bitcount

Count the number of set bits in each element. Time and memory efficient

bitcount Counts the number of set bits in each element

Usage:

   B = bitcount(A)

Counts the number '1' bits in each element

 B = bitcount(A, bitValue)

"bitValue" = 1 = default = counts the occurance of '1'

if bitValue = 0; counts the number '0'

The total bits to verify is [8,16,32,or 64] based on the maximal value of A

B = bitcount(A, bitValue, maxBits)

the total # of bits to examine

Ref: https://www.mathworks.com/matlabcentral/fileexchange/44371-bitcount-bitwise-hamming-distance

The "bitcount" I have implemented here - is much more time/memory efficient vs the Ref above

Efficiency evaluation test: x 10:

 maxData = 1e4;
    data = uint32(randi([1,intmax('uint32')],1,maxData));
    tic;cell2mat(arrayfun(@(x) sum(bitget(x,1:32)),data,'UniformOutput',false)');t_elapsed1 = toc;
    tic;bitcount(data);t_elapsed2 = toc;
    eff = t_elapsed1 / t_elapsed2;
 
     maxData = 1e4: eff =  10
     maxData = 1e5: eff =  20
     maxData = 5e5: eff =  40
     maxData = 1e6: eff =  56
     maxData = 1e7: eff = 128

Examples:

data = 0:15;
out = bitcount(data);
bin_representation = cellfun(@(x) dec2bin(x,4),num2cell(data),'uni',0)';
nBits = num2cell(out)';
disp([num2cell(data(:)), bin_representation, nBits])

% Counts the #'0', default # of bits to examine = 8
data = 0:15;
out = bitcount(data,0);   % out = [8 7 7 6 7 6 6 5 7 6 6 5 6 5 5 4]

% Counts the #'0', # of bits to examine = 4
 data = 0:15;
out = bitcount(data,0,4); % out = [4 3 3 2 3 2 2 1 3 2 2 1 2 1 1 0]

Created by Noah Safra May 2025