RE: [Bmelibpy-devel] rank function

[Dimitri D'O. <dim...@fs...>]

Didrik,

What do you propose as a solution for this potential bug with "rank"?

I frequently use "rank" to test the dimensionality of a matrix or a =
vector.
Should I replace it by "len(A.shape)" or import it as Nrank?

The second solution seems to be clearer for me. But then how do I have =
to
import the modules?

from scipy import *
from scipy import rank as Nrank
from pylab import *

Another solution would be to import "rank" as "dim" to match the Matlab
function name.

Or

from scipy import *
from scipy import Numeric as N
from pylab import *

With this last import scheme, N.rank should be used to test =
dimensionality
while rank gives the rank of a matrix (from pylab).

Which would be the best way to import the modules?

Dimitri
_______________________________________________
Dr. Dimitri D'Or
Agricultural Engineer, Ph.D.=20
Environmental Resources Consultant for FSS International
Tienne-de-Mont, 10
5140 Sombreffe
Belgique
=20
T=E9l. : +32-71-81.43.60
e-mail : dim...@fs...
Web site : http://www.fssintl.com/
Le groupe de travail fran=E7ais GeoSiPol : http://www.geosipol.org
=20
Upcoming events:
Sites pollu=E9s : Pratique de l'investigation et de l'assainissement,
Neuch=E2tel (Suisse)
Cours de formation continue du 31 mai au 3 juin et 13-14 octobre 2005.
_______________________________________________
> -----Message d'origine-----
> De=A0: bme...@li... =
[mailto:bmelibpy-devel-
> ad...@li...] De la part de Didrik Pinte
> Envoy=E9=A0: jeudi 14 juillet 2005 15:05
> =C0=A0: bme...@li...
> Objet=A0: [Bmelibpy-devel] rank function
>=20
> For your information, potential bug source:
>=20
> ----------------------------------------------------------------------
> Help on function rank in module Numeric:
>=20
> rank(a)
>     Get the rank of sequence a (the number of dimensions, not a matrix
> rank)
>     The rank of a scalar is zero.
>=20
> ----------------------------------------------------------------------
>=20
> Help on function rank in module matplotlib.mlab:
>=20
> rank(x)
>     Returns the rank of a matrix.
>     The rank is understood here as the an estimation of the number of
>     linearly independent rows or columns (depending on the size of the
>     matrix).
>     Note that numerix.mlab.rank() is not equivalent to Matlab's rank.
>     This function is!
>=20
> Didrik