Frequency of data occurrence using histc

I have a matrix called data which is 40*40*50 and I want to find all the distinct numbers in the matrix and count their occurrences. This code below seems to work when data is 40*40*39 or smaller, but as soon as I try to run the code on the entire data matrix it starts giving 0.001 as the frequency for all of the unique occurrences

     % code
out = [unique(data),histc(data(:),a)]
plot(a, histc(data(:),a));

I know that using histc is not recommended, but it was the easiest way I could find to do this.

Thanks for any and all help.

[Megan]

what is data?

Answer 1

[Guillaume]

I presume that your numbers are integer (otherwise, you're playing dangerous games using unique with floating point values), in which case:

out = histcounts(data, 'BinMethod', 'integers');

[Edward Jahoda]

Thanks for the response. I am using floating point values (with 6 values after the decimal point), and consequently wondering if you have some suggestion as to how to deal with the floating points or some information on why its dangerous.

[Guillaume]

"some information on why its dangerous"

See the output of

unique([0.3, 0.1+0.1+0.1])

If you want 6 digits of precision, I would do this:

out = histcounts(floor(data * 1e6), 'BinMethod', 'integers');

to move back to integer bins.

Answer 2

[Bruno Luong]

You might take a look at uniquetol()