I have a 25 x 25 numpy array filled with 0s. I want to randomly replace 0s within the array with 1s, with the number of 0s to be replaced defined by the user and ranging from 1 to 625. How can I do this without replacement becoming an issue given there are two coordinate values, x and y?

Code for generating a 25 x 25 array:

 import numpy as np
 Array = np.zeros((25,25))

Expected result on a 4 x 4 array:

Before

0,0,0,0
0,0,0,0
0,0,0,0
0,0,0,0

After (replace 1)

0,0,0,0
0,0,1,0
0,0,0,0
0,0,0,0

After (replace 16)

1,1,1,1
1,1,1,1
1,1,1,1
1,1,1,1

CodePudding user response：

Notes:

• One may generate a unique sequence of random integers from 0 to 624, and mutate the corresponding element of the flattened array according to the generated sequence. Working with the flattened array, in this specific case, enables the development of a (relatively) concise code.

• One can use np.random.choice together with the argument replace=False to generate a non-repetitive sequence of random numbers. You can see more information on

  CodePudding user response：

  Try:

  import random
  
  import numpy as np
  
  howmany = 5
  
  array = np.zeros((25, 25), dtype=int)
  x = set(range(25))
  y = set(range(25))
  
  for num in range(howmany):
      xi = random.choice(list(x))
      yi = random.choice(list(y))
  
      array[xi, yi] = 1
  
      x.discard(xi)
      y.discard(yi)
  
  print(array)