Copy or view numpy subarray using boolean indexing

Question

Given a 2D numpy array, i.e.;

import numpy as np

data = np.array([
     [11,12,13],
     [21,22,23],
     [31,32,33],
     [41,42,43],         
     ])

I need to both create a new sub-array or modify the selected elements in place based on two masking vectors for the desired rows and columns;

rows = [False, False, True, True]
cols = [True, True, False]

Such that

print subArray

# [[31 32]
#  [41 42]]

Note that copy and view (as in the title) are very different things. — askewchan
– askewchan, Commented Feb 28, 2013 at 20:05

Pierre GM · Accepted Answer · 2012-09-15 11:44:33Z

5

First, make sure that your rows and cols are actually boolean ndarrays, then use them to index your data

rows = np.array([False, False, True, True], dtype=bool)
cols = np.array([True, True, False], dtype=bool)
data[rows][:,cols]

Explanation If you use a list of booleans instead of an ndarray, numpy will convert the False/True as 0/1, and interpret that as indices of the rows/cols you want. When using a bool ndarray, you're actually using some specific NumPy mechanisms.

edited Sep 15, 2012 at 11:44

answered Sep 15, 2012 at 11:38

Pierre GM

20.5k3 gold badges58 silver badges67 bronze badges

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

1 Comment

Marcus Jones Over a year ago

Great, thats clear now thanks. Just not as straight-forward as Matlab for same result. I have a follow up - how do I do this in-place, without creating a new array?

Collectives™ on Stack Overflow

Copy or view numpy subarray using boolean indexing

1 Answer 1

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related