So.. yeah... that's the question. I've to count the number of 1 in a positive number. So for this, the approach was to do the bitwise and . We take the number and subtract 1 from it and and the two number until it becomes zero.
Here is the code:

def countOnes(num):
    cnt = 0
	while num:
	    num = num & (num-1)
		cnt += 1
	
	return cnt

So, for num = 3, let's simulate the code:

1) cnt = 0
	num = 3(11)
	num - 1 = 2(10)
	num = num & (num-1) = 
	11 
  & 10
  ------
  = 10
	cnt = 1
2) cnt = 1
	num = 2(10)
	num - 1 = 1(01)
	num = num & (num-1) = 
	10 
  & 01 
  -----
  = 00
	cnt = 2
3) The loop breaks as now num = 0

So this method works. But what is the significant here? What's the pattern? Why does this work? I mean I understand that you and two adjacent number and after some times it becomes 0 but why? Does it have any proof?

Any answer to the question is helpful.
Thanks