Mirror
/
cpython
mirror of https://github.com/python/cpython
4
1
Fork 0
Code Issues Releases Wiki Activity

Fixed a bug in the gauss() function. The bug was reported by Mike 

Miller, who complained that its kurtosis was bad, and then fixed by
Lambert Meertens (author of the original algorithm) who discovered
that the mathematical analysis leading to his solution was wrong, and
provided a corrected version.  Mike then tested the fix and reported
that the kurtosis was now good.
This commit is contained in:
Guido van Rossum 1998-02-19 21:17:42 +00:00
parent 9824509d3e
commit 72c2e1b56e
1 changed files with 6 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
Lib/random.py
View File

@ -182,12 +182,13 @@ def gauss(mu, sigma):
# When x and y are two variables from [0, 1), uniformly # When x and y are two variables from [0, 1), uniformly
# distributed, then # distributed, then
# #
# cos(2*pi*x)*log(1-y) # cos(2*pi*x)*sqrt(-2*log(1-y))
# sin(2*pi*x)*log(1-y) # sin(2*pi*x)*sqrt(-2*log(1-y))
# #
# are two *independent* variables with normal distribution # are two *independent* variables with normal distribution
# (mu = 0, sigma = 1). # (mu = 0, sigma = 1).
# (Lambert Meertens) # (Lambert Meertens)
# (corrected version; bug discovered by Mike Miller, fixed by LM)
global gauss_next global gauss_next
@ -196,9 +197,9 @@ def gauss(mu, sigma):
gauss_next = None gauss_next = None
else: else:
x2pi = random() * TWOPI x2pi = random() * TWOPI
log1_y = log(1.0 - random()) g2rad = sqrt(-2.0 * log(1.0 - random()))
z = cos(x2pi) * log1_y z = cos(x2pi) * g2rad
gauss_next = sin(x2pi) * log1_y gauss_next = sin(x2pi) * g2rad
return mu + z*sigma return mu + z*sigma