这段python代码是我想要的吗？

Question

I have this code

import random
b = 20
x = random.randrange(0,b)
y = random.randrange(0,b)
ab = 0
xc = 0

while ab != 10:
    if x != y:
        x = random.randrange(0,b)
        y = random.randrange(0,b)
        xc = xc + 1
    elif x == y:
        print ab
        print 'number of tries out of', b, ' : ', xc
        ab = ab + 1
        xc = 0
        y = 999999

it's essentially a stats program. I want to see out of 10 tries how many times a random number needs to be generated in order to match. So far the numbers i have been getting after running this program i get these values, i ran the program 5 times so that would be a total of 50 tries.

9
26
6
1
5
109
5
42
12
63

All of those low numbers has led me to wondering if my program is just extremely lucky or if something is wrong my code. Thanks!

Note: is there a way i can get XC to add up after each while loop and when the while loop is true for it to show the total?

Answer 1

I'd say that you were somewhat unlucky.

The average there is 27.8. But if you pick two numbers between 0 and 20, you'd expect them to match about 1/20 of the time, and hence you expect to wait about 20 times before you get a match.

Answer 2

Check it with a larger number of iterations:

#!/usr/bin/env python
import random

max_num = 20
x = random.randrange(0, max_num)
y = random.randrange(0, max_num)
tries = 0

i = 0
iterations = 1000
total_tries = 0
while i < iterations:
    if x != y:
        x = random.randrange(0, max_num)
        y = random.randrange(0, max_num)
        tries += 1
    else:
        print(i)
        print('number of tries out of %s : %s ' % (max_num, tries))
        total_tries += tries
        i += 1
        tries = 0
        y = -1

print("Average tries: %s" % (total_tries / iterations))

If I run this, I get ~20 every time.

Note: This could be simplified to this:

#!/usr/bin/env python
import random

max_num = 20
iterations = 1000
total_tries = 0

for i in range(iterations):
    print(i)

    tries = 0
    x = 0
    y = -1
    while x != y:
        x = random.randrange(0, max_num)
        y = random.randrange(0, max_num)
        tries += 1

    print('number of tries out of %s : %s ' % (max_num, tries))
    total_tries += tries
    i += 1
    tries = 0

print("Average tries: %s" % (total_tries / iterations))

Answer 3

Your logic is very strange.

• You don't want to use artificial sentinel values to break out of a loop. Just explicitly break out of it.

• You fundamentally have two loops: one to iterate over trials, and one to see how many attempts are needed in a given trial. Don't hide that structure.

• Don't cover all the cases with if / else if . That's what else is for.

• Within a loop, generate numbers before you test them, not after. That way it's clearer what you're doing, and you don't need an extra generation step before the loop. Similarly, re-initialize values within a loop at the top, rather than the bottom.

• Use clear variable names. When there isn't a meaningful variable name, avoid creating a variable. You don't actually need to set two variables to random.randrange(0, b) to compare the results. OTOH, if you want to simplify the logic and avoid writing a seemingly weird comparison of random.randrange(0, b) to itself, then you can note (and should be able to prove, if you're interested enough in this kind of stuff to write a program) that you can arbitrarily select a target value and get the same results. Also, use variables to name your arbitrarily selected numeric constants.

• You can use += in Python to update a variable.

• Use print formatting.

---

import random
range_size = 20
total = 0
iterations = 1000 # or however many

for trial in xrange(iterations):
    attempts = 0
    while random.randrange(0, range_size) != 0: attempts += 1
    print "Trial #{0}: Took {1} tries to match one of {2} numbers.".format(
        trial, attempts, range_size
    )
    total += attempts

print "Average trials: {0}".format(float(total) / iterations)

If you don't need the debugging information, we can make things even cleaner by using built-in functions to do the summation and looping logic for us:

from random import randrange
from itertools import *
range_size = 20
total = 0
iterations = 1000 # or however many

print "Average trials: {0}".format(sum(
    sum(takewhile(lambda x: randrange(0, range_size) != 0, repeat(1)))
    # I tested that way, but this is probably more logical
    # even if it's more verbose:
    # len(list(takewhile(
    #     lambda x: x != 0,
    #     imap(randrange, repeat(0), repeat(range_size))
    # )))
    # 'lambda x: x != 0' can also be spelled 'bool' in this context,
    # but explicit is better than implicit...
    for i in xrange(iterations)
) / float(iterations))