Skip to content
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

Multivariate Gaussian sensitivities #978

Merged
merged 4 commits into from
Oct 1, 2019
Merged

Multivariate Gaussian sensitivities #978

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
1fa643c
Added MV Gaussian sensitivities and tests
ben18785 Sep 29, 2019
2718dc3
Update test_log_priors.py
ben18785 Sep 30, 2019
e73566e
Fixed log prior tests
MichaelClerx Sep 30, 2019
94c429c
Simplified log prior tests
MichaelClerx Sep 30, 2019
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
5 changes: 5 additions & 0 deletions pints/_log_priors.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -660,12 +660,17 @@ def __init__(self, mean, cov):
self._mean = mean
self._cov = cov
self._n_parameters = mean.shape[0]
self._cov_inverse = np.linalg.inv(self._cov)

def __call__(self, x):
return np.log(
scipy.stats.multivariate_normal.pdf(
x, mean=self._mean, cov=self._cov))

def evaluateS1(self, x):
""" See :meth:`LogPDF.evaluateS1()`. """
return self(x), -np.matmul(self._cov_inverse, x - self._mean)

def mean(self):
""" See :meth:`LogPrior.mean()`. """
return self._mean
Expand Down
30 changes: 29 additions & 1 deletion pints/tests/test_log_priors.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -604,7 +604,13 @@ def test_multivariate_normal_prior(self):
# Basic test
covariance = [[1]]
p = pints.MultivariateGaussianLogPrior(mean, covariance)
self.assertEqual(p([0]), -0.5 * np.log(2 * np.pi))
x = [0]
y = p(x)
self.assertEqual(y, -0.5 * np.log(2 * np.pi))
y1, dy = p.evaluateS1(x)
MichaelClerx marked this conversation as resolved.
Show resolved Hide resolved
self.assertEqual(y, y1)
self.assertTrue(len(dy), 1)
self.assertEqual(dy[0], 0)

# 5d tests
mean = [1, 2, 3, 4, 5]
Expand All @@ -623,6 +629,28 @@ def test_multivariate_normal_prior(self):
ValueError, pints.MultivariateGaussianLogPrior, [1, 2],
[[1, 0, 0], [0, 1, 0], [0, 0, 1]])

# Test sensitivities
mean = [1, 3]
covariance = [[2, 0.5], [0.5, 2]]
p = pints.MultivariateGaussianLogPrior(mean, covariance)
y, dy = p.evaluateS1([4, 5])
self.assertEqual(len(dy), 2)
self.assertAlmostEqual(y, -5.165421653067172, places=6)
dy_test = [-float(4 / 3), -float(2 / 3)]
self.assertAlmostEqual(dy[0], dy_test[0], places=6)
self.assertAlmostEqual(dy[1], dy_test[1], places=6)

mean = [-5.5, 6.7, 3.2]
covariance = [[3.4, -0.5, -0.7], [-0.5, 2.7, 1.4], [-0.7, 1.4, 5]]
p = pints.MultivariateGaussianLogPrior(mean, covariance)
y, dy = p.evaluateS1([4.4, 3.5, -3])
self.assertEqual(len(dy), 3)
self.assertAlmostEqual(y, -20.855279298674258, places=6)
dy_test = [-2.709773397444412, 0.27739553170576203, 0.7829609754801692]
self.assertAlmostEqual(dy[0], dy_test[0], places=6)
self.assertAlmostEqual(dy[1], dy_test[1], places=6)
self.assertAlmostEqual(dy[2], dy_test[2], places=6)

def test_multivariate_normal_sampling(self):
d = 1
mean = 2
Expand Down