Python: rounding error distorts uniform distribution

Question

I need to sample 10 uniformly distributed random numbers between 0 and 1. So I thought the following code in python would do this:

positions = []
for dummy_i in range(1000000):
    positions.append(round(random.random(),1))

However when putting the result into a histogram the result looks as follows:

So the rounding seems to destroy the uniform distribution generated by random.random(). I wonder what causes this and how to prevent this from happening. Thanks for your help!

Answer 1

It seems you have a problem later in the code... (e.g. when collecting the statistics). Check this smaller snippet:

import random, collections
data = collections.defaultdict(int)
for x in range(1000000):
    data[round(random.random(),1)] += 1
print(data)

You will see that 0 and 1 of course have about half of the samples of the other values that are all pretty much uniform.

For example I got:

defaultdict(<class 'int'>,
            {0.4: 100083,
             0.9: 99857,
             0.3: 99892,
             0.8: 99586,
             0.5: 100108,
             1.0: 49874,     # Correctly about half the others
             0.7: 100236,
             0.2: 99847,
             0.1: 100251,
             0.6: 100058,
             0.0: 50208})    # Correctly about half the others