Hello,
I am using GLPK and i am quite satisfied with it, it is really a gem
of the open source community.
I would like to compile it with icc on Linux i686-smp, 32 bit. I use
it for solving (continuous) LP problems through the API, so i do not
have to compile the routines related to mps file
Hello,
I have a problem with setting the message level:
lpx_set_int_parm(lp, LPX_K_MSGLEV, 1);
works fine if lpx_simplex(lp) is invoked. If i call lpx_interior(lp)
instead, it has no effect on the message level.
Could anyone help me, please?
Many thanks,
Ali
Hello,
I have a problem with setting the message level:
lpx_set_int_parm(lp, LPX_K_MSGLEV, 1);
works fine if lpx_simplex(lp) is invoked. If i call lpx_interior(lp)
instead, it has no effect on the message level.
Could anyone help me, please?
Many thanks,
Ali
The issue is unclear. Counting the number of integer feasible solutions
(as well as generating them) for your problem is a trivial task. Since
the problem has no objective, what does a better solution mean?
I guess he mean more efficient or elegant way of solving the problem.
It would be nice
Hello,
It is part of the GLPK solver.
Just type glpsol --model transportation.mod --output
transportation.sol . I have attached an example file
transportation.mod from the documentation.
Good luck,
Ali
On 5/29/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
I find references to MathProg for
Hello,
At the second call of lpx_simplex your problem is primal infeasible so
phase I of the primal simplex is invoked, that is why you see 0 at the
objective value. At iteration 44, it reaches feasibilty and switches
to phase II.
It seems to me that you would like to re-optimize your problem.
Here is a C implementation of the Hungarian method:
http://www.informatik.uni-freiburg.de/~stachnis/misc.html
Good luck,
Ali
On 6/12/07, Andrew Makhorin [EMAIL PROTECTED] wrote:
\sum_{i=1}^{375} x_i_j = 1, \all j
\sum_{j=1}^{375} x_i_j = 1, \all i
where x_i_j denotes that the combination
Hello,
You can download v4.9 from the GnuWin32 project:
https://sourceforge.net/project/showfiles.php?group_id=23617release_id=189708
Note this is not the latest version.
I think it shouldn't be too difficult to compile the latest source
package on your machine.
Good luck,
Ali
On 6/15/07,
Hello,
Try this:
http://downloads.sourceforge.net/gnuwin32/glpk-4.9.exe?use_mirror=surfnet
Good luck!
On 7/11/07, tobias anglevik [EMAIL PROTECTED] wrote:
Hello,
I´m trying to download GLPK and install it but I don´t know how to do it.
Does anyone know how to explain this for a guy how is a
Dear Colleagues,
I have the following problem. In my computations i have to solve LP
problems repeatedly. The LP problems are not at all difficult, the
difficulty lies in the enormous number of LP problems to be solved.
One thing is common in all LP problems: they have the same form, i.e.
the
If i am not mistaken, the function you are looking for is
glp_set_col_kind(glp_prob* mip, int j, int kind)
where kind is GLP_BV. It is discussed in section Mixed integer
programming routines.
Do you have to use the C API? Using Mathprog could be easier.
Good luck,
Ali
On 9/15/07, Nicholas
Dear Marc,
This exact arithmetic feature in GLPK is very interesting but i am
afraid it is not applicable to relatively large linear programs.
If you need some kind of verification of your result, C Jansson
published a rigorous method and numerical results for real life
problems illustrating its
Dear Enrico,
I could not understand your problem clearly but there was a similar
discussion. I think you can implement what you need via callback
routines. Please see Andrew's responses:
http://lists.gnu.org/archive/html/help-glpk/2007-08/msg8.html
Thank you for your response.
One reason is debugging. I generate the LP problems by linearizing
nonlinear problems. A redundant nonlinear constraint may or may not be
nonredundant in the LP problem after linearization. Identifying
redundant rows / columns could help me debugging the nonlinear
Thank you for your detailed e-mail.
I tried parm.nrs_max = 10; but it did not help.
