def test_posterior_cost():
likelihood = MagicMock(return_value=4)
prior = MagicMock(return_value=5)
posterior = Posterior(likelihood, prior)
assert posterior(44.4) == -posterior.cost(44.4)
def test_posterior_cost_gradient():
likelihood = MagicMock(return_value=4)
likelihood.gradient.return_value = 0
prior = MagicMock(return_value=5)
prior.gradient.return_value = 1
posterior = Posterior(likelihood, prior)
assert posterior.gradient(44.4) == -posterior.cost_gradient(44.4)
def test_posterior_gradient():
likelihood = MagicMock(return_value=4)
prior = MagicMock(return_value=5)
posterior = Posterior(likelihood, prior)
posterior.gradient(44.4)
likelihood.gradient.assert_called()
prior.gradient.assert_called()
def test_posterior_initial_guess_default_args():
likelihood = MagicMock()
likelihood.side_effect = lambda x: x
prior = MagicMock()
prior.side_effect = lambda x: x
prior.sample.side_effect = range(200)
posterior = Posterior(likelihood, prior)
samples = posterior.generate_initial_guesses()
assert samples == [99]
def test_posterior_call():
likelihood = MagicMock(return_value=4)
prior = MagicMock(return_value=5)
posterior = Posterior(likelihood, prior)
result = posterior(44.4)
assert result == 9
cost = posterior