unknown_bugs.py

import matplotlib.pyplot as plt
import random
import numpy as np
from scipy.optimize import curve_fit

INITIAL_BUGS = 50
LINES_OF_CODE = 1000
TESTERS = 5
RUNS = 10


def random_bugs_found(remaining_bugs):
    probability = remaining_bugs / LINES_OF_CODE
    if random.random() < probability:
        return 1
    else:
        return 0


def run_simulation():
    remaining_bugs = INITIAL_BUGS
    remaining_bugs_each_t = [remaining_bugs]
    t = 0
    t_array = [0]

    while remaining_bugs > 0:
        t += 1

        for _ in range(0, TESTERS):
            if remaining_bugs > 0:
                remaining_bugs -= random_bugs_found(remaining_bugs)

        remaining_bugs_each_t.append(remaining_bugs)
        t_array.append(t)

    return t_array, remaining_bugs_each_t


def b_of_t(t, a):
    return INITIAL_BUGS * np.power(a, np.divide(np.multiply(-TESTERS, t), LINES_OF_CODE))


a_values = []
fig, ax = plt.subplots()
for _ in range(0, RUNS):
    x, y = run_simulation()
    ax.plot(x, y)

    popt, pcov = curve_fit(b_of_t, x, y)
    a_values.append(popt[0])
    #plt.plot(x, b_of_t(x, *popt), 'r-', label='fit: a=%5.3f' % tuple(popt))

# Print average fitted value of parameter "a" in b_of_t()
print(sum(a_values) / len(a_values))

plt.title('Remaining unknown bugs')
plt.xlabel('time')
plt.ylabel('bugs')
plt.show()