Reputation: 45
Slicing, indexing and iterating over 2D numpy arrays
I am enrolled in a beginners class Python at my university. We've got a programming assignment which I'm stuck on.
We got an assignment to find a route between two point in a map, the map is an 2D numpy array. One of the first tasks is to convert the array consisting of free road(1) and buildings(0) to an array where all the locations surrounding a building (left, right, below, above) are converted to parking spots (-1)
I started with writing a function to generate a 2D numpy array in the right size:
def get_map():
map = np.random.randint(2, size=(12, 10))
return map
Now I want to write another function that takes the map as an argument, and returns the map where the parking spots are converted from 1's to -1's.
def adjusted_map(map):
map_adjusted =
return map_adjusted
I am mostly stuck on the elements above and below the 0's. Left and right I can do because that's no different from 1D arrays or normal lists, strings etc. Im sorry if this is a stupid question but I looked into the numpy documentation about indexing, slicing and iterating over numpy arrays, but I could not find a solution for my problem.
Upvotes: 1
Views: 195
Answers (2)
Reputation: 53089
Here is one simple way using standard numpy techniques:
1) Make a map consisting of 3x3 blocks with 80% road
>>> map_ = np.kron(np.random.random((6, 5)) < 0.8, np.ones((3, 3), int))
>>> map_
array([[1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1],
[1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1],
[1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1],
[1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 1],
[0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 1],
[0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 1],
[0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 1],
[1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 1],
[1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 1],
[0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0],
[0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0],
[0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0]])
2) Invert the map and zero-pad it
>>> helper = np.pad(1-map_, ((1, 1), (1, 1)), 'constant')
>>> helper
array([[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0],
[0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0],
[0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0],
[0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0],
[0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0],
[0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0],
[0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0],
[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]])
3) Cut out shifted versions (up, down, left, right) of the inverse map, use (bitwise) or to find all neighbors of buildings, use and to only keep neighbors that are road
>>> parking = map_ & (helper[2:, 1:-1] | helper[:-2, 1:-1] | helper[1:-1, 2:] | helper[1:-1, :-2])
>>> parking
array([[0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0],
[0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0],
[0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0],
[0, 0, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0],
[0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0],
[1, 1, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0],
[0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0],
[0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0],
[0, 0, 0, 1, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 1, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0],
[1, 1, 1, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0],
[1, 1, 1, 0, 0, 1, 0, 0, 0, 1, 0, 0, 1, 1, 1],
[0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 0, 1, 0, 0, 0],
[0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 0, 1, 0, 0, 0],
[0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 0, 1, 0, 0, 0]])
4) Mark the parking spots on the map
>>> result = map_ - 2*parking
>>> result
array([[ 1, 1, -1, 0, 0, 0, 0, 0, 0, -1, 1, 1, 1, 1, 1],
[ 1, 1, -1, 0, 0, 0, 0, 0, 0, -1, 1, 1, 1, 1, 1],
[ 1, 1, -1, 0, 0, 0, 0, 0, 0, -1, 1, 1, 1, 1, 1],
[ 1, 1, -1, 0, 0, 0, -1, -1, -1, 1, 1, 1, 1, 1, 1],
[ 1, 1, -1, 0, 0, 0, -1, 1, 1, 1, 1, 1, 1, 1, 1],
[-1, -1, -1, 0, 0, 0, -1, -1, -1, 1, 1, 1, 1, 1, 1],
[ 0, 0, 0, -1, -1, -1, 0, 0, 0, -1, 1, 1, 1, 1, 1],
[ 0, 0, 0, -1, 1, -1, 0, 0, 0, -1, 1, 1, 1, 1, 1],
[ 0, 0, 0, -1, 1, -1, 0, 0, 0, -1, 1, 1, 1, 1, 1],
[ 0, 0, 0, -1, 1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1],
[ 0, 0, 0, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[ 0, 0, 0, -1, 1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1],
[-1, -1, -1, 1, 1, -1, 0, 0, 0, -1, 1, 1, 1, 1, 1],
[ 1, 1, 1, 1, 1, -1, 0, 0, 0, -1, 1, 1, 1, 1, 1],
[-1, -1, -1, 1, 1, -1, 0, 0, 0, -1, 1, 1, -1, -1, -1],
[ 0, 0, 0, -1, 1, -1, 0, 0, 0, -1, 1, -1, 0, 0, 0],
[ 0, 0, 0, -1, 1, -1, 0, 0, 0, -1, 1, -1, 0, 0, 0],
[ 0, 0, 0, -1, 1, -1, 0, 0, 0, -1, 1, -1, 0, 0, 0]])
5) Bonus: Prettify
>>> symbols = np.array(('x', '.', 'P'))
>>> rowtype = f'U{map_.shape[1]}'
>>> rowtype
'U15'
>>> print('\n'.join(symbols[map_].view(rowtype).ravel()))
...xxxxxx......
...xxxxxx......
...xxxxxx......
...xxx.........
...xxx.........
...xxx.........
xxx...xxx......
xxx...xxx......
xxx...xxx......
xxx............
xxx............
xxx............
......xxx......
......xxx......
......xxx......
xxx...xxx...xxx
xxx...xxx...xxx
xxx...xxx...xxx
>>> print('\n'.join(symbols[result].view(rowtype).ravel()))
..PxxxxxxP.....
..PxxxxxxP.....
..PxxxxxxP.....
..PxxxPPP......
..PxxxP........
PPPxxxPPP......
xxxPPPxxxP.....
xxxP.PxxxP.....
xxxP.PxxxP.....
xxxP..PPP......
xxxP...........
xxxP..PPP......
PPP..PxxxP.....
.....PxxxP.....
PPP..PxxxP..PPP
xxxP.PxxxP.Pxxx
xxxP.PxxxP.Pxxx
xxxP.PxxxP.Pxxx
Upvotes: 2
Reputation: 6101
so you need something like this:
import numpy as np
mapper = np.random.randint(2, size=(12, 10))
buildings = np.nonzero(mapper)
nonzero_row = buildings[0]
nonzero_col = buildings[1]
for row, col in zip(nonzero_row, nonzero_col):
if row > 0:
mapper[row-1, col] = -1
if col > 0:
mapper[row, col-1] = -1
if row < mapper.shape[0]:
mapper[row+1,col] = -1
if col < mapper.shape[1]:
mapper[row,col+1] = -1
Upvotes: 0
Related Questions
- Slicing multi-dimensional array with another array
- Pythonic way of retrieving elements of 2D array by indices
- 2-D arrays - Slicing and Indexing
- Slicing Numpy Array by 2 index arrays
- Slicing a 2D array using indices 1D array
- Slicing using arrays/indices
- slicing 2d numpy array
- Slicing a 2D numpy array in python
- numpy: slicing and vectorized looping with 1d and 2d arrays
- Slicing 2D arrays using indices from arrays in python