Numpy slicing a fixed length on two axis based on different starting index given by two arrays

Question

For example, I have nparray:

a = np.arange(48).reshape((3,4,4))
'''
[[[ 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]]]
'''

I have two arrays that used as the starting point of slicing on axis=1, axis=2 respectively:

b1 = [0,1,2]
b2 = [1,0,0]

I want to achieve, a slicing like:

a[:,b1:b1+2, b2:b2+2] # but this syntax is wrong

To get

[
[
[1,2]
[5,6]
]

[
[20 21]
[24 25]
]

[
[40 41]
[44 45]
]
]

Please let me know if you know the proper syntax for doing this?

kederrac · Accepted Answer · 2020-03-15 22:55:18Z

1

you can use the built-in functions enumerate with zip:

list(a[i][f:f+2, s:s+2].tolist() for i, (f, s) in enumerate(zip(b1, b2)))

output:

[[[1, 2], [5, 6]], [[20, 21], [24, 25]], [[40, 41], [44, 45]]]

answered Mar 15, 2020 at 22:55

kederrac

17.4k6 gold badges36 silver badges58 bronze badges

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

1 Comment

Jason Over a year ago

Thanks for mentioning zip, I almost got it by breaking it down to two steps without using zip, but with zip, it could be done within the list comprehension. My final solution is: d = np.array([a[i,c1:c1+2, c2:c2+2] for i, (c1,c2) in enumerate(zip(b1,b2))])

Collectives™ on Stack Overflow

Numpy slicing a fixed length on two axis based on different starting index given by two arrays

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related