[Numpy-discussion] Wanted: Numeric/NumPy compatible array creation expression

Travis Oliphant oliphant.travis at ieee.org
Tue Nov 28 01:10:49 CST 2006


konrad.hinsen at laposte.net wrote:
> I am looking for a way to create an array of the same type as another  
> given array (but of different shape) that works both with Numeric and  
> NumPy without being unreasonably slow. In other words, I am looking  
> for a a replacement for the expression
>
> 	array2 = Numeric.zeros(shape, array1.typecode())
>
> that will *also* (not *only*) work under NumPy (where typecode()  
> became dtype.char). An obvious idea is reshaping array1 and  
> multiplying by 0., but that can become quite costly.
>
> Any ideas?
>   

Matplotlib uses the numerix layer to handle compatibility between all 
three array packages. 

It defines the typecode function in that package and then uses it 
instead of the method.


-Travis



More information about the Numpy-discussion mailing list