Simplify Matrix decimal to integers

I do not want

881/2158 985/1393 780/1351

881/2158 -985/1393 780/1351

-881/1079 0 780/1351

But I want

1 1 1

1 -1 1

-2 0 1

How do I do this using one single function ?

I get this error if I use simplify

Check for incorrect argument data type or missing argument in call to function 'simplify'.

Answer 1

[Alan Stevens]

One way is

M = [881/2158       985/1393       780/1351  
     881/2158      -985/1393       780/1351  
    -881/1079             0        780/1351]

M = 3×3

0.4082 0.7071 0.5774 0.4082 -0.7071 0.5774 -0.8165 0 0.5774

ip = find(M>=0);
in = find(M<0);
M(ip) = ceil(M(ip));
M(in) = floor(M(in));
disp(M)

     1     1     1
     1    -1     1
    -1     0     1

Don't know where you got the -2 from though!

[Shaunak Bagade]

I don't want to round the numbers ... I need to simplify it like 881/2158 is common factor in column 1 ... 985/1393 in column 2 ... 780/1351 in column 3

Answer 2

[Cris LaPierre]

It's likely going to take two functions: one to convert your symbolic expressions to numeric, and a second to round.

I would probably do something like

new_var = round(double(old_var))

[Shaunak Bagade]

I don't want to round the numbers ... I need to simplify it like 881/2158 is common factor in column 1 ... 985/1393 in column 2 ... 780/1351 in column 3

[Cris LaPierre]

You say you are getting an error message, so you must have code already. Please share what you have tried.

[Cris LaPierre]

Do you mean to do something like this? If so, look into elementwise division.

M = [881/2158       985/1393       780/1351  
     881/2158      -985/1393       780/1351  
    -881/1079             0        780/1351];

B = M./M(1,:)

B = 3×3

1 1 1 1 -1 1 -2 0 1

[Steven Lord]

Or you could normalize by the first element in each column.

M = [881/2158       985/1393       780/1351  
     881/2158      -985/1393       780/1351  
    -881/1079             0        780/1351];

B = normalize(M, 1, 'scale', 'first') % Normalize in dimension 1

B = 3×3

1 1 1 1 -1 1 -2 0 1

[Shaunak Bagade]

Thanks But now we know M but what if we need to normalize it when M is not known