Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- # Made by Wout12345
- import numpy as np
- import matplotlib.pyplot as plt
- days = 365
- n = 61
- min_pairs = 5
- experiments = int(1e6)
- passes = 0
- probabilities = np.empty(dtype=np.float32, shape=(experiments,))
- # Some helper variables for blocking
- block_size = int(1e5)
- experiments = int(np.ceil(experiments/block_size))*block_size
- # Initialize constant indices in birthdays array
- random_birthdays_amount = block_size*n
- birthdays = np.empty(dtype=np.uint32, shape=(2, random_birthdays_amount))
- for i in range(block_size):
- for j in range(n):
- birthdays[0, i*n + j] = i
- birthday_counts = np.empty(dtype=np.uint32, shape=(block_size, days))
- success = np.empty(dtype=np.uint8, shape=(block_size,))
- for i in range(0, experiments, block_size):
- # Assign random birthdays
- birthdays[1, :] = np.random.randint(0, days - 1, (random_birthdays_amount,))
- # Count number of birthdays per day
- birthday_counts.fill(0)
- np.add.at(birthday_counts, (birthdays[0, :], birthdays[1, :]), 1)
- # Check if there are at least 4 pairs of people sharing their birthday without 3 or more people ever sharing a birthday
- success.fill(1)
- #success[np.amax(birthday_counts, axis=1) > 2] = 0
- success[np.sum((birthday_counts >= 2).astype(int), axis=1) < min_pairs] = 0
- # success has now been determined
- probabilities[i:i + block_size] = (round(probabilities[i - 1]*i) if i > 0 else 0) + np.cumsum(success)
- probabilities[i:i + block_size] /= np.linspace(i + 1, i + block_size, block_size)
- print("Empirical probability after %s experiments: %.8f"%(experiments, probabilities[-1]))
- # The part below is only relevant if you have some theoretical value you want to verify
- theoretical_probability = 0.1720784820786103795516077320188488593260442214894743886987922653 # For exactly 4 pairs
- #theoretical_probability = 0.76 # For 4+ pairs normally
- print("Relative error from theoretical probability: %s"%((probabilities[-1] - theoretical_probability)/theoretical_probability))
- """
- plt.plot(range(1, experiments+1), probabilities)
- plt.plot([1, experiments], [theoretical_probability,]*2)
- plt.xlabel("Experiments")
- plt.ylabel("Probability")
- plt.xlim(1, experiments)
- plt.ylim(0, 1)
- plt.show()
- plt.plot(range(1, experiments+1), np.abs(probabilities - theoretical_probability))
- plt.xscale("log", base=2)
- plt.yscale("log", base=2)
- plt.xlabel("Experiments")
- plt.ylabel("Error of empirical probability")
- plt.xlim(1, experiments)
- plt.grid()
- plt.show()
- """
Advertisement
Add Comment
Please, Sign In to add comment