BUG: stats: Fix zipf.pmf and zipfian.pmf for int32 k
#20702
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
github-actions
bot
added
scipy.stats
defect
dschmitz89
changed the title
zipf.pmf for integer kzipf.pmf for int32 k
|
scipy/scipy/stats/_discrete_distns.py Line 1415 in 2f16622 I am not familiar with these distributions though and could not cook up a failing example in a quick session, so let me know if this would be important to fix as well. |
tylerjereddy
added
the
backport-candidate
| dist = zipf(9) | ||
| pmf = dist.pmf(k) | ||
| pmf_k_int32 = dist.pmf(k_int32) | ||
| assert_equal(pmf, pmf_k_int32) |
There was a problem hiding this comment.
Hey Daniel, I just messed around for a bit and found this case that failed before/passed after a similar patch per your comment on zipfian:
diff --git a/scipy/stats/_discrete_distns.py b/scipy/stats/_discrete_distns.py
index cda7c4e11..2ee610827 100644
--- a/scipy/stats/_discrete_distns.py
+++ b/scipy/stats/_discrete_distns.py
@@ -1412,7 +1412,8 @@ class zipfian_gen(rv_discrete):
return 1, n
def _pmf(self, k, a, n):
- return 1.0 / _gen_harmonic(n, a) / k**a
+ k = k.astype(np.float64)
+ return 1.0 / _gen_harmonic(n, a) * k**-a
def _cdf(self, k, a, n):
return _gen_harmonic(k, a) / _gen_harmonic(n, a)
diff --git a/scipy/stats/tests/test_discrete_distns.py b/scipy/stats/tests/test_discrete_distns.py
index c5993ebde..bbe9085ce 100644
--- a/scipy/stats/tests/test_discrete_distns.py
+++ b/scipy/stats/tests/test_discrete_distns.py
@@ -627,3 +627,12 @@ class TestBetaNBinom:
# return float(fourth_moment/var**2 - 3.)
assert_allclose(betanbinom.stats(n, a, b, moments="k"),
ref, rtol=3e-15)
+
+
+def test_zipfian_gh_20692():
+ k = np.arange(0, 1000)
+ k_int32 = k.astype(np.int32)
+ dist = zipfian(111, 22)
+ pmf = dist.pmf(k)
+ pmf_k_int32 = dist.pmf(k_int32)
+ assert_equal(pmf, pmf_k_int32)You shouldn't trust my judgement on stats stuff, but it does look quite similar in nature as you note. The obtained pmf floats are ridiculously tiny, maybe 111 is a bit much, but anyway I just pushed it to error out for integers. It does run lightning fast at least.
There was a problem hiding this comment.
Added in the last commit.
Co-authored-by: Tyler Reddy <tyler.je.reddy@gmail.com>
dschmitz89
changed the title
zipf.pmf for int32 kzipf.pmf and zipfian.pmf for int32 k
lucascolley
changed the title
zipf.pmf and zipfian.pmf for int32 kzipf.pmf and zipfian.pmf for int32 k
|
I went ahead and merged this since it is a quite simple bug fix. |
tylerjereddy
pushed a commit
to tylerjereddy/scipy
that referenced
this pull request
May 22, 2024
…) [wheel build] * BUG: fix zipf.pmf for integer k * ENH: increase range for zipfian.pmf for integer k --------- Co-authored-by: Tyler Reddy <tyler.je.reddy@gmail.com> [wheel build]
10 tasks
tylerjereddy
removed
the
backport-candidate
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Reference issue
Closes gh-20692
In the long run, these issues should hopefully be taken care of by the infrastructure (#15928 ).