intersection of two sets
> >
> > > r...@... wrote:
> > > > Hello,
> >
> > > > how can I check that the intersection of two sets is empty?
> >
> > > > Thank you
> >
> > > set A;
> > > set B;
> > > # ...
> > > set C := A inter B;
> > > if card(C) == 0 then {
> > > #...
> > > }
> >
> >
>
> --
>
> You received this message because you are subscribed to the Google Groups
> "AMPL Modeling Language" group.
> To post to this group, send email to ampl@....
> For more options, visit this group at
> http://groups.google.com/group/ampl?hl=en.
>

[AMPL 2962] Impossible deduced bounds ...

2009-11-04T15:14:32Z

SnsrAssignRMtd.mod (8K) Download Attachment
SnsrAssignRMtd7.dat (3K) Download Attachment

[AMPL 2954] Re: using branch and bound

2009-11-03T05:39:01Z

enhany75 wrote:
> Hi,
>
> Is branch and cut is the default algorithm to solve MIP models?

It is in all the solvers I'm familiar with.

> How to use branch and bound instead of branch and cut to solve MIP
> problems?

The solver should allow you to set some parameters to turn off the
various types of cuts.

/Paul

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

[AMPL 2960] Re: Problem with AMPL 10 together with CPLEX 11

2009-11-03T02:06:57Z

Thank you very much for your quick answers!

Yes, when I remove the line

option solver cplex;

then the problem disappears, however, then AMPL uses CPLEX 10 rather
than CPLEX 11. I will do what Paul suggested and try to compile
cplexampl with CPLEX 11.
Best,
Andy

> On Mon, 02 Nov 2009, Andreas wrote:
> > Hello,
> > I am having the following problem: I am usingAMPL10 together with
> >CPLEX11. When I specify inAMPLto useCPLEXwith
>
> > option solvercplex;
>
> > and then run the
>
> > solve;
>
> > command, I get the following output:
>
> you dont have to use the 'option solvercplex' line. remove this line and your
> problem should get solved.
>
> --
> regards
> Ashutosh Mahajanhttp://coral.ie.lehigh.edu/~asm4

RE: [AMPL 2950] Re: Why is ampl+snopt slow in my program?

2009-11-02T12:08:50Z

Cheng,

The #execute phase is devoted to executing AMPL commands -- this is why the
line for #execute comes before the "solve" statement. Do you have some AMPL
statements that are run to set up the data or to perform some other tasks
before solving? Especially, a lot of looping can be time-consuming. But
possibly these statements won't have to be executed again before later
solves.

At present AMPL is limited by the time it takes to exchange problem and
solution files with the solver. You'll have to experiment to see how many
solves you can get per second. Subsequent solves should be much faster than
the first one, and you might be able to speed things up by incorporating
some of the presolve phase's simplifications directly into your model, and
then turning presolve off ("option presolve 0;").

Bob Fourer
4er@...

> -----Original Message-----
> From: ampl@... [mailto:ampl@...]
> On Behalf Of Cheng [rencheng.daniel@...]
> Sent: Thursday, October 29, 2009 9:25 PM
> To: AMPL Modeling Language
> Subject: [AMPL 2935] Re: Why is ampl+snopt slow in my program?
>
>
> Thank you Robert.
>
> I tested to use ipopt, but it took longer time than using snopt.
>
> Here is the result of using "option times 1" with snopt:
>
> Call SNOPT:
> # incremental total
> #phase seconds memory memory
> #execute 0.483603 99645780 99645784
> ### .\ampl_files\optimization_solve.run:34(1649) solve ...
> #compile 0.0156001 0 99645784
> #genmod 0 0 99645784
> #merge 0 0 99645784
> #collect 0.0468003 0 99645784
> #presolve 0.124801 41984 99687768
> #solveout 0.109201 0 99687768
>
> Presolve eliminates 16623 constraints and 17823 variables.
> Substitution eliminates 83359 variables.
> Adjusted problem:
> 144 variables, all nonlinear
> 0 constraints
> 1 nonlinear objective; 144 nonzeros.
>
> #presolve 0 0 99687768
> #output 0.0156001 0 99687768
> SNOPT 6.2-1: Superbasics_limit=249
> Crash_option=3
> Crash_tolerance=0.1
> #Total 0.795605
> SNOPT 6.2-1: Optimal solution found.
> 0 iterations, objective 0
> #execute 0 1483800 101171568
>
> The first #execute took the most time, what does #execute represent
> here?
>
> Here is the result of using "option gentimes 1" with snopt:
> Call SNOPT:
> ### .\ampl_files\optimization_solve.run:34(1649) solve ...
> ##genmod times:
> ##seq seconds cum. sec. mem. inc. name
> ## 513 0 0 0 phase_time_const
>
> Presolve eliminates 16623 constraints and 17823 variables.
> Substitution eliminates 83359 variables.
> Adjusted problem:
> 144 variables, all nonlinear
> 0 constraints
> 1 nonlinear objective; 144 nonzeros.
>
> SNOPT 6.2-1: Superbasics_limit=249
> Crash_option=3
> Crash_tolerance=0.1
> SNOPT 6.2-1: Optimal solution found.
> 0 iterations, objective 0
>
> I'm doing work related with computer games, I'm wondering if AMPL is
> applicable for highly interactive program(say run 30 times per
> second)?
>
> On Oct 30, 12:49 am, "Robert Fourer" <4...@...> wrote:
> > _ampl_elapsed_time is the amount of time that you actually spent waiting
> for
> > your computer to run AMPL. (It is sometimes called the "wall clock"
> time
> > because it is the time you would measure using a clock unconnected to
> the
> > computer.) _ampl_time is an estimate of the amount of time that the
> > computer's processor actually devoted to running AMPL (and not to the
> other
> > processes that were active at the same time); it is basically the sum of
> > _ampl_system_time and _ampl_user_time, which are two components of the
> > processor time that Unix-based systems (such as Linux) measure
> separately.
> >
> > I had interpreted your original post as implying that the excessive time
> was
> > spent in SNOPT. If it's in AMPL then maybe the presolve phase is taking
> a
> > lot of time to eliminate all those variables and constraints, but you
> can
> > tell for sure where AMPL is spending its time by setting "option times
> 1;"
> > as described in question 3.6 ofwww.ampl.com/FAQ.
> >
> > Bob Fourer
> > 4...@...
> >
> >
> >
> > > -----Original Message-----
> > > From: ampl@... [mailto:ampl@...]
> > > On Behalf Of Cheng [rencheng.dan...@...]
> > > Sent: Wednesday, October 28, 2009 9:10 PM
> > > To: AMPL Modeling Language
> > > Subject: [AMPL 2929] Re: Why is ampl+snopt slow in my program?
> >
> > > Thanks.
> >
> > > I used the ampl command to test time:
> > > _ampl_elapsed_time = 0.887
> > > _ampl_system_time = 0.0780005
> > > _ampl_user_time = 0.780005
> > > _ampl_time = 0.858005
> >
> > > _solve_system_time = 0
> > > _total_solve_system_time = 0
> > > _total_solve_user_time = 0
> >
> > > It seems that the solver didn't take time, but it's ampl which cost
> > > much.
> > > Can anyone explain to me what is _ampl_user_time, what's difference
> > > between _ampl_elapsed_time and _ampl_time ?
> >
> > > Also I will try IPOPT.
> >
> > > Cheng
> >
> > > On Oct 29, 5:40 am, Hans Mittelmann <mittelm...@...> wrote:
> > > > You may be measuring the preprocessing time by AMPL and/or SNOPT. If
> > > > you want to solve problems with 10,000-100,000 variables, SNOPT-6
> may
> > > > be slow and SNOPT-7 may be faster but better may be to use an
> interior
> > > > point solver such as IPOPT.
> >
> > > > On Oct 27, 8:32 pm, Cheng <rencheng.dan...@...> wrote:
> >
> > > > > Hi,
> >
> > > > > I'm using ampl with snopt6.2-1, I find it slow even if I give it
> an
> > > > > perfect initial guess(optimal solution as the initial guess).
> > > > > Just now I tested an easy problem, which has no constraints, and
> all
> > > > > the constraints could be kept as unchanged(all 0) for the optimal
> > > > > solution, but still it took 0.99 seconds.
> > > > > Here is the output:
> >
> > > > > Call snopt:
> > > > > Presolve eliminates 16623 constraints and 17823 variables.
> > > > > Substitution eliminates 83359 variables.
> > > > > Adjusted problem:
> > > > > 144 variables, all nonlinear
> > > > > 0 constraints,
> > > > > 1 nonlinear objective, 144 nonzeros.
> >
> > > > > snopt 6.2-1: superbasics_limit=249
> > > > > crash_option = 3
> > > > > crash_tolerance = 0.1
> > > > > snopt 6.2-1: optimal solution found
> > > > > 0 iterations, objective 0
> >
> > > > > Can anyone tell me why it took almost 1 second to solve? Thanks in
> > > > > advance.
> >
> > > > > Cheng
>
> --~--~---------~--~----~------------~-------~--~----~
> You received this message because you are subscribed to the Google Groups
> "AMPL Modeling Language" group.
> To post to this group, send email to ampl@...
> To unsubscribe from this group, send email to
> ampl+unsubscribe@...
> For more options, visit this group at
> http://groups.google.com/group/ampl?hl=en
> -~----------~----~----~----~------~----~------~--~---

