将概率添加到列表列表中的项目

Question

I want to add probabilities to each item in a list, where that list is in another list.

Some psuedo-code:

myList = [ [a, b, c, d], [e, f, g, h], [i, j, k, l], [m, n, o], [p, q, r], [s, t, u] ]

probabilities = [ [0.6, 0.3, 0.075, 0.025], [0.6, 0.3, 0.075, 0.025], [0.6, 0.3, 0.075, 0.025], [0.55, 0.35, 0.1], [0.55, 0.35, 0.1], [0.55, 0.35, 0.1] ]

Is there any way to do achieve this?

Further:

My need for this is to create another list that would look similar to the below...

newList = [ [b, e, k, o, p, s], [a, f, i, m, r, t], ... etc. ] 

where each element was chosen randomly given the probabilities, and no two list in newList are the same. Which I am not sure is achievable.

My code so far:

layers = [list(Path(directory).glob("*.png")) for directory in ("dir1/", "dir2/", "dir3/", "dir4/", "dir5/", "dir6/")]

list_of_prob = [[0.6, 0.3, 0.075, 0.025], [0.6, 0.3, 0.075, 0.025], [0.6, 0.3, 0.075, 0.025], [0.6, 0.3, 0.1], [0.6, 0.3, 0.1], [0.6, 0.3, 0.1]]

rwp = [choices(layers, list_of_prob, k=????)]

rand_combinations = [([choice(k) for k in layers]) for i in choice_indices]

I am not entirely sure what k would be in choices(), ex. number of lists or number of total elements in the lists. Layers is a list of image paths, .pngs, which is identical to the format of "myList" provided above in pseudo code (4 images in dir1, 4 images in dir2, 4 images in dir3, 3 in dir4, 3 in dir5, 3 in dir6).

I already have code to iterate through a list and create random images, but I want some of the images to only be generated x% of the time. Hence my original question. Sorry if I just complicated things, I tried to simplify it.

Answer 1

I converted myList to strings just to make things easy.

This will create combinations and append them to newList , disregarding any combinations that already exist in newList

While loop ends when length of newList is equal to the length of myList

import random

myList = [['a', 'b', 'c', 'd'], 
          ['e', 'f', 'g', 'h'], 
          ['i', 'j', 'k', 'l'], 
          ['m', 'n', 'o'], 
          ['p', 'q', 'r'], 
          ['s', 't', 'u']]

probabilities = [[0.6, 0.3, 0.075, 0.025], 
                 [0.6, 0.3, 0.075, 0.025], 
                 [0.6, 0.3, 0.075, 0.025], 
                 [0.55, 0.35, 0.1], 
                 [0.55, 0.35, 0.1], 
                 [0.55, 0.35, 0.1]]
newList = []

def random_list():
    combo = []
    for d, p in zip(myList, probabilities):
        choice = random.choices(d, p)
        combo.append(''.join(choice))
    return combo

while len(newList) < len(myList):
    a = random_list()
    if a not in newList:
        newList.append(a)

Results of newList:

[['b', 'f', 'k', 'm', 'q', 's'],
 ['a', 'e', 'j', 'm', 'q', 't'],
 ['b', 'f', 'k', 'm', 'q', 'u'],
 ['a', 'f', 'i', 'm', 'p', 's'],
 ['a', 'e', 'i', 'n', 'p', 't'],
 ['b', 'f', 'k', 'm', 'r', 'u']]

Answer 2

So if I understood correctly, you want to take one element of each sublist of myList based on the probabilities to get an element, and do this multiple times to get a list of lists.

With Python 3, you can do:

import random
# the new set since two elements cannot be the same
result_set = set()
# the number of times you want to do it, you can change it.
for i in range(5):
    # A single result as a str, hashable.
    result = ""
    for chars, char_probabilities in zip(myList, probabilities):
    # weighted random choice
    char = random.choices(chars, char_probabilities)
    result += char[0]
    result_set.add(result)

result_list = [list(elt) for elt in result_set]