diff options
| author | Raymond Hettinger <rhettinger@users.noreply.github.com> | 2023-08-27 13:59:40 (GMT) |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2023-08-27 13:59:40 (GMT) |
| commit | 042aa88bcc6541cb8b312f1119452f7a58a5b4df (patch) | |
| tree | a9cc25f87cf66e82fbd0277db1ab1644a0b14d27 | |
| parent | 09343dba44cdb5c279ec51df34552ef451434958 (diff) | |
| download | cpython-042aa88bcc6541cb8b312f1119452f7a58a5b4df.zip cpython-042aa88bcc6541cb8b312f1119452f7a58a5b4df.tar.gz cpython-042aa88bcc6541cb8b312f1119452f7a58a5b4df.tar.bz2 | |
gh-108322: Optimize statistics.NormalDist.samples() (gh-108324)
| -rw-r--r-- | Doc/library/statistics.rst | 5 | ||||
| -rw-r--r-- | Lib/statistics.py | 12 | ||||
| -rw-r--r-- | Misc/NEWS.d/next/Library/2023-08-22-12-05-47.gh-issue-108322.kf3NJX.rst | 2 |
3 files changed, 14 insertions, 5 deletions
diff --git a/Doc/library/statistics.rst b/Doc/library/statistics.rst index 483ebea..368b2a1 100644 --- a/Doc/library/statistics.rst +++ b/Doc/library/statistics.rst @@ -828,6 +828,11 @@ of applications in statistics. number generator. This is useful for creating reproducible results, 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) Using a `probability density function (pdf) diff --git a/Lib/statistics.py b/Lib/statistics.py index a8036e9..96c8034 100644 --- a/Lib/statistics.py +++ b/Lib/statistics.py @@ -1135,7 +1135,7 @@ def linear_regression(x, y, /, *, proportional=False): >>> noise = NormalDist().samples(5, seed=42) >>> y = [3 * x[i] + 2 + noise[i] for i in range(5)] >>> 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 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)] >>> 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) @@ -1279,9 +1279,11 @@ class NormalDist: def samples(self, n, *, seed=None): "Generate *n* samples for a given mean and standard deviation." - gauss = random.gauss if seed is None else random.Random(seed).gauss - mu, sigma = self._mu, self._sigma - return [gauss(mu, sigma) for _ in repeat(None, n)] + rnd = random.random if seed is None else random.Random(seed).random + inv_cdf = _normal_dist_inv_cdf + mu = self._mu + sigma = self._sigma + return [inv_cdf(rnd(), mu, sigma) for _ in repeat(None, n)] def pdf(self, x): "Probability density function. P(x <= X < x+dx) / dx" diff --git a/Misc/NEWS.d/next/Library/2023-08-22-12-05-47.gh-issue-108322.kf3NJX.rst b/Misc/NEWS.d/next/Library/2023-08-22-12-05-47.gh-issue-108322.kf3NJX.rst new file mode 100644 index 0000000..5416c01 --- /dev/null +++ b/Misc/NEWS.d/next/Library/2023-08-22-12-05-47.gh-issue-108322.kf3NJX.rst @@ -0,0 +1,2 @@ +Speed-up NormalDist.samples() by using the inverse CDF method instead of +calling random.gauss(). |