Attempting to take the sum of an index in my loop

Hello everyone,

In my script, I am calculating RMSDs of a variety of things, and creating a 155x2 matrix.

predictions=load('Sparta_Predictions2.txt');
experimental=load('ExperimentalValue_Unk2.txt');
x=predictions(:,1);
error=predictions(:,2);
y=experimental(:,1);
z = zeros(1,6);

sizevaly = length(y)/6;
sizevalx=length(x(:,1))/6;

b = zeros(sizevaly,sizevalx);
d=(1:sizevalx);
% e=zeros(sizevaly,6);
e=zeros(1,6);
for n=1:sizevalx
    for j=1:sizevaly   
        for i=1:6
            xindex = i+(6*(n-1));
            yindex=i+(6*(j-1));
            z(i)=((x(xindex)-y(yindex)))^2;
            e(i)=(z(i)/(error(xindex)^2)); 
            if e(i)>1000
                e(i)=0;
            end
            
            b(j,n)=sqrt((1/5)*sum(e,2)); 
        end     
    end
    
   
   A=b';
    
end

I also have a diagonal vector, that is filtering my 155x2 matrix by setting the values to 0 and 1 according to a certain threshold.

[m,n] = size(A);
pattern = false(m,n);
dvec = 1 : m+1 : 1 + (n-1)*(m+1);
for ni = 1 : m-n+1
    pattern(dvec) = all(A(dvec)<2);
    dvec = dvec + 1;
end

What I also wanted to do, in addition to the above filtering, was take the sum of that diagonal vector. To this end I have tried to write a simple loop similar to the above:

a=zeros(m)
for nt=1:m-n+1
    a(nt)=sum(A(dvec))
    dvec=dvec+1;
end

In short, I have made a vector of zeros the same size as my data. Then I want to take the sum of the daigonal vector and store it in my vector a.

The issue is I get this error:

Index exceeds the number of array elements (310).

Error in For_Sparta (line 45)
    a(nt)=sum(A(dvec))

From what I've looked up, it appears to mean that I'm not actually storing the sum values from each iteration. However from my understanding, all you have to do is index your loop values and that should store them for every iteration (which is what I have attempted to do with a(nt)). So a(1) is the sum of my daigonal vector in my data A, and then dvec is added by one, and then a(2) is the sum of my diagonal vector+1.

I.E. In other words:

If we say I have a matrix

G =

     1     1
     1     2
     1     3

Then I want to make a vector with the sum of the diagonal in this. So say vector a would be

     a =

     3
     4

In this example.

[darova]

where is data

[Sam Mahdi]

those data files are relatively large (over 1000 values), so I thought including them would just make things more difficult. They are also modified and vary quite a bit, so including them would be pointless. This script is set up to theoretically work with multiple variations in the data set. The only thing that is really "set" is the size of m. M will always be 155; however n could change depending on the input. Due to all this, I haven't included the data.

Also, the question is really just asking how to store values in an index for the loop I've written. So the data is somewhat irrelevant (it could be anything). It could be that my loop is not actually doing what I want it to do either. So just looking for feedback on my script and error.

[darova]

I can't run your script without data.

My computer says:

Error using load
Unable to read file 'Sparta_Predictions2.txt': no such file
or directory. 

[Sam Mahdi]

Here I just made up some random numbers in the proper format for the scirpt. The values are shit, but the script will run fine. Use test1 in replace of Sparta_Predictions

[darova]

Is this correct diagonal?

[Sam Mahdi]

Yes. On a numberical level here is what it should look like:

>> A

A =

    0.5621    1.7063
    1.4955    2.3217
    1.3758    3.2607

>> sum(A(dvec))

ans =

    2.8839

So if you run the sample set I provided, you will get those values of A. If you just use dvec at its initial value of 1 and 5, that will give you the above answer. The thing is, I want to take this basic concept and make it into a loop for larger data sets. So the loop I have now should do dvec+1, so that'll give 2 and 6.

sum(A(dvec+1))

ans =

    4.7562

And so on and so forth. I see that one of my issues is the final iteration where dvec is 3 and 7 will cause an error, since there is no 7th value. However from my understanding, setting my iterations to 1:m-n+1 should limit the number of diagonal vectors (so in this case it would only be 1:2 since you can only get 2 diagonals since there is no 7th value.

Answer 1

[Sam Mahdi]

I guess I'll post this as an answer. The issue appears to be my diagonal vector had to be redfined. It was using the diagonal vector from my previous loop it seems, and that would cause an error since the last iteration of that loop will actually have values missing in the diagonals, hence causing the error. All I had to do was redfine my diagonal vector.

a=zeros(m);
dvec = 1 : m+1 : 1 + (n-1)*(m+1);
for nt=1:m-n+1
    a(nt)=sum(A(dvec));
    dvec=dvec+1;
end