Thread Subject: Randomization_same number

Hi all,

I am trying to present 4 different numbers 120 times in a vector, in a randomized order. But the thing is that I try to avoid repeating the same number twice in a row, and in the same time to keep the repetition frequency of each number the same (= 30).

Do you have any ideas?

Thanks

"Kostas " <eclipse_dsotm@hotmail.com> wrote in message <hd249a$im2$1@fred.mathworks.com>...
> Hi all,
>
> I am trying to present 4 different numbers 120 times in a vector, in a randomized order. But the thing is that I try to avoid repeating the same number twice in a row, and in the same time to keep the repetition frequency of each number the same (= 30).
>
> Do you have any ideas?
>
> Thanks

BUild the 120 element vector in a sorted order than use "randperm"

The "randperm" will help you, but you need to postprocess the data to change duplicate values on a single row. A swap should be good.

http://www.advancedmcode.org/

>BUild the 120 element vector in a sorted order than use "randperm"
>
>The "randperm" will help you, but you need to postprocess the data to change duplicate >values on a single row. A swap should be good.
>
>http://www.advancedmcode.org/


Thanks for your reply. I think I should have mentioned that I must generate this vector a lot of times in the future, so it must me always different. :)

Thanks again

"Kostas " <eclipse_dsotm@hotmail.com> wrote in message <hd26rr$q78$1@fred.mathworks.com>...
> >BUild the 120 element vector in a sorted order than use "randperm"
> >
> >The "randperm" will help you, but you need to postprocess the data to change duplicate >values on a single row. A swap should be good.
> >
> >http://www.advancedmcode.org/
>
>
> Thanks for your reply. I think I should have mentioned that I must generate this vector a lot of times in the future, so it must me always different. :)
>
> Thanks again

So? Then be efficient in your code. Randperm is
fast. It takes little effort to test to see if elements
are consecutive. Use diff to do the test.

John

> So? Then be efficient in your code. Randperm is
> fast. It takes little effort to test to see if elements
> are consecutive. Use diff to do the test.
>
> John

Thanks John. I do use the randperm function, but of course i have same trials in a row. I can find where (using diff as well), but still cannot solve the problem.

I created a mask to get rid of all the numbers i've used, but i am not sure if it is efficient and worth trying -i am new in matlab. And besides, it doesn't work properly.

Thanks

"D " <eclipse_dsotm@hotmail.com> wrote in message <hd2cjh$evv$1@fred.mathworks.com>...
> > So? Then be efficient in your code. Randperm is
> > fast. It takes little effort to test to see if elements
> > are consecutive. Use diff to do the test.
> >
> > John
>
> Thanks John. I do use the randperm function, but of course i have same trials in a row. I can find where (using diff as well), but still cannot solve the problem.
>
> I created a mask to get rid of all the numbers i've used, but i am not sure if it is efficient and worth trying -i am new in matlab. And besides, it doesn't work properly.
>
> Thanks

So use randperm. Then set up a loop. Scan through
the sequence. If the ith element is equal to the previous
element, then swap it with the next element in the
sequence which is not equal to it. Continue this
operation until done. If you get to the end without
success, then wrap around to the beginning.

This will be fast.

John