The objective
may be inaccurate also due to reduced cost tolerance which is 1e-7 by
default. Besides, if the objective coefficients are too small (much
less than 1.0), it would be desired to scale them.
My
, Michael Hennebry
[EMAIL PROTECTED] wrote:
On Mon, 14 Jan 2008, Ali Baharev wrote:
I faced the following problem. I repeatedly call glp_simplex on the
same lp object (only continuous variables) after manipulating the
objective function. The objective function value i get seems to be a
bit
Thanks for the tips.
There was an LP problem where 2.0 = x_j = 3.0 and solving min x_j
resulted 1.5. I have lost this example and i do not remember the
numbers but it was something like that. It would be difficult to
explain the inconsistency with the current one.
It is OK if the result is
Thanks for the tip.
I did try the exact solver but it is very-very slow, as i expected.
My LP problems are generated by successive linearization of a
nonlinear problem, and i need to automate the solution process. So my
problem is not only for this particular LP problem, i need an error
estimate
My LP problems are generated by successive linearization of a
nonlinear problem, and i need to automate the solution process. So my
problem is not only for this particular LP problem, i need an error
estimate on the objective function value for each solved LP problem.
I think that in
After i had spent more than a week with hopeless debugging i found
that the true solution was lost due to a bug in my source code: an
array was treated as the indexing started from one but for that
particular array indexing started from zero. As a result the LP solver
got incorrect input data, and
Hi,
Why don't you call glp_simplex with zero (or constant) objective?
Good luck!
Ali
On Feb 13, 2008 12:50 AM, Dragos Ilie [EMAIL PROTECTED] wrote:
Hi!
I am interested in using glpk to find a (initial) feasible solution to a
problem with linear constraints. This should be equivalent the
Maybe i miss something import but a intersects b if and only if
the intersection of internal areas is not empty. One has to solve the
following LP problem (feasibility test):
min z=0
s.t.
x = 1
x = 3
y = 2
y = 3
x = 2
x = 4
y = 1
y = 4
If the question would be that how to compute the convex hull
Make sure that the objective is not set (i.e. zero), then call
glp_simplex. The procedure stops where the Phase 1 stops.
Best,
Ali
___
Help-glpk mailing list
Help-glpk@gnu.org
http://lists.gnu.org/mailman/listinfo/help-glpk
Probably this thread helps:
http://lists.gnu.org/archive/html/help-glpk/2006-12/msg00018.html
Best,
Ali
___
Help-glpk mailing list
Help-glpk@gnu.org
http://lists.gnu.org/mailman/listinfo/help-glpk
Dear Paul,
The number of integer variables is just one thing. I think i have a
MIP problem with less than 20 integer variables and glpsol would not
be able to solve it.
You should first improve and / or simplify your model rather than
change the code of the GLPK.
Or try a commercial solver such
Hello,
I am afraid setting the objective to zero is probably not the best
solution for Chris. He would like to have 'the current best
(feasible) solution' however setting the objective to zero yields the
first integer feasible which may not be a good solution.
Probably setting a gap limit would
Dear List Members,
We often see posts from confused users who cannot understand why GLPK
behaves weirdly for some LP instances. The suspected reason is usually
excessive round-off error. I also had a long discussion on this topic
in January:
Hello,
In the Reference Manual at 1.3 Brief Example you can find an example C
program with explanation.
Good luck,
Ali
On Fri, Aug 8, 2008 at 6:54 PM, KAKAR, RATIKA [EMAIL PROTECTED] wrote:
Hi,
Could someone please send me a simple program for MIP problem, in Java
or C? I need help with
I am convinced that interval arithmetic provide the tools to overcome
this and similar other problems, see for example:
http://www.ti3.tu-harburg.de/~keil/#software
http://www.ti3.tu-harburg.de/cgi-bin/cjbibsearch/publications/ti3.html?author=jansson
(all papers containing the word rigorous in
Dear Vijay Patil,
You are right. I should have used term repeating decimal instead of
irrational. Should have looked up before using it.
I would rather say not machine / internally representable.
