Slice a 3D numpy array by a list of indices

Question

I can't seeem to figure out how to slice an array so that I get the indices of interest in the 3rd dimension. Here is an example 3D numpy array.

data = np.arange(60).reshape(5,4,3)  
print data

[[[ 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 50]   [51 52 53]   [54 55 56]   [57 58 59]]]

Now here are the indices that I would like to grab from the 3rd dimension.

indices_of_interest = np.random.randint(3, size=5) print indices_of_interest

[0 2 2 2 0]

So basically I want values

[[[ 0] [ 3] [ 6] [ 9]]

[[14] [17] [20] [23]]

[[26] [29] [32] [35]]

[[38] [41] [44] [47]]

[[48] [51] [54] [57]]]

Is there some way to do this? When I try and index the array directly it broadcasts the dimension rather than providing me a subset of the data.

Answer 1

We can use advanced-indexing to grab them by the third dim -

data[np.arange(len(indices_of_interest)),:, indices_of_interest]

Sample run -

In [65]: data = np.arange(60).reshape(5,4,3)

In [66]: indices_of_interest = [0,2,2,2,0]

In [67]: data[np.arange(len(indices_of_interest)),:, indices_of_interest]
Out[67]: 
array([[ 0,  3,  6,  9],
       [14, 17, 20, 23],
       [26, 29, 32, 35],
       [38, 41, 44, 47],
       [48, 51, 54, 57]])