How to generate a random number of a numpy array?

Question

There is a numpy array like following numpy array. I want to generate a random integer number for specific string of the numpy array.. How can I do this task?

# input
X = np.array([['a', 'p', 'b'],
              ['a', 'p', 'd'],
              ['c', 'p', 'd'],
              ['c', 'p', 'e'],
              ['c', 'p', 'f']])

# output
X = np.array([[100, 5, 101],
              [100, 5, 103],
              [102, 5, 103],
              [102, 5, 104],
              [102, 5, 105]])

What do you mean by "random value"? It seems the values shown in the output are not random at all. — mkrieger1, Jun 06 '20 at 12:00
@mkrieger1, `a = 100` or `a=1`. Whatever you want to assign. — Md. Rezwanul Haque, Jun 06 '20 at 12:03
And what exactly is the problem that occurred when you tried to do this? Are you asking how to generate a random number? — mkrieger1, Jun 06 '20 at 12:04
You mean like this? https://stackoverflow.com/questions/3996904/generate-random-integers-between-0-and-9 — mkrieger1, Jun 06 '20 at 12:06
@mkrieger1, Nope. I have an input graph. Form there how can I generate like above graph.? — Md. Rezwanul Haque, Jun 06 '20 at 12:13
I don't see a graph in your question. I see two numpy arrays. Can you please be more specific about what you are trying to do and give the required context? — mkrieger1, Jun 06 '20 at 12:14
Let us [continue this discussion in chat](https://chat.stackoverflow.com/rooms/215417/discussion-between-md-rezwanul-haque-and-mkrieger1). — Md. Rezwanul Haque, Jun 06 '20 at 12:20

arshovon · Accepted Answer · 2020-06-06T12:30:20.197

As you want to assign a random integer values for each unique value of the array, you can map the unique values to a dictionary.

Update

We need to create check if the random number is used as a value of the mapper.

from random import randint
import numpy as np


# input
X = np.array([['a', 'p', 'b'],
              ['a', 'p', 'd'],
              ['c', 'p', 'd'],
              ['c', 'p', 'e'],
              ['c', 'p', 'f']])

mapper = {}

output = []
for ar in X:
    temp = []
    for i in range(len(ar)):
        if ar[i] not in mapper:
            random_number = None
            while True:
                random_number = randint(1,120)
                if random_number not in mapper.values():
                    break
            mapper[ar[i]] = random_number
        temp.append(mapper[ar[i]])
    output.append(temp)
output = np.array(output)
print(output)

Output:

[[  7  86 120]
 [  7  86  14]
 [ 51  86  14]
 [ 51  86 113]
 [ 51  86  54]]

Good point. Updated the answer. The type of the initial array was `unicode string` with length or equal to 1. That is why it was showing only the first character of the mapper value. — arshovon, Jun 06 '20 at 12:32

How to generate a random number of a numpy array?

1 Answers1