[Numpy-discussion] A question about dtype syntax

Pierre GM pgmdevlist@gmail....
Wed Aug 31 05:42:21 CDT 2011


On Aug 31, 2011, at 12:20 PM, Jean-Baptiste Marquette wrote:

> 
> Hi Pierre,
> 
> Thanks for the guess. Unfortunately, I got the same error:
> 
> [('bs3000k.cat', 280.60341, -7.09118, 9480, 0.2057, 0.14)]
> Traceback (most recent call last):
>   File "/Users/marquett/workspace/Distort/src/StatsSep.py", line 40, in <module>
>     StatsAll = np.array(np.asarray(Stats), dtype=('a15, f8, f8, i4, f8, f8'))
> ValueError: could not convert string to float: bs3000k.cat

Of course, silly me

Your line 40 is actually
>>>    StatsAll = np.array(np.asarray(Stats), dtype=('a15, f8, f8, i4, f8, f8'))

With np.asarray(Stats), you're trying to load Stats as an array using a dtype of float by default. Of course, np.asarray is choking on the first element.

So, try to use instead
>>>    StatsAll = np.array(Stats, dtype=('a15, f8, f8, i4, f8, f8'))




More information about the NumPy-Discussion mailing list