numpy-discussion

[Numpy-discussion] astype() behaviour

[Joris De R. <jo...@st...>]

Hi,

Is the following behaviour of astype() intentional in NumPy 1.0?

>>> x = array([1,2,3])
>>> x.astype(None)
array([ 1.,  2.,  3.])

That is, the int32 is converted to float64.

J.


Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm

Re: [Numpy-discussion] astype() behaviour

[Francesc A. <fa...@ca...>]

A Divendres 27 Octubre 2006 17:58, Joris De Ridder va escriure:
> Hi,
>
> Is the following behaviour of astype() intentional in NumPy 1.0?
>
> >>> x =3D array([1,2,3])
> >>> x.astype(None)
>
> array([ 1.,  2.,  3.])
>
> That is, the int32 is converted to float64.

Yes, I think the behaviour is intended. This is because 'float64' is the=20
default type in NumPy from some months ago (before the default was 'int_')

HTH,

=2D-=20
>0,0<   Francesc Altet =A0 =A0 http://www.carabos.com/
V   V   C=E1rabos Coop. V. =A0=A0Enjoy Data
 "-"

Re: [Numpy-discussion] astype() behaviour

[Joris De R. <jo...@st...>]

On Friday 27 October 2006 19:01, Francesc Altet wrote:
   [FA]: A Divendres 27 Octubre 2006 17:58, Joris De Ridder va escriure:
   [FA]: > Hi,
   [FA]: >
   [FA]: > Is the following behaviour of astype() intentional in NumPy 1.0?
   [FA]: >
   [FA]: > >>> x = array([1,2,3])
   [FA]: > >>> x.astype(None)
   [FA]: >
   [FA]: > array([ 1.,  2.,  3.])
   [FA]: >
   [FA]: > That is, the int32 is converted to float64.
   [FA]: 
   [FA]: Yes, I think the behaviour is intended. This is because 'float64' is the 
   [FA]: default type in NumPy from some months ago (before the default was 'int_')

OK, updated the astype() example in the www.scipy.org/Numpy_Example_List .

J.


Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm