Skip to content

Added ToroidalFourierDistribution #282

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 1 commit into
base: main
Choose a base branch
from
Open

Added ToroidalFourierDistribution #282

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
81 changes: 81 additions & 0 deletions pyrecest/tests/distributions/test_toroidal_fourier_distribution.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,81 @@
import unittest
import numpy as np
from pyrecest.distributions.hypertorus.toroidal_fourier_distribution import ToroidalFourierDistribution
from pyrecest.distributions.hypertorus.toroidal_von_mises_sine_distribution import ToroidalVonMisesSineDistribution
from pyrecest.distributions.hypertorus.hypertoroidal_fourier_distribution import HypertoroidalFourierDistribution


class ToroidalFourierDistributionTest(unittest.TestCase):
def setUp(self):
self.tvm = ToroidalVonMisesSineDistribution(np.array([1, 4]), np.array([0.3, 0.7]), 0.5)
self.tfd = ToroidalFourierDistribution.from_distribution(self.tvm, [15, 15], 'sqrt')
self.hfd = HypertoroidalFourierDistribution.from_distribution(self.tvm, [15, 15], 'sqrt')
self.test_points = np.random.rand(100, 2) * 4 * np.pi - np.pi

def test_single_element_vector(self):
ToroidalFourierDistribution([1])

def test_from_function(self):
tvm = ToroidalVonMisesSineDistribution(np.array([1, 2]), np.array([0.3, 0.5]), 0.5)
tfd1_fun = ToroidalFourierDistribution.from_function(
lambda x, y: np.reshape(tvm.pdf(np.column_stack([x.ravel(), y.ravel()])), x.shape),
[15, 15],
desired_transformation='identity'
)
tfd2_fun = ToroidalFourierDistribution.from_function(
lambda x, y: np.reshape(tvm.pdf(np.column_stack([x.ravel(), y.ravel()])), x.shape),
[15, 15],
desired_transformation='sqrt'
)
for i in range(-2, 4):
tvm_moment = tvm.trigonometric_moment(i)
np.testing.assert_allclose(tfd1_fun.trigonometric_moment(i), tvm_moment, atol=1e-6)
np.testing.assert_allclose(tfd1_fun.trigonometric_moment_numerical(i), tvm_moment, atol=1e-6)
np.testing.assert_allclose(tfd2_fun.trigonometric_moment(i), tvm_moment, atol=1e-6)
np.testing.assert_allclose(tfd2_fun.trigonometric_moment_numerical(i), tvm_moment, atol=1e-6)

def test_from_function_and_moments(self):
tvm = ToroidalVonMisesSineDistribution([1, 2], [0.3, 0.5], 0.5)
tfd1 = ToroidalFourierDistribution.from_distribution(tvm, [15, 15], 'identity')
tfd2 = ToroidalFourierDistribution.from_distribution(tvm, [15, 15], 'sqrt')
for i in range(-2, 4):
tvm_moment = tvm.trigonometric_moment(i)
np.testing.assert_allclose(tfd1.trigonometric_moment(i), tvm_moment, atol=1e-6)
np.testing.assert_allclose(tfd1.trigonometric_moment_numerical(i), tvm_moment, atol=1e-6)
np.testing.assert_allclose(tfd2.trigonometric_moment(i), tvm_moment, atol=1e-6)
np.testing.assert_allclose(tfd2.trigonometric_moment_numerical(i), tvm_moment, atol=1e-6)

with np.testing.assert_warns(Warning):
np.testing.assert_allclose(tfd1.trigonometric_moment(16), [0, 0])
np.testing.assert_allclose(tfd2.trigonometric_moment(30), [0, 0])

def test_integrate(self):
tvm = ToroidalVonMisesSineDistribution([1, 2], [0.6, 1], 0.2)
tfd_id = ToroidalFourierDistribution.from_distribution(tvm, [45, 49], 'identity')
tfd_sqrt = ToroidalFourierDistribution.from_distribution(tvm, [45, 49], 'sqrt')
l = [0.3, 0.3]
r = [1.5, 1.5]
np.testing.assert_allclose(tfd_id.integrate(l, r), tvm.integrate(l, r), atol=1e-6)
np.testing.assert_allclose(tfd_id.integral_numerical(l, r), tvm.integrate(l, r), atol=1e-6)
np.testing.assert_allclose(tfd_sqrt.integrate(l, r), tvm.integrate(l, r), atol=1e-6)
np.testing.assert_allclose(tfd_sqrt.integral_numerical(l, r), tvm.integrate(l, r), atol=1e-6)

tfd_simple = ToroidalFourierDistribution(np.diag([0, 1 / (4 * np.pi ** 2), 0]), 'identity')
np.testing.assert_allclose(tfd_simple.integrate(), 1, atol=1e-6)

def test_to_twn(self):
mu = [1, 3]
C = [[9, 0.3], [0.3, 2]]
twn = HypertoroidalWrappedNormalDistribution(mu, C)
tfd = ToroidalFourierDistribution.from_distribution(twn, 9, 'identity')
twn_conv = tfd.to_twn()
np.testing.assert_allclose(twn_conv.mu, twn.mu, atol=1e-4)
np.testing.assert_allclose(twn_conv.C, twn.C, atol=1e-4)
tfd = ToroidalFourierDistribution.from_distribution(twn, 9, 'sqrt')
twn_conv = tfd.to_twn()
np.testing.assert_allclose(twn_conv.mu, twn.mu, atol=1e-4)
np.testing.assert_allclose(twn_conv.C, twn.C, atol=1e-4)


if __name__ == '__main__':
unittest.main()
Loading