Fwd: numpy/matlab compatibility

[Andrea Ambu]

I replied to Matt only ARGH!


---------- Forwarded message ----------
From: Andrea Ambu <andreambu at gmail.com>
Date: 25 January 2011 22:36
Subject: Re: numpy/matlab compatibility
To: Matt Funk <mafunk at nmsu.edu>




On Tue, Jan 25, 2011 at 9:13 PM, Matt Funk <mafunk at nmsu.edu> wrote:
>
> Hi,
>
> i am fairly new to python. I was wondering of the following is do-able
> in python:
>
> 1) a = rand(10,1)
> 2) Y = a
> 3) mask = Y > 100;
> 4) Y(mask) = 100;
> 5) a = a+Y
>

No. Not like that.
You do literally:
a = rand(10, 1)
Y = a
mask = Y>100
Y = where(mask, 100, Y)
a = a+Y

More Pythonically:
a = rand(10, 1)
a = where(a > 100, a + 100, a + a)

For those who don't speak Matlab:
1) a = rand(10,1) ; generates a 10x1 matrix for random number 0 < n < 1
2) Y = a
3) mask = Y > 100; similar to: mask = [i>100 for i in Y]
4) Y(mask) = 100; sets to 100 elements of Y with index i for which
mask[i] = True
5) a = a+Y ; sums the two matrices element by element (like you do in
linear algebra)

Anyway... rand generates number from 0 up to 1 (both in python and
matlab)... when are they > 100?

>
> Basically i am getting stuck on line 4). I was wondering if it is
> possible or not with python?
> (The above is working matlab code)
>
> thanks
> matt
> --
> http://mail.python.org/mailman/listinfo/python-list




-- 
Andrea