mirrors
/
cpython
mirror of https://github.com/python/cpython.git
2
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

gh-108322: Optimize statistics.NormalDist.samples() (gh-108324)

This commit is contained in:
Raymond Hettinger 2023-08-27 08:59:40 -05:00 committed by GitHub
parent 09343dba44
commit 042aa88bcc
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
3 changed files with 14 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
Doc/library/statistics.rst
View File

@ -828,6 +828,11 @@ of applications in statistics.
number generator. This is useful for creating reproducible results, number generator. This is useful for creating reproducible results,
even in a multi-threading context. even in a multi-threading context.
.. versionchanged:: 3.13
Switched to a faster algorithm. To reproduce samples from previous
versions, use :func:`random.seed` and :func:`random.gauss`.
.. method:: NormalDist.pdf(x) .. method:: NormalDist.pdf(x)
Using a `probability density function (pdf) Using a `probability density function (pdf)

12
Lib/statistics.py
View File

@ -1135,7 +1135,7 @@ def linear_regression(x, y, /, *, proportional=False):
>>> noise = NormalDist().samples(5, seed=42) >>> noise = NormalDist().samples(5, seed=42)
>>> y = [3 * x[i] + 2 + noise[i] for i in range(5)] >>> y = [3 * x[i] + 2 + noise[i] for i in range(5)]
>>> linear_regression(x, y) #doctest: +ELLIPSIS >>> linear_regression(x, y) #doctest: +ELLIPSIS
LinearRegression(slope=3.09078914170..., intercept=1.75684970486...) LinearRegression(slope=3.17495..., intercept=1.00925...)
If *proportional* is true, the independent variable *x* and the If *proportional* is true, the independent variable *x* and the
dependent variable *y* are assumed to be directly proportional. dependent variable *y* are assumed to be directly proportional.
@ -1148,7 +1148,7 @@ def linear_regression(x, y, /, *, proportional=False):
>>> y = [3 * x[i] + noise[i] for i in range(5)] >>> y = [3 * x[i] + noise[i] for i in range(5)]
>>> linear_regression(x, y, proportional=True) #doctest: +ELLIPSIS >>> linear_regression(x, y, proportional=True) #doctest: +ELLIPSIS
LinearRegression(slope=3.02447542484..., intercept=0.0) LinearRegression(slope=2.90475..., intercept=0.0)
""" """
n = len(x) n = len(x)
@ -1279,9 +1279,11 @@ def from_samples(cls, data):
def samples(self, n, *, seed=None): def samples(self, n, *, seed=None):
"Generate *n* samples for a given mean and standard deviation." "Generate *n* samples for a given mean and standard deviation."
gauss = random.gauss if seed is None else random.Random(seed).gauss rnd = random.random if seed is None else random.Random(seed).random
mu, sigma = self._mu, self._sigma inv_cdf = _normal_dist_inv_cdf
return [gauss(mu, sigma) for _ in repeat(None, n)] mu = self._mu
sigma = self._sigma
return [inv_cdf(rnd(), mu, sigma) for _ in repeat(None, n)]
def pdf(self, x): def pdf(self, x):
"Probability density function. P(x <= X < x+dx) / dx" "Probability density function. P(x <= X < x+dx) / dx"

2
Misc/NEWS.d/next/Library/2023-08-22-12-05-47.gh-issue-108322.kf3NJX.rst Normal file
View File

@ -0,0 +1,2 @@
Speed-up NormalDist.samples() by using the inverse CDF method instead of
calling random.gauss().