All you can expect from Python is to generate for you a random number in \([1,0)\). In an unbiased flip all the shots of random below 0.5 would be True and above would be False, or vice versa. Now you can adjust the bias of the flip with the parameter p.

def flip(p):
    """Return True with probability p"""
    from random import random
    return random() < p

You can test your flip by:

len([k for k in [flip(0.8) for _ in range(int(1e5))] if k])/1e5

0.80005
Download .ipynb Download .py