Python indexing 2D array

Question

How can i do indexing of a 2D array column wise. For example-

array([[ 0,  1,  2,  3,  4],
   [ 5,  6,  7,  8,  9],
   [10, 11, 12, 13, 14],
   [15, 16, 17, 18, 19],
   [20, 21, 22, 23, 24],
   [25, 26, 27, 28, 29],
   [30, 31, 32, 33, 34],
   [35, 36, 37, 38, 39],
   [40, 41, 42, 43, 44],
   [45, 46, 47, 48, 49]])

This is a 2D array. I can access to it column wise by using a[:,0] which will give me the first column. But if I want to read all column at a time and want to pick values for example

[5]

[10][15]

[20][25][37]

then it should pick the values like

I know it must be easy. But i am learning the stuff.

DSM · Accepted Answer · 2013-05-07 09:53:55Z

2

If you want [5] to give 20, you must be starting to count from 1. Since Python starts counting from 0, that's a habit to break now: it'll only cause headaches.

I'm not sure what output format you want because numpy doesn't support ragged arrays, but maybe

>>> idx = np.array([5, 10, 15, 20, 25, 37])
>>> a.T.flat[idx-1]
array([20, 45, 21, 46, 22, 33])

would suffice? Here I had to take the transpose, view it as a flat array, and then subtract 1 from the indices to match the way you seem to be counting.

We can use a list instead of an array (but then we'd need to do a listcomp or something to subtract the 1s.) For example:

>>> a.T.flat[[4, 9, 14, 19, 24, 36]]
array([20, 45, 21, 46, 22, 33])

edited May 7, 2013 at 9:53

answered May 7, 2013 at 9:40

DSM

355k67 gold badges606 silver badges504 bronze badges

Sign up to request clarification or add additional context in comments.

4 Comments

haq Over a year ago

Thank you for your answer. the transpose, flat array is something new for me as I am learning still. I will try to understand them well.

Henry Gomersall Over a year ago

Just to be clear, flat isn't a view but an iterator (albeit one that supports fancy indexing). You can return a genuine view by using ravel() if you want to play around with the array itself as a 1D object without making copies (or more generally, reshape()).

DSM Over a year ago

@HenryGomersall: ah, I wasn't even thinking about that. I just meant "view" as "think of it as"/"treat it as", which is an unfortunate choice of words given that "view" has a technical meaning in a numpy context.

Henry Gomersall Over a year ago

@DSM :) I realise that, I just wanted Enamul to be clear since he mentioned not knowing about flat.

Collectives™ on Stack Overflow

Python indexing 2D array

1 Answer 1

4 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

4 Comments

Your Answer

Sign up or log in

Post as a guest

Related