RE: [AMPL 2952] I want to save my AMPL output as a matrix in a file

2009-11-02T12:04:01Z

Thank you Robert. I am working on some trickery right now, when I figure it out I will post it in case someone else ever needs to do that.

The AMPL book you co authored is a great resource.

Thank You,
Matt Kaspari
President
Kaspo Inc
cell: 720.373.3920
fax: 303.255.1740
www.kaspoinc.com

-----Original Message-----
From: "Robert Fourer" <4er@...>
To: <ampl@...>, "'mkaspari'" <mkaspari@...>
Date: Mon, 2 Nov 2009 13:47:59 -0600
Subject: RE: [AMPL 2934] I want to save my AMPL output as a matrix in a file

>
> You can write a script that uses printf statements to create a text
> file in
> just about any format you would like. Matlab has an option for reading
> text
> files into n-dimensional arrays. So what you might need here is a
> Matlab
> guru who knows the format for those text files.
>
> Bob Fourer
> 4er@...
>
>
> > -----Original Message-----
> > From: ampl@... [mailto:ampl@...]
> > On Behalf Of mkaspari [mkaspari@...]
> > Sent: Thursday, October 29, 2009 1:08 PM
> > To: AMPL Modeling Language
> > Subject: [AMPL 2934] I want to save my AMPL output as a matrix in a
> file
> >
> >
> > Hello AMPL gurus. I am having trouble. My AMPL program is outputting
> a
> > 3 dimensional matrix. I want to be able to read the output in matlab.
> > So I am trying to create a file that has the AMPL output as a matrix.
> > I am looking at the printf command. Does anyone know a quick way of
> > doing this? Thank you for your time.
> >
>
>
>

RE: [AMPL 2949] I want to save my AMPL output as a matrix in a file

2009-11-02T11:47:59Z

You can write a script that uses printf statements to create a text file in
just about any format you would like. Matlab has an option for reading text
files into n-dimensional arrays. So what you might need here is a Matlab
guru who knows the format for those text files.

Bob Fourer
4er@...

> -----Original Message-----
> From: ampl@... [mailto:ampl@...]
> On Behalf Of mkaspari [mkaspari@...]
> Sent: Thursday, October 29, 2009 1:08 PM
> To: AMPL Modeling Language
> Subject: [AMPL 2934] I want to save my AMPL output as a matrix in a file
>
>
> Hello AMPL gurus. I am having trouble. My AMPL program is outputting a
> 3 dimensional matrix. I want to be able to read the output in matlab.
> So I am trying to create a file that has the AMPL output as a matrix.
> I am looking at the printf command. Does anyone know a quick way of
> doing this? Thank you for your time.
>

[AMPL 2948] using branch and bound

2009-11-02T09:22:46Z

Hi,

Is branch and cut is the default algorithm to solve MIP models?
How to use branch and bound instead of branch and cut to solve MIP
problems?

thanks

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

[AMPL 2947] Re: Problem with AMPL 10 together with CPLEX 11

2009-11-02T08:25:24Z

On Nov 2, 5:57 am, Andreas <andreas.wiese.ber...@...>
wrote:

> Hello,
> I am having the following problem: I am using AMPL 10 together with
> CPLEX 11. When I specify in AMPL to use CPLEX with
>
> option solver cplex;
>
> and then run the
>
> solve;
>
> command, I get the following output:
>
> ILOG AMPL 10.000, licensed to "tu-berlin".
> AMPL Version 20051214 (Linux 2.6.9-5.ELsmp)
>
> Presolve eliminates 1536 constraints and 720 variables.
> Adjusted problem:
> 11224 variables:
> 4544 binary variables
> 6680 linear variables
> 36014 constraints, all linear; 176104 nonzeros
> 0 objectives.
>
> ILOG CPLEX 11.010, licensed to "tu-berlin", options: e m b q use=10
>
> Welcome to CPLEX Interactive Optimizer 11.0.1
> with Simplex, Mixed Integer & Barrier Optimizers
> Copyright (c) ILOG 1997-2008
> CPLEX is a registered trademark of ILOG
>
> Type 'help' for a list of available commands.
> Type 'help' followed by a command name for more
> information on commands.
>
> CPLEX> can't open /tmp/at6353.sol
>
> this confuses me, since I thought that CPLEX reads the input and then
> creates the output file (with the suffix .sol). If I add a command
> like
>
> write moutput;
>
> I get a similar error message ("CPLEX> can't open output.sol").
>
> Thank you very much in advance!
> Kind regards,
> Andy

To get AMPL to pass files directly to CPLEX, and CPLEX to pass
solutions directly to AMPL, you need to use a driver program that
bridges the two. (Not sure if my terminology is exactly correct
there.) CPLEX/ILOG/IBM/whoever sells an AMPL/CPLEX bundle that does
this, but since you apparently have separate copies of AMPL and CPLEX,
you'll need to compile the bridge program yourself. There's C code
available at http://www.netlib.org/ampl/solvers/cplex/.

/Paul

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

Re: [AMPL 2953] Problem with AMPL 10 together with CPLEX 11

2009-11-02T07:28:40Z

On Mon, 02 Nov 2009, Andreas wrote:

you dont have to use the 'option solver cplex' line. remove this line and your
problem should get solved.

--
regards
Ashutosh Mahajan
http://coral.ie.lehigh.edu/~asm4

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

[AMPL 2951] Re: Problem with AMPL 10 together with CPLEX 11

2009-11-02T07:23:30Z

I think the answer has been given in

http://groups.google.com/group/ampl/browse_thread/thread/46a3bf513b2b4f35/3b43882c2da298a2?lnk=gst&q=solver+cplex#3b43882c2da298a2

the thread can be closed.
Thanks,
Andy

On Nov 2, 11:57 am, Andreas <andreas.wiese.ber...@...>
wrote:

[AMPL 2946] Poisson comand in AMPL

2009-11-02T04:11:47Z

Hello can any body suggest me how to solve the below given problem
model file:

set N; # N denotes the set of nodes
set L within {N,N}; # L denotes the set of links
set T := 1..NP; # T denotes the set all paths
set P{T} within {N,N}; # P[p] denotes the arcs in path p That is,
paths are directed.
set Q{N,N} within T; # Q[a,b] denotes the paths that can be
used
to satisfy demand pair (a,b)

param d{N,N} default 0.0; # d[a,b] denotes the demand between node a
& b
set lamda := 1..36;

var y{T,lamda} binary;
var c{L} >= 0;

minimize WorkingCapacity: sum {(i,j) in L} c[i,j];
subject to WorkingDemand {(i,j) in D}:
sum {b in lamda,p in Q[i,j]} y[p,b] = d[i,j];

data file:
param d :=
2 4 10
3 5 10
1 5 10
;

I try to route the above traffic and optimize, which is a static one.

But I need these demand/traffic modelling as per poissons
distribution
and comes with randomly and optimize the result. I think this may be
done using matlab.

Thanks to R.Fourer for the solution and on which I am trying.

Though one solution to the above problem is in AMPL
param d {i in N,j in N: i < j} := Poisson (2) ;

but I am unable to restrict the total number of demand by the above
method, is it possible to resttict it

Secondly is it possible to generate the above data periodically in
reference to system time in AMPL .

thanks & regards,
bm

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

[AMPL 2944] Problem with AMPL 10 together with CPLEX 11

2009-11-02T02:57:09Z

Hello,
I am having the following problem: I am using AMPL 10 together with
CPLEX 11. When I specify in AMPL to use CPLEX with

option solver cplex;

and then run the

solve;

command, I get the following output:

ILOG AMPL 10.000, licensed to "tu-berlin".
AMPL Version 20051214 (Linux 2.6.9-5.ELsmp)

Presolve eliminates 1536 constraints and 720 variables.
Adjusted problem:
11224 variables:
4544 binary variables
6680 linear variables
36014 constraints, all linear; 176104 nonzeros
0 objectives.

ILOG CPLEX 11.010, licensed to "tu-berlin", options: e m b q use=10

Welcome to CPLEX Interactive Optimizer 11.0.1
with Simplex, Mixed Integer & Barrier Optimizers
Copyright (c) ILOG 1997-2008
CPLEX is a registered trademark of ILOG

Type 'help' for a list of available commands.
Type 'help' followed by a command name for more
information on commands.

CPLEX> can't open /tmp/at6353.sol

this confuses me, since I thought that CPLEX reads the input and then
creates the output file (with the suffix .sol). If I add a command
like

write moutput;

I get a similar error message ("CPLEX> can't open output.sol").

Thank you very much in advance!
Kind regards,
Andy

--

You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to ampl@....
For more options, visit this group at http://groups.google.com/group/ampl?hl=en.

[AMPL 2945] Re: checking intersection of two sets

2009-11-02T00:41:54Z

Thank you Paul and Bob for your replies.

In fact, I don't need the intersection to do something, but I must
ensure that the intersection is empty to keep my model coherent. So, I
think the "check" command is more useful in this case, since it throws
an error if the condition is not satisfied.

My problem is that I didn't find any example on the usage of the
"check" command applied to sets. Is that possible? Or can I throw an
error using the "if" statement as you proposed?

