Re: Solver, MILP and Boolean constraints



I join the sample file Massi sent me. Effectively, there are problems.
First, if we use a constraint relative to a range as E2:E15, 
global_range_contained fails because a->v_range.cell.b.sheet is NULL and
is different from a->v_range.cell.a.sheet. IMHO a better test would be:

        if ((a->v_range.cell.a.sheet != a->v_range.cell.b.sheet)
                && (a->v_range.cell.b.sheet != NULL))
                return FALSE;

We have probably the same problem with b, I propose the following:

        if ((a->v_range.cell.a.sheet != b->v_range.cell.a.sheet)
                && (a->v_range.cell.a.sheet != b->v_range.cell.b.sheet)
                && (b->v_range.cell.b.sheet != NULL))
                return FALSE;

Worse, when I replaced the E2:E15 Bool test by E2 Bool, E3 Bool,...,
gnumeric dies silently on attempting to solve the problem. I don't know
why. After the moduifications I made in the global_range_contained code,
I can't anymore reproduce the problem.

Anyway, the solver does not seem able to solve the problem (the message
is: "Unknown error").

Le lundi 18 octobre 2004 à 16:02 +0200, Jean Bréfort a écrit :
It might be a bug. Can you join your file?

Le lundi 18 octobre 2004 à 15:52 +0200, Massi A Milani a écrit :
Thank you very much for your clarification.
I am sorry that I was not too much clear on the issue.

The reason I have sent the email to the list is exactly the following
Cells in my problem (T2:T204) are in the "By changing cells" entry.
Therefore, Gnumeric should not give me that message.

Massi


On Mon, 18 Oct 2004 15:44:02 +0200, Jean Bréfort
<jean brefort normalesup org> wrote:
Le samedi 16 octobre 2004 à 14:16 +0000, Massi a écrit :


Hi guys,

I am running a MILP problem with Gnumeric 1.2.13.
The Solver of Gnumeric is able to solve it in less than 1 second, 5
time faster than Excel. However, introducing a Boolean constraint
(i.e. modifiable cells can only assume a binary value, 0 or 1), I get
the following message

Constraint `T2:T204 Bool ' is for a cell that is not an input cell

Since the Handbook of Gnumeric does not actually specify how to input
a binary constraint, I included it as I usually do in Excel.
I am just wondering whether I made a mistake or Boolean constraints do
not work properly.

I would really appreciated your help.
Massi

PS: I have been pretty happy to notice that Gnumeric does not have the
200-constraint limit as Excel does. Good job, guys!

The cell for which you add a constraint must be an input cell of the
problem. In other words, it must already be listed in the "By Changing
Cells" entry.





_______________________________________________
gnumeric-list mailing list
gnumeric-list gnome org
http://mail.gnome.org/mailman/listinfo/gnumeric-list

Attachment: example.xls
Description: MS-Excel spreadsheet



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]