python: generating a histogram

this:

 numpy.histogram([1,3,2,3,1,1,1,1,2,3,2,5,6,6],bins=numpy.arange(0,7,1))
开发者_JAVA百科

yields:

(array([0, 5, 3, 3, 0, 3]), array([0, 1, 2, 3, 4, 5, 6]))

why does it count three 6's? there are only 2!

because bins defines the bin edges you need to add one more bin

numpy.histogram([1,3,2,3,1,1,1,1,2,3,2,5,6,6],bins=numpy.arange(0,8,1))

There is one 5 and two 6's in the last bin. Quoting the doc "All but the last (righthand-most) bin is half-open", so the last bin includes the 2 6's.

I believe this is related to the semantics of the first and last bins being underflow and overflow. Your total bucket count is correct.

It looks like it's lumping the 5's and 6's together - maybe the last bin is 5 to 6, inclusive?

Edit: Looking at the docs, I'm guessing that 6 is the rightmost edge, so values between five and six are in that final bin.