Help understanding matlab code ?

I am completely new to Matlab and I also don't have Matlab - but I have been asked to convert some Matlab code to another language, which is proving a little bit tricky for me. At the moment I am stuck with the following lines which are within a loop:

    d=s2(1)*ones(dim(1),1);
    for i=2:R,
        d=[d;s2(i)*ones(dim(i),1)];

s2 is a 1x2 matrix [0.1 1], dim = [3], and R is equal to 1 (the first time this code is evaluated)

My understanding is that the first line results in the 3x1 matrix

    0.1
    0.1
    0.1

But I am confused by what happens in the for loop a) since R is 1, is this a decrementing loop (evaluated with i=2 then i=1 ?) b) I really don't get the line d=[d;s2(i)*ones(dim(i),1)] at all

Could anyone help explain this to me ?

Answer 1

[Image Analyst]

No. First of all dim(1) doesn't work - throws error. But I'll assume you copied it incorrectly and that it is a 3 by 1 column vector. What the code does is to tack on additional columns with different values that depend on s2's value. For example if s2(2) = 12.34 then the second column of d will be [12.34;12.34;12.34]. Then the third column of d will have elements with the value of s2(3), and so on. If R equals 1, then it doesn't even enter the loop at all so d stays the same as before the loop.

The loop does not decrement from 2 to R - it increments. To decrement, there would have to be a step of -1, like

for i = R : -1 : 2

[Modus Operandi]

Thank you. In the code I'm looking at, there is

    dim=[3];

immediately before this code block. Doesn't dim(1) just result in 3 ? As for the loop, since R = 1, I don't get what this

    for i=2:R,

does ?

[Image Analyst]

If you defined dim that way, then yes, dim(1) will be 3. Also, since R=1 the loop never gets entered, as I said earlier, and so basically the loop does nothing and can be deleted, unless you might ever have situations where R does not equal anything less than 2.

[Modus Operandi]

Ahh - OK - that makes sense - thank you !

Answer 2

[dpb]

for i=2:R

for R=1 is an empty loop

Matlab for loop control is

for start:[increment:]end

with the default for increment when not given of +1. If initial end < start, the loop body is never entered. The code as written is the same as

for i=2:1:R

and a decrementing loop would require

for i=2:-1:R

the step variable cannot be skipped in that case. It's same as Fortran DO with the optional increment excepting Fortran writes the increment last instead of in the middle. The difference is that Matlab syntax follows that for vectors with the colon operator for consistency.

I really don't get the line d=[d;s2(i)*ones(dim(i),1)]

If R>=2, then it concatenates the results to the initial d to build a longer column vector of the elements of s repeated the number of times in the dim array.

Answer 3

[Harmon Harmon]

Hi, I'm trying my best to understand this code and how everything works together. I know that there are some comments, which are in green and I understand that.

[Image Analyst]

How is this an answer of Modus's original question?

Do you have a question? Or is this just an announcement?