How to create a random list that satisfy a condition (in one try)?-CodePudding

I have written the following code to generate a random list. I want the list to have elements between 0 and 500, but the summation of all elements does not exceed 1300. I dont know how to continue my code to do that. I have written other codes; for example, to create a list of random vectors and then pick among those that satisfy the condition. But here I want to create such a list in one try.

nv = 5
bounds = [(0, 500), (0, 500), (0, 500), (0, 500), (0, 500)]
var =[]
for j in range(nv):
    var.append(random.uniform(bounds[j][0], bounds[j][1]))
    summ = sum(var)
    if summ > 1300:
        ????

CodePudding user response：

Don't append until after you've validated the value.

Use while len() < maxLen so that you can handle repeat attempts.

You don't really need nv since len(bounds) dictates the final value of len(var).

len(var) is also the next index of the var list that is unused so you can use that to keep track of where you are in bounds.

A running sum is more efficient than using sum() on every check. (Though on small lists, it's not going to make a noticeable difference.)

The * in the .uniform() call splits a list into individual arguments. (Asterisks in Python: what they are and how to use them seems like a good tutorial on the subject.)

import random

bounds = [(0, 500), (0, 500), (0, 500), (0, 500), (0, 500)]
var = []
runningSum = 0
while len(var) < len(bounds):
    sample = random.uniform(*bounds[len(var)])
    if runningSum   sample < 1300:
        runningSum  = sample
        var.append(sample)

print(repr(var))

CodePudding user response：

Per the comments, you're able to use a third-party library, so you can do this with Numpy. This question is where I got some of this code, because I'm not very familiar with Numpy.

import numpy as np
import random

limits = [500, 500, 500, 500, 500]
total = 1300
gen = np.random.Generator(np.random.PCG64(random.randrange(1, 100)))
random_list = gen.multivariate_hypergeometric(limits, total)
print(list(random_list))

CodePudding user response：

Without the aid of numpy you could do this:

from random import uniform

def func1():
    LIMIT = 1_300
    bounds = [(0, 500), (0, 500), (0, 500), (0, 500), (0, 500)]

    while sum(result := [uniform(lo, hi) for lo, hi in bounds]) > LIMIT:
        pass

    return result