Global optimization toolbox - GA

When I run the GA solver, I got the following warning: What does it mean? What action should I take or I just ignore it?

Warning: Length of upper bounds is > length(x); ignoring extra bounds.

Answer 1

[Matt J]

You've specified more upper bounds than there are variables.

[Sata]

Hi Matt J,

The lb and ub for all my variables are within 0 and 5 respectively. I am not very clear about your answer, can you please help to explain a bit more?

[Walter Roberson]

How many upper bounds did you specify? Not what values they are, but numel() of the array of upper bounds you passed in? Then, what is numel() of your x0 vector ? If the number of upper bounds is more than the number of elements in x0 then you get the warning.

[Sata]

TQ Walter Roberson for your comments.

If we don't get rid of the warning messages, will it affect the results?

In the constrain option, if I don’t have any constraints, do I just put [] for both linear equalities and linear inequalities?

[Walter Roberson]

There will be no effect if you ignore the warning. However, warnings often indicate that you might not be passing in the values that you think you are passing in.

Yes, for ga() you can use [] if the constraint does not exist.

[Matt J]

If we don't get rid of the warning messages, will it affect the results?

It sounds like it could affect the results and it doesn't make sense to accept the warnings whose apparent cause is so simple to solve.

If the unknown vector x is of length N, then ub and lb should be length N vectors as well. The warning is telling you that this is not the case. This could very well mean, as Walter intimated, that you are not applying bounds where you think you are.