Le 18/07/2014 20:42, Charles G. Waldman a écrit :
Well, if the goal is "shorthand", typing numpy.array(numpy.mat()) won't please many users.
But the more I think about it, the less I think Numpy should support this (non-Pythonic) input mode. Too much molly-coddling of new users! When doing interactive work I usually just type:
np.array([[1,2,3], ... [4,5,6], ... [7,8,9]])
which is (IMO) easier to read: e.g. it's not totally obvious that "1,0,0;0,1,0;0,0,1" represents a 3x3 identity matrix, but
[[1,0,0], [0,1,0], [0,0,1]]
is pretty obvious.
Compare what's comparable: [[1,0,0], [0,1,0], [0,0,1]] vs "1 0 0;" "0 1 0;" "0 0 1" or """ 1 0 0; 0 1 0; 0 0 1 """ [[1,0,0], [0,1,0], [0,0,1]] vs "1 0 0; 0 1 0; 0 0 1"
The difference in (non-whitespace) chars is 19 vs 25, so the "shorthand" doesn't seem to save that much.
Well, it's easier to type "" (twice the same character) than [], and you have no risk in swapping en opening and a closing bracket. In addition, you have to use AltGr on some keyboards to get the brackets. It doesn't boils down to a number of characters.
Just my €0.02,
- C
On Fri, Jul 18, 2014 at 10:05 AM, Alan G Isaac <alan.isaac@gmail.com> wrote:
On 7/18/2014 12:45 PM, Mark Miller wrote:
If the true goal is to just allow quick entry of a 2d array, why not just advocate using a = numpy.array(numpy.mat("1 2 3; 4 5 6; 7 8 9"))
It's even simpler: a = np.mat(' 1 2 3;4 5 6;7 8 9').A
I'm not putting a dog in this race. Still I would say that the reason why such proposals miss the point is that there are introductory settings where one would like to explain as few complications as possible. In particular, one might prefer *not* to discuss the existence of a matrix type. As an additional downside, this is only good for 2d, and there have been proposals for the new array builder to handle other dimensions.
fwiw, Alan Isaac
_______________________________________________ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion
NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion