Hi Greg,

I don't know how specific it is to NumPy, but that's definitely the correct way 
to talk about it in NumPy, and your understanding in your example is spot-on. 
This is true of many NumPy functions.

Juan.

On 30 Dec. 2016, 9:08 AM +1100, greg g <[email protected]>, wrote:
> Thanks
> Is this a numpy specific terminology ?
> For a multidimensionnal array with dimension=n and size l1 x l2 x ... x ln, 
> does "along axis=0" mean that l2 x..x ln operations are performed scrolling 
> first dimension, each operation on l1 elements,  and that an array with 
> dimension n-1 and size l2 x..x ln containing the operations results is 
> returned?
> ( Finally I'm not sure this sentence really clarify ... ;-) )
>
> De : scikit-learn <[email protected]> de la 
> part de Jacob Schreiber <[email protected]>
> Envoyé : jeudi 29 décembre 2016 20:22
> À : Scikit-learn user and developer mailing list
> Objet : Re: [scikit-learn] numpy.amin behaviour with multidimensionnal arrays
>
> It means that instead of returning the minimum value anywhere in the entire 
> matrix, it will return the minimum value for each column or each row 
> depending on which axis you put in, so a vector instead of a scalar.
>
> > On Thu, Dec 29, 2016 at 6:00 AM, greg g <[email protected]> wrote:
> > > Hi,
> > > I would like to understand the behaviour of the scipy.spatial.kdtree 
> > > class that uses numpy.amin function.
> > > In the numpy.amin description, we find that it returns the "minimum value 
> > > along a given axis"
> > > What does it mean exactly ?
> > >
> > > Thanks for any help
> > > Gregory
> > >
> > >
> > > _______________________________________________
> > > scikit-learn mailing list
> > > [email protected]
> > > https://mail.python.org/mailman/listinfo/scikit-learn
> > >
>
> _______________________________________________
> scikit-learn mailing list
> [email protected]
> https://mail.python.org/mailman/listinfo/scikit-learn
_______________________________________________
scikit-learn mailing list
[email protected]
https://mail.python.org/mailman/listinfo/scikit-learn

Reply via email to