For example 0.1 is neither irrational nor repeating decimal, and still
cannot be represented
I would like to note that 0.1 is a rational number and therefore its
decimal representation is periodic: 0.1...000... = 0.0...999...,
i.e. it is a repeating decimal.
You are right.
I misunderstood the meaning of repeating decimal, my fault.
I just wanted to say that interval
Could you please e-mail the source code too?
It is hard to figure out without that.
Ali
On Tue, Aug 19, 2008 at 7:41 PM, Jose Monreal [EMAIL PROTECTED] wrote:
Good afternoon everyone,
I have a problem with glpk, after I create a problem with lpx_create_prob in
a c function, it changes the
Please provide more information on how the glpsol is started, by a shell script?
If you use the C API to call glp_simplex, please send us the source file.
It seems that the scheduler or maybe memory is causing the program to be
killed.
Please send us the source file in which you suspect the
And you have the PDF with the new API? does it exist?
There is no PDF version. However in every distribution contains the
documentaion also, if i am not mistaken PS version is available. With
the free Ghostscript:
http://pages.cs.wisc.edu/~ghost/
you should be able to read that. If not, i will
I was always curious why there are so many problems with 64 bit
applications. If someone could e-mail me an explanation (preferably
written for dummies) i would appreciate.
In the wikipedia i read this:
SSE2 instructions on x86 and x64 CPUs do require the data to be
128-bit (16-byte) aligned
How can I add a column like glp_set_col_ but I do not want that variable
to apper at the Objectve Function.
If you do not set the coefficient of a variable then it will not
appear in the objective. In other words it will appear with zero
coefficient by default.
Best,
Ali
Thanks! :)
___
Help-glpk mailing list
Help-glpk@gnu.org
http://lists.gnu.org/mailman/listinfo/help-glpk
lpx_load_matrix(lp,1,ia,ja,ar);
You did not load all the matrix, only one non-zero element but you
have 4 of them. Try:
lpx_load_matrix(lp, 4,ia,ja,ar);
It seems to me you are using an old version of GLPK, try to update to
a more recent one.
Good luck,
Ali
First of all: it is really good to see GLPK becoming better and better.
There are a number of primal heuristics which would be nice to have in GLPK.
Would it be difficult to implement the following proposals, regarding
how often the problem of big M formulation comes up here in the
mailing list?
You are completely correct. Though, I'm finding the relaxed solution right
now via a decomposition method (which is why I have no basis at the end) and
I'm finding it up to 100X faster than glp_simplex would otherwise. Since
this is the case, I'd love to use that as a starting point for
If finding the minimal set of conflicting constraints is a well known
problem, are there any solutions provided in GLPK for it?
Not that i know of.
It is outside of the scope of GLPK.
And if it is, what other tools could be used to find
out the minimal set?
I am sorry i cannot be more
Hi,
Try this:
int iter_count = lpx_get_int_parm(lp, LPX_K_ITCNT);
May or may not work with you distribution, i used it a long while ago...
Good luck,
Ali
___
Help-glpk mailing list
Help-glpk@gnu.org
http://lists.gnu.org/mailman/listinfo/help-glpk
The current version is glpk-4.36.tar.gz
___
Help-glpk mailing list
Help-glpk@gnu.org
http://lists.gnu.org/mailman/listinfo/help-glpk
Hi,
Once upon a time, I had a discussion with one of the GPULib developers
and he told me GPUs were not good for the simplex method, because the
gain is not so significant for sparse matrix-vector operations. (Or at
least it was what i understood.)
Ali
Hi,
Your question is a little off-topic.
Ipopt is very robust / stable but slow. The other open-source
nonlinear solvers i tried so far failed to solve my problems.
https://projects.coin-or.org/Ipopt
The commercial solver CONOPT is said to be both fast and robust / stable.
Hi,
Download the GLPK source distribution. It includes the manual in PDF format.
http://ftp.gnu.org/gnu/glpk/
Ali
___
Help-glpk mailing list
Help-glpk@gnu.org
http://lists.gnu.org/mailman/listinfo/help-glpk
In the documentation, find:
4 Advanced API Routines
4.3 Simplex tableau routines
4.3.1 glp_eval_tab_row—compute row of the tableau . . . . . 119
4.3.2 glp_eval_tab_col—compute column of the tableau . . . 120
Good luck,
Ali
___
Help-glpk mailing list
The time complexity of the Hungarian method is O(n^3). I was unable to
determine how this relates to the complexity of the network simplex
method.
Ali
___
Help-glpk mailing list
Help-glpk@gnu.org
http://lists.gnu.org/mailman/listinfo/help-glpk
Hello,
There is nothing wrong.
It seems to me you have chosen the bounds to be the minimum and
maximum value for a 32-bit signed integer.
But your bounds are converted to double precision variables, and GLPK
checks bounds accordingly, the relative error is less than 1.0e-9
which is OK.
You can
Another interesting thing is that, using EXACTLY the same parameters, the
compilation of glpsol 4.9 I downloaded from http://winglpk.sourceforge.net/
is unable to get the gap below 4.6% (the optimum is 0%), whereas it seems to
solve fairly easily in the GUSEK 0.2 Windows IDE. This is exactly
Hi Yaron,
The cplex lp file format is smarter than the MPS that is why i suggested that.
However you should not get incorrect results with the MPS... :(
It may be some MPS format problems (SCIP expects or handles
differently than GLPK) or some numerical problems you avoided by using
the smarter
Dear Andrew,
Thank you for your response. I decided to move our conversation
http://lists.gnu.org/archive/html/bug-glpk/2009-10/msg00018.html
here from the bug-glpk because we are not dealing with a bug anymore.
I just added a limit on the number of restarts in the simplex
algorithm. In this
This thread is getting more and more exciting :)
The Zimpl User Guide gives examples of reformulations that made the
solution process much faster:
Section 5 - Modeling examples
http://zimpl.zib.de/download/zimpl.pdf
It appeared earlier on this mailing list before:
AIMMS Modeling Guide -
Hello,
You can tell GLPK to use either primal or dual, see the documentation
(a pdf) included in the distribution:
glp simplex—solve LP problem with the primal or dual simplex method
and also:
Control parameters
GLP_PRIMAL—use two-phase primal simplex;
GLP_DUAL —use two-phase dual simplex;
Hi,
However, when I did the LPR, x1 and x2 can become 0.5. Though it still
satisfies the constraint x1+x2=1, but that is NOT what I want.
Please re-read Andrew's e-mail:
LP relaxation is just an LP problem, where all variables are allowed
to take any *continuous* values, if only they are
No, it is not.
Could you please send us your code so we have the chance to help?
Ali
On Fri, Jun 11, 2010 at 10:41 AM, Fabio pruef...@me.com wrote:
Is it normal that solving the problem again with fixed integer variables at
the optimal values takes nearly as long as solving the original
Dear Andrew,
In my application I would like to get the computational efforts
through the C API. In the previous releases I could use
lpx_get_int_parm with LPX_K_ITCNT but it does not seem to work
anymore. Please provide API functions to get the computational efforts
such as iteration count.
If I
for an
official solution.
Ali
On Mon, Nov 29, 2010 at 2:01 AM, Ali Baharev ali.baha...@gmail.com wrote:
Dear Andrew,
In my application I would like to get the computational efforts
through the C API. In the previous releases I could use
lpx_get_int_parm with LPX_K_ITCNT but it does not seem to work
Dear Andrew,
Thank you for the detailed explanation.
I decided not to include routines like lpx_get_parm/lpx_set_parm in the
new api, because most of control parameters in the old api are solver
properties rather than model properties. Probably such statistics could
be reported through the
GLP_MSG_ERR is a value of a control parameter passed to glp_simplex, and
glp_scale_prob knows nothing about it. I think it is normal for
processing routines to display some brief information about what is
going on. In any case you can suppress/resume the terminal output with
glp_term_out.
I
60 matches
Mail list logo
