Python 2D array processing

Question

I have an input array of the form

I want to calculate the density map and get output in the form (summing up all the values around a value including the value)

3 4 2 2 1
4 5 2 2 1
3 4 3 3 2
2 2 2 2 2
1 1 2 2 2

import numpy as np

def get_value(arr,i,j):
    """
    This function will return the value at the index if present. If list index out of range, it will return 0
    """
    try:
        return arr[i][j]
    except:
        return 0

#input array
arr = [[1, 0, 0, 0, 1],
 [1, 1, 1, 0, 0],
 [1, 0, 0, 0, 0],
 [0, 0, 0, 1, 1],
 [0, 1, 0, 0, 0]]

np_arr = np.array(arr) # converting to numpy array

r =1 # initializing the radius r as 1
n =5 # initializing the size of the input matrix as 5
np_arr_result = np.zeros(shape=(5,5)) # initializing the output array

for i in range(0,5):
    for j in range(0,5):
        np_arr_result[i][j] = get_value(np_arr,i,j) + get_value(np_arr,i-r,j-r) + get_value(np_arr,i-r,j)  + get_value(np_arr,i-r,j+r) + get_value(np_arr,i,j-r)  + get_value(np_arr, i, j+r) + get_value(np_arr, i+r, j-r) + get_value(np_arr, i+r, j) + get_value(np_arr, i+r, j+r)

print("output")
print(np_arr_result)

However, I am not getting the right output and getting the result as:

output
[[ 5.  5.  3.  2.  1.]
 [ 5.  5.  2.  2.  1.]
 [ 4.  4.  3.  3.  2.]
 [ 3.  2.  2.  2.  2.]
 [ 2.  1.  2.  2.  2.]]

What could have gone wrong?

I think the misunderstanding arises from assuming arr[0][-1] will raise an IndexError, when it will actually return 1 (the last value of the first row). This indexing operation will return something for any negative number up to -5, which I think is not the intended behavior, given your code. — Tomas Farias
– Tomas Farias, Commented Jul 3, 2018 at 15:08

Nils Werner · Accepted Answer · 2018-07-03 15:02:53Z

2

This is a perfect task for a 2D convolution:

data = numpy.array([
    [1, 0, 0, 0, 1],
    [1, 1, 1, 0, 0],
    [1, 0, 0, 0, 0],
    [0, 0, 0, 1, 1],
    [0, 1, 0, 0, 0],
])

kernel = numpy.ones((3, 3), dtype=data.dtype)
# array([[1, 1, 1],
#        [1, 1, 1],
#        [1, 1, 1]])

scipy.signal.convolve2d(data, kernel, mode='same')
# array([[3, 4, 2, 2, 1],
#        [4, 5, 2, 2, 1],
#        [3, 4, 3, 3, 2],
#        [2, 2, 2, 2, 2],
#        [1, 1, 2, 2, 2]])

answered Jul 3, 2018 at 15:02

Nils Werner

37.2k7 gold badges85 silver badges108 bronze badges

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

Comments

one · Accepted Answer · 2018-07-03 16:01:26Z

0

That is because for numpy, the index -1 is for the last element, so it will not assert IndexError.

just modify the get_value() function as :

def get_value(arr,i,j):
    """
    This function will return the value at the index if present. If list index out of range, it will return 0
    """
    if (i>-1 and j >-1 and i < arr.shape[0] and j < arr.shape[1]):
        return arr[i][j]
    else:
        return 0

answered Jul 3, 2018 at 16:01

one

2,6252 gold badges22 silver badges40 bronze badges

Collectives™ on Stack Overflow

Python 2D array processing

2 Answers 2

Comments

Comments

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Related