Randomizing Sentences In Python

Question

I not long ago finished my project which comments on a video based on a keyword on YouTube, it will pick a random comment using the random library. The program has been acting strange when "randomizing" the comments to add to the YouTube videos. I'm starting to think I may need to improve the randomness, the fact being that it has chosen the 4th comment (out of 9) 7 times, and the 9th one 3 times. Both of these results occur after each other, in other words, it is sending the same message in a row when it's meant to pick randomly and not repeat. Is there any way I can increase how random this is? If you know, please do tell me, I will appreciate it a bunch!

PS: This is more of stopping it from repeating the same sentence on YouTube.

code:

if __name__ == "__main__":
    from googleapiclient.errors import HttpError
    import random
    import time
    import sys
    
    # Comments are getting loaded
    comments = load_comments('Comments.txt')
    # Getting the number of comments you want to add
    number_of_comments = int(input('Enter the number of comments: '))
    count, cycle, videoid_store = 0, 1, []
    # Getting the keyword
    keyword = input('Enter the Keyword: ')
    
    
    # This loop keeps running until all comments have been added
    youtube = authentication()
    while count < number_of_comments:
        
        print("Searching for videos ..  (Cycle:%d)" %cycle)
        time.sleep(10)
        
        random.shuffle(comments)
        

Answer 1

If you want to keep the randomness but avoid repeats or frequent re-occurrences, you have to remember the previous outcomes, to manipulate the odds in favour of the ones that have occurred less frequently so far.

Some examples:

1. Shuffle the outcomes. Go through all, by the order obtained. Repeat.
2. Keep a count of the last few outcomes. If we get an outcome that has been drawn X times already, draw again up to X times.

Code, for the above examples:

import random
import collections

class ShuffleAndExhaust(): # cycle in random order each time
    
    def __init__(self,N): # N: number of possible outcomes
        self.N     = N
        self.queue = list()

    def draw(self):
        if not self.queue: # exhausted (or 1st ever draw)
            self.queue.extend(range(self.N))
            random.shuffle(self.queue)
        return self.queue.pop()

class RememberAndRetry(): # re-draw, if reoccurring
    
    def __init__(self,N): # N: number of possible outcomes
        self.N       = N
        self.history = collections.deque(maxlen=3*self.N) # maxlen = rule-of-thumb constant * N

    def draw(self):
        triesleft = collections.Counter(self.history)
        while True:
            n = random.randint(0,self.N-1)
            if triesleft[n] > 0: # drew this number N times already
                triesleft[n] -= 1 # 1 less try left drawing this same number until we accept it
            else: break
        self.history.append(n)
        return n

Testing 100 draws of 10 possible outcomes:

>>> random.seed(0) # for reproducibility in testing only
>>> r = ShuffleAndExhaust(10)
>>> print(*(r.draw() for i in range(100)))
6 9 0 2 4 3 5 1 8 7 5 3 2 7 1 0 6 8 4 9 4 1 8 6 5 2 3 9 0 7 5 6 9 4 7 1 3 8 2
0 0 8 9 7 5 3 6 2 1 4 5 3 9 7 0 1 4 6 2 8 8 7 1 0 2 4 3 6 9 5 8 4 1 9 2 6 7 5
3 0 7 1 6 2 4 8 9 0 3 5 2 0 4 3 8 5 1 7 9 6
>>> r = RememberAndRetry(10)
>>> print(*(r.draw() for i in range(100)))
1 8 6 4 3 9 7 5 1 3 0 2 0 9 1 0 4 6 2 7 8 5 9 3 6 7 0 1 4 5 2 5 4 8 7 6 3 1 7
9 8 9 4 0 7 9 1 0 8 6 1 5 3 2 5 2 3 6 6 2 7 4 0 4 1 8 6 7 0 1 5 3 9 8 0 9 5 4
7 7 1 6 8 3 8 5 2 3 9 0 5 0 2 6 2 1 4 4 7 4

Use N = the number of comments and each draw as the index of the comment to pick.

Answer 2

Try random.choice()

x=random.choice(['option1','option2'])