Floating representation in python

Question

It is pretty self-explanatory (see when pct = 0.07) :

def what_the_foo(range_pct):
    for pct in range_pct:
        print("pct: %s, (pct*100): %s, int(pct * 100) : %s" %
              (pct, pct*100, int(pct*100)))

range_pct
Out[9]: array([ 0.01,  0.02,  0.03,  0.04,  0.05,  0.06,  0.07,  0.08,  0.09,  0.1 ])

what_the_foo(range_pct)
pct: 0.01, (pct*100): 1.0, int(pct * 100) : 1
pct: 0.02, (pct*100): 2.0, int(pct * 100) : 2
pct: 0.03, (pct*100): 3.0, int(pct * 100) : 3
pct: 0.04, (pct*100): 4.0, int(pct * 100) : 4
pct: 0.05, (pct*100): 5.0, int(pct * 100) : 5
pct: 0.06, (pct*100): 6.0, int(pct * 100) : 6
pct: 0.07, (pct*100): 7.0, int(pct * 100) : 6
pct: 0.08, (pct*100): 8.0, int(pct * 100) : 8
pct: 0.09, (pct*100): 9.0, int(pct * 100) : 9
pct: 0.1, (pct*100): 10.0, int(pct * 100) : 10

I know this must be about the floating representation of 0.07, but everything seems fine outside the function :

0.07 * 100
Out[11]: 7.000000000000001

int(0.07 * 100)
Out[12]: 7

Why int(0.07 * 100) returns 6 inside the function and 7 oustide ? Actually, why would it return 6 anyway ?

This is due to float representation. `int` will simply `floor` that giving you the closest integer (`7`). — Dimitris Fasarakis Hilliard, Oct 04 '16 at 16:20
Because you're casting it to `int` in the second example, so `.000000000000001` precision is removed. Why would you expect a decimal point in an `int`? — roganjosh, Oct 04 '16 at 16:22
Also, things might be self-explanatory until someone misjudges your confusion. Be explicit with your questions so we're all on the same page :-) — Dimitris Fasarakis Hilliard, Oct 04 '16 at 16:22
@FlorianGauthier running your function which the same input data gives me expected results. — Dimitris Fasarakis Hilliard, Oct 04 '16 at 16:29
@JimFasarakis-Hilliard also confirmed in Python 2.7. Works either with python list or `numpy` array. — roganjosh, Oct 04 '16 at 16:30

Floating representation in python

0 Answers0