Thank you

Raffaele

On Nov 2, 1:45 am, "Robert Fourer" <4...@...> wrote:

> In fact you can write the expression for the intersection of A and B
> directly inside the cardinality function:
>
> if card(A inter B) = 0 then {
> #...
> }
>
> That way you don't have to define another set just in order to assign it the
> value of A inter B.
>
> Bob Fourer
> 4...@...
>
> > -----Original Message-----
> > From: ampl@... [mailto:ampl@...]
> > On Behalf Of Paul
> > Sent: Friday, October 30, 2009 10:21 AM
> > To: AMPL Modeling Language
> > Subject: [AMPL 2937] Re: checking intersection of two sets
>
> > r...@... wrote:
> > > Hello,
>
> > > how can I check that the intersection of two sets is empty?
>
> > > Thank you
>
> > set A;
> > set B;
> > # ...
> > set C := A inter B;
> > if card(C) == 0 then {
> > #...
> > }
>
>

RE: [AMPL 2941] Re: checking intersection of two sets

2009-11-01T16:45:55Z

In fact you can write the expression for the intersection of A and B
directly inside the cardinality function:

if card(A inter B) = 0 then {
#...
}

That way you don't have to define another set just in order to assign it the
value of A inter B.

Bob Fourer
4er@...

> -----Original Message-----
> From: ampl@... [mailto:ampl@...]
> On Behalf Of Paul
> Sent: Friday, October 30, 2009 10:21 AM
> To: AMPL Modeling Language
> Subject: [AMPL 2937] Re: checking intersection of two sets
>
>
> ruf@... wrote:
> > Hello,
> >
> > how can I check that the intersection of two sets is empty?
> >
> > Thank you
>
> set A;
> set B;
> # ...
> set C := A inter B;
> if card(C) == 0 then {
> #...
> }
>

RE: [AMPL 2940] AMPL/CPLEX from Matlab

2009-11-01T15:22:36Z

There is a random Poisson function built into AMPL -- see Table A-3 on page
459 of the AMPL book (2nd edition).

If for some reason you need to generate the random values of d in MATLAB,
you will need to communicate them to AMPL through a text file. You can use
MATLAB's formatted printing statements to create an AMPL data file (which
can be separate from the file that contains all the other data -- just use
two "data" statements). Or you can write just the values into the file and
use AMPL's "read" command to process them as described on page 163 of the
book.

Bob Fourer
4er@...

> -----Original Message-----
> From: ampl@... [mailto:ampl@...]
> On Behalf Of Mohapatra [bmohapatra9@...]
> Sent: Thursday, October 29, 2009 11:50 AM
> To: AMPL Modeling Language
> Subject: [AMPL 2933] AMPL/CPLEX from Matlab
>
>
> Hello
> Can any body plaease help me how to interface matlab with AMPL/CPLEX.
>
> model file:
>
> set N; # N denotes the set of nodes
> set L within {N,N}; # L denotes the set of links
> set T := 1..NP; # T denotes the set all paths
> set P{T} within {N,N}; # P[p] denotes the arcs in path p That is,
> paths are directed.
> set Q{N,N} within T; # Q[a,b] denotes the paths that can be used
> to satisfy demand pair (a,b)
>
> param d{N,N} default 0.0; # d[a,b] denotes the demand between node a
> & b
> set lamda := 1..36;
>
> var y{T,lamda} binary;
> var c{L} >= 0;
>
> minimize WorkingCapacity: sum {(i,j) in L} c[i,j];
> subject to WorkingDemand {(i,j) in D}:
> sum {b in lamda,p in Q[i,j]} y[p,b] = d[i,j];
>
> data file:
> param d :=
> 2 4 10
> 3 5 10
> 1 5 10
> ;
>
> I try to route the above traffic and optimize, which is a static one.
>
> But I need these demand/traffic modelling as per poissons distribution
> and comes with randomly and optimize the result. I think this may be
> done using matlab.
>
> thanks & regards,
> bm
>
>
> --~--~---------~--~----~------------~-------~--~----~
> You received this message because you are subscribed to the Google Groups
> "AMPL Modeling Language" group.
> To post to this group, send email to ampl@...
> To unsubscribe from this group, send email to
> ampl+unsubscribe@...
> For more options, visit this group at
> http://groups.google.com/group/ampl?hl=en
> -~----------~----~----~----~------~----~------~--~---

[AMPL 2943] bug in the Linux student version

2009-11-01T15:06:48Z