[Numpy-discussion] abs for max negative integers - desired behavior?
Tue Oct 11 14:51:32 CDT 2011
On Tue, Oct 11, 2011 at 3:16 PM, Charles R Harris
> On Tue, Oct 11, 2011 at 12:23 PM, Matthew Brett <email@example.com>
>> I recently ran into this:
>> In : arr = np.array(-128, np.int8)
>> In : arr
>> Out: array(-128, dtype=int8)
>> In : np.abs(arr)
>> Out: -128
> This has come up for discussion before, but no consensus was ever reached.
> One solution is for abs to return an unsigned type, but then combining that
> with signed type of the same number of bits will cause both to be cast to
> higher precision. IIRC, matlab was said to return +127 as abs(-128), which,
> if true, is quite curious.
octave-3.2.3:1> a = int8([-128, 127])
Matlab is the same. That is curious...
More information about the NumPy-Discussion