Question

Take a look at this:

print 41063625 ** (1.0/3)  # cube-root(41063625) = 345
print int(345.0)
print int(41063625 ** (1.0/3))

It outputs:

345.0
345
344

I was expecting the last line to be 345, since I was expecting int(41063625 ** (1.0/3)) to equal int(345.0) to in turn equal 345, as the other two outputs suggest. However, this is evidently not the case. Can anyone give me any insight as to what’s going on here?

Answer 1

Print (or rather float.__str__) is rounding the output.

In [22]: str( 41063625 ** (1.0/3) )
Out[22]: '345.0'

The floating point representation for
41063625 ** (1.0/3) is less than 345, so when you take the int of it, you get 344 rather than 345.

In [15]: 41063625 ** (1.0/3)
Out[15]: 344.9999999999999

In [16]: int(345.0)
Out[16]: 345

In [17]: int(41063625 ** (1.0/3))
Out[17]: 344

If you want the closest int, you could use round:

In [18]: round(41063625 ** (1.0/3))
Out[18]: 345.0

or, to get an int:

In [19]: int(round(41063625 ** (1.0/3)))
Out[19]: 345