code gives me a false values?

Hi to all, i have a table which contain 3 columns. my goal is to divise the values of column1 by 64 but i find a value which > 128, i should substract it from 256 . after i should do the same think for the others columns. The problem here when i execute my program until matlab, it gives me a right values for column 2 and colum 3, but for column 1 it gives me a false values.

for x = finale(:,1)
if x >  128
x = x - 256
end
e1 = x / 64
end

for x = finale(:,2)
if x >  128
x = x - 256
end
e1 = x / 64
end

for x = finale(:,3)
if x >  128
x = x - 256
end
e1 = x / 64
end

spreadsheet of finale

255  199  23
7  199  23
7  199  23
7  199  23
9  199  23
7  199  23
7  199  23
7  199  23
7  199  23
7  199  23
7  199  23
7  199  23
7  199  23
3  201  25
13  197  23
13  197  23
13  199  19
255  197  23
11  199  23

This is the problem in colum 1 , it gives me these values :

255
7
7
7
9
7
7
7
7
7
7
7
7
3
13
13
13
255
11
1

Thanks in advance

Answer 1

[Geoff Hayes]

There is no need for a for loop (and I don't think that it is working as you expect anyway). Step through the code to see why it is unnecessary.

Instead, do the following

 % initialize x to the first column
 x    = finale(:,1);

 % determine all indices of elements that are greater than 128
 idcs = x>128;

 % do the subtraction
 x(idcs) = x(idcs) - 256;

 % do the division
 e1 = x/64;

Try the above and see what happens, repeating for each of the three columns.

Though from your question you state that find a value which > 128, i should subtract it from 256. The code finds the number and subtracts 256 from it. Which should it be?

Answer 2

[Matz Johansson Bergström]

You want to store the result in the correct positions in the resulting vector. You keep placing the value in e1. It is better to use indexing when you loop with "for", instead of passing the vector finale(:, 1). For instance

finale2 = []; %we don't overwrite finale, just create a new matrix

for i=1:size(finale,1)
   if finale(i,1)>126
      finale2(i,1)=256-finale(i,1);
   end
end

Please note: we have to do this for all three columns in finale. In this way, we step through finale and use "i" to store the element in finale2. Now finale2 contains the correct elements.

Another way to do it (which is more elegant) is to use vectorization.

finale2 = finale;
inds = find(finale2 > 256); %store indices of the elements that are larger than 256
finale2(inds) = 256-finale2(inds);
finale2 = finale2./64; %divide all elements by 64

This code is much more compact and neater and we don't need to check for each column, which is nice.

[inti]

Thanks for your reply.

 i make a mistake by writing e1 three times. i'm a searching for e,e1,e2 values 

if 
for x = finale(:,1)
if x >  128
x = x - 256
end
e = x / 64
end
for x = finale(:,2)
if x >  128
x = x - 256
end
e1 = x / 64
end
for x = finale(:,3)
if x >  128
x = x - 256
end
e2 = x / 64
end

end

Answer 3

[Jos]

you can do it for all columns in one go (e1 will be a 3 column matrix)

e1=finale;

e1(e1>128)=e1(e1>128)-256; (or e1(e1>128)=256-e1(e1>128); your description and code are different as Geoff mentioned)

e1=e1./64;

[inti]

Thanks for your reply.

 i make a mistake by writing e1 three times. i'm a searching for e,e1,e2 values 

if 
for x = finale(:,1)
if x >  128
x = x - 256
end
e = x / 64
end
for x = finale(:,2)
if x >  128
x = x - 256
end
e1 = x / 64
end
for x = finale(:,3)
if x >  128
x = x - 256
end
e2 = x / 64
end

end

Answer 4

[inti]

Thank you veryyyyyy much.

Now , it works well