[Numpy-discussion] Bug in logaddexp2.reduce

[T J]

On Wed, Mar 31, 2010 at 1:21 PM, Charles R Harris
<charlesr.harris at gmail.com> wrote:
>
> Looks like roundoff error.
>

So this is "expected" behavior?

In [1]: np.logaddexp2(-1.5849625007211563, -53.584962500721154)
Out[1]: -1.5849625007211561

In [2]: np.logaddexp2(-0.5849625007211563, -53.584962500721154)
Out[2]: nan

In [3]: np.log2(np.exp2(-0.5849625007211563) + np.exp2(-53.584962500721154))
Out[3]: -0.58496250072115608

Shouldn't the result at least behave nicely and just return the larger value?