Skip to content

Stabilize concentrated Watson density evaluation #5259

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
FlorianPfaff wants to merge 2 commits into
base: main
Choose a base branch
from
Open

Stabilize concentrated Watson density evaluation #5259

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
65266c5
Stabilize Watson density evaluation
FlorianPfaff Aug 7, 2026
45cf957
Test concentrated Watson density stability
FlorianPfaff Aug 7, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion src/pyrecest/distributions/hypersphere_subset/watson_distribution.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -155,7 +155,7 @@ def pdf(self, xs):
raise ValueError(
f"xs must have trailing dimension {self.input_dim}, got {xs.shape}."
)
p = self.norm_const * exp(self.kappa * (xs @ self.mu) ** 2)
p = exp(self.ln_norm_const + self.kappa * (xs @ self.mu) ** 2)
return p

def ln_pdf(self, xs):
Expand Down
6 changes: 6 additions & 0 deletions tests/distributions/test_watson_distribution.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -81,6 +81,12 @@ def test_pdf(self):
pdf_values = w.pdf(self.xs)
npt.assert_array_almost_equal(pdf_values, expected_pdf_values, decimal=5)

def test_pdf_stays_finite_for_large_positive_kappa(self):
mu = array([1.0, 0.0, 0.0])
dist = WatsonDistribution(mu, 1000.0)

npt.assert_allclose(dist.pdf(mu), 159.07528584318962, rtol=1e-6)

def test_pdf_accepts_list_inputs(self):
mu = array([1.0, 2.0, 3.0])
mu = mu / linalg.norm(mu)
Expand Down
Loading