Mercurial > public > think_complexity
annotate ch5ex3.py @ 39:49db586c727a
Chapter 6, exercise 1: simple program to play with CA and CADrawer.
| author | Brian Neal <bgneal@gmail.com> |
|---|---|
| date | Fri, 11 Jan 2013 21:18:45 -0600 |
| parents | 65b70b4e12ad |
| children |
| rev | line source |
|---|---|
| bgneal@29 | 1 """Chapter 5.3 exercise 3 in Allen Downey's Think Complexity book. |
| bgneal@29 | 2 |
| bgneal@29 | 3 "Write a function called plot_ccdf that takes a list of values and the |
| bgneal@29 | 4 corresponding list of probabilities and plots the CCDF on a log-y scale. |
| bgneal@29 | 5 |
| bgneal@29 | 6 To test your function, use expovariate from the random module to generate 100 |
| bgneal@29 | 7 values from an exponential distribution. Plot the CCDF on a log-y scale and see |
| bgneal@29 | 8 if it falls on a straight line." |
| bgneal@29 | 9 |
| bgneal@29 | 10 """ |
| bgneal@29 | 11 import random |
| bgneal@29 | 12 import sys |
| bgneal@29 | 13 |
| bgneal@29 | 14 import matplotlib.pyplot as pyplot |
| bgneal@29 | 15 |
| bgneal@29 | 16 |
| bgneal@29 | 17 def plot_ccdf(x_vals, y_vals, title=''): |
| bgneal@29 | 18 """Given a set of x-values and y-values from a continuous distribution, plot |
| bgneal@29 | 19 the complementary distribution (CCDF) on a log-y scale. |
| bgneal@29 | 20 |
| bgneal@29 | 21 """ |
| bgneal@29 | 22 if len(x_vals) != len(y_vals): |
| bgneal@29 | 23 raise ValueError |
| bgneal@29 | 24 |
| bgneal@29 | 25 ys = [1.0 - y for y in y_vals] |
| bgneal@29 | 26 |
| bgneal@29 | 27 pyplot.clf() |
| bgneal@29 | 28 pyplot.xscale('linear') |
| bgneal@29 | 29 pyplot.yscale('log') |
| bgneal@29 | 30 pyplot.title(title) |
| bgneal@29 | 31 pyplot.xlabel('x') |
| bgneal@29 | 32 pyplot.ylabel('1-y') |
| bgneal@29 | 33 pyplot.plot(x_vals, ys, label='1-y', color='green', linewidth=3) |
| bgneal@29 | 34 pyplot.legend(loc='upper right') |
| bgneal@29 | 35 pyplot.show() |
| bgneal@29 | 36 |
| bgneal@29 | 37 |
| bgneal@29 | 38 def main(script, lambd): |
| bgneal@29 | 39 |
| bgneal@29 | 40 lambd = float(lambd) |
| bgneal@29 | 41 |
| bgneal@29 | 42 x_vals = range(100) |
| bgneal@29 | 43 y_vals = [random.expovariate(lambd) for x in x_vals] |
| bgneal@29 | 44 y_vals.sort() |
| bgneal@29 | 45 |
| bgneal@29 | 46 plot_ccdf(x_vals, y_vals, 'expovariate w/lambda = {}'.format(lambd)) |
| bgneal@29 | 47 |
| bgneal@29 | 48 |
| bgneal@29 | 49 if __name__ == '__main__': |
| bgneal@29 | 50 main(*sys.argv) |