convert a numbere from integer to double

Hi,

does anyone know how can I covert 1 in 1.0000, please?

Thank you!

[Stephen23]

What is the difference? In MATLAB a number of class double can also display in the command window without trailing digits, even though it is of class double:

>> 1
ans =
     1
>> class(ans)
ans =
double

Is your number really of an integer class, or is it simply a whole (round) number of class double? Are you only interested in the displaying of this value, or are you trying to save it somehow? What problem are you facing that requires these digits?

[Natalia Molinero Mingorance]

Those numbers are double, and they are the BPSK values of a vector (-1, 1, 1, -1...) that I need to save into a .txt file as (-1.0000, 1.0000, 1.0000) in order to upload this file into a software to create a waveform. Any clue, please?

[James Tursa]

So, the real problem isn't conversion ... the real problem is that you need to write a text file in a specific format? That was not mentioned at all in your original post. Please tell us the size and type of the variable(s) involved, and the exact text format you want in the file.

[Andrew Gerlach]

Most of the time Matlab handles this fine, but there are cases when you need to explicitly convert a number type. For example, I just loaded imaging data from a .nii file. When I tried to get the standard deviation from a time series, I got the following error:

>> std(ts)
Error using var (line 74)
Invalid data type. First input argument must be single or double.

Error in std (line 59)
y = sqrt(var(varargin{:}));

The fix was to convert to double manually:

>> std(double(ts))

ans =

  128.1063

Answer 1

[James Tursa]

Your question is not clear. If the number is a scalar double and exactly 1, it will display as 1 without the trailing 0's. If it is very close to 1 but not exactly 1, it will display with 0's. E.g.,

>> 1
ans =
     1
>> 1+eps(1)
ans =
    1.0000

To get more digits to display you can use format. E.g.,

>> format long
>> 1
ans =
     1
>> 1+eps(1)
ans =
   1.000000000000000

If you really have an integer type and need to convert it to double type, use the double function. E.g.,

>> k = int32(1)
k =
           1
>> class(k)
ans =
int32
>> d = double(k)
d =
     1
>> class(d)
ans =
double

[Natalia Molinero Mingorance]

I don't really understand how to use those function, I mean, the syntax: if if got a vector x=[1,1,1], how can I convert it into x=[1.00000, 1.0000, 1.0000]? Thank you!!

Answer 2

[Stephen23]

Here is one easy way that you can save a vector of values with your chosen precision: with dlmwrite:

>> A = [-1, 1, 1, -1];
>> dlmwrite('temp.txt',A, 'precision','%.4f');

produces me a file which includes this text:

-1.0000,1.0000,1.0000,-1.0000

See the fprintf documentation for a list of the precision formatting options.

[Natalia Molinero Mingorance]

that's great! is it possible to add a newline after a pair of values and a space before the -1 too? thank you very much!

[Stephen23]

dlmwrite and the similar csvwrite simply write the matrix to a file with exactly the same shape as it has in the supplied variable: so if you give it a 2x3 matrix, it will write a file with two rows and three columns. To get the desired output you just need to rearrange the input matrix to have the desired shape. You can rearrange the shape of an array using reshape, but note that this keeps the columns together:

B = reshape(A,[],2)

If you want to keep the rows together you can transpose before and after the reshape command:

B = reshape(A.',2,[]).';

Note that dlmwrite seems to use unix newlines by default, which means viewing this file using Windows NotePad it will look like there are no newlines: the problem is Notepad, not the file. You can select the 'newline' option 'pc' to change the newline character used.

You can insert spaces by adding a space character into the format specifier string, like this:

dlmwrite('temp.txt',A, 'precision',' %.4f');

which will place a space in front of each number in the file.

[Natalia Molinero Mingorance]

that's fine, but I just want the space before the -1 and that will insert the space before the 1s too, right? :( thank you!

[Stephen23]

If you want to control the spaces individually then you will need to use a lower-level function, such as fprintf:

>> A = [-1, 1, 1, -1];
>> fid = fopen('temp.txt','wt');
>> fprintf(fid,' %.4f,%.4f\n',A);
>> fclose(fid);

gives me a file with this inside (there is a space in front of the first column):

   -1.0000,1.0000
   1.0000,-1.0000

Note that fprintf will simply keep applying that formatting to the input variables until it runs out of variable... and this is columnwise! You might need to transpose the data before inputting it into fprintf.

If this does not resolve your formatting issues, please upload an example file with the exact formatting that you need, and I will show you how to generate it using dlmwrite or fprintf.

Answer 3

[Shantanu Jana]

you can do this process and latter use 'ans' as variable that hold your answer

>> a=1

a =

     1
>>a = double(a)

>> sprintf('%0.4f', a)

ans =

1.0000

>>

[Guillaume]

 >>a = 1

creates a of type double. Therefore:

>>a = double(a)

does nothing whatsoever.

[Stephen23]

Lets try the suggested code:

>> a = 1;
>> class(a)
ans = double

Lo! we already have a double! Which means that the code suggested by Shantanu Jana:

>> a = double(a);

does nothing useful at all, as a is already of class double.

MATLAB's documentation about floating point numbers states that "...The default is double precision...". My first comment to the original question also clarifies this.