[Numpy-discussion] NumPy C-API equivalent of np.float64()

Matthew Brett matthew.brett@gmail....
Wed Dec 29 11:48:05 CST 2010


Hi,

On Wed, Dec 29, 2010 at 5:37 PM, Robert Bradshaw
<robertwb@math.washington.edu> wrote:
> On Wed, Dec 29, 2010 at 9:05 AM, Keith Goodman <kwgoodman@gmail.com> wrote:
>> On Tue, Dec 28, 2010 at 11:22 PM, Robert Bradshaw
>> <robertwb@math.washington.edu> wrote:
>>> On Tue, Dec 28, 2010 at 8:10 PM, John Salvatier
>>> <jsalvati@u.washington.edu> wrote:
>>>> Wouldn't that be a cast? You do casts in Cython with <double>(expression)
>>>> and that should be the equivalent of float64 I think.
>>>
>>> Or even <numpy.float64_t >(expression) if you've cimported numpy
>>> (though as mentioned this is the same as double on every platform I
>>> know of). Even easier is just to use the expression in a the right
>>> context and it will convert it for you.
>>
>> That will give me a float object but it will not have dtype, shape,
>> ndim, etc methods.
>>
>>>> m = np.mean([1,2,3])
>>>> m
>>   2.0
>>>> m.dtype
>>   dtype('float64')
>>>> m.ndim
>>   0
>>
>> using <np.float64_t> gives:
>>
>> AttributeError: 'float' object has no attribute 'dtype'

Forgive me if I haven't understood your question, but can you use
PyArray_DescrFromType with e.g  NPY_FLOAT64 ?

Best,

Matthew


More information about the NumPy-Discussion mailing list