def test_logprior(self):
t0 = [10.0]
lpost = PoissonPosterior(self.x, self.y, self.model)
lpost.logprior = set_logprior(lpost, self.priors)
lp_test = lpost.logprior(t0)
lp = np.log(scipy.stats.norm(self.countrate, self.countrate).pdf(t0))
assert lp == lp_test
def test_counts_are_nan(self):
y = np.nan * np.ones(self.x.shape[0])
t0 = [10.0]
self.model.amplitude = t0[0]
mean_model = self.model(self.x)
lpost = PoissonPosterior(self.x, y, self.model)
lpost.logprior = set_logprior(lpost, self.priors)
assert np.isclose(lpost(t0), logmin, 1e-5)
def test_negative_loglikelihood(self):
t0 = [10.0]
self.model.amplitude = t0[0]
mean_model = self.model(self.x)
loglike = -np.sum(-mean_model + self.y*np.log(mean_model) - scipy_gammaln(self.y+1))
lpost = PoissonPosterior(self.x, self.y, self.model)
lpost.logprior = set_logprior(lpost, self.priors)
loglike_test = lpost.loglikelihood(t0, neg=True)
assert np.isclose(loglike, loglike_test)
def test_negative_posterior(self):
t0 = [10.0]
self.model.amplitude = t0[0]
mean_model = self.model(self.x)
lpost = PoissonPosterior(self.x, self.y, self.model)
lpost.logprior = set_logprior(lpost, self.priors)
post_test = lpost(t0, neg=True)
loglike = np.sum(-mean_model + self.y*np.log(mean_model) - scipy_gammaln(self.y+1))
logprior = np.log(scipy.stats.norm(self.countrate, self.countrate).pdf(t0))
post = -loglike - logprior
assert np.isclose(post_test, post, atol=1.e-10)
def test_correct_number_of_parameters(self):
lpost = PoissonPosterior(self.x, self.y, self.model)
lpost.logprior = set_logprior(lpost, self.priors)
with pytest.raises(IncorrectParameterError):
lpost([2, 3])
def test_making_posterior(self):
lpost = PoissonPosterior(self.x, self.y, self.model)
lpost.logprior = set_logprior(lpost, self.priors)
assert lpost.x.all() == self.x.all()
assert lpost.y.all() == self.y.all()
def test_logprior_fails_without_prior(self):
lpost = PoissonPosterior(self.x, self.y, self.model)
with pytest.raises(AttributeError):
lpost.logprior([10])