Re: [Matplotlib-users] reshape not like matlab

[Chris B. <Chr...@no...>]

Dimitri D'Or wrote:
> I was surprised to discover recently that the RESHAPE function in Python 
> and Matlab do not yield the same result. In fact, Python reshapes along 
> the rows while Matlab does it along the columns.

> Does anyone have an explanation for this?

AFAIK, the reason is that MATLAB has Fortran in it's history, ans thus 
uses Fortran semantics laying out arrays, whereas Numeric and numarray 
use C semantics for laying out arrays.

NumPy is not Matlab, nor is it trying to be.

> Is it possible to change the 
> axis priority like e.g., in concatenate?

It doesn't look like it, from the docs. The reason is that re-shape 
doesn't move any data around in memory, it just changes the way the data 
is interpreted.

What you want really is a transpose, as you've discovered, but think 
about whether you really want that... it might make more sense to just 
stick with NumPy's natural ordering.

This, by the way is a Numeric or numarray question, you'd be best off 
referring to the docs for those packages, and asking questions on the 
numpy-discussion list.

-Chris



-- 
Christopher Barker, Ph.D.
Oceanographer
                                     		
NOAA/OR&R/HAZMAT         (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception

Chr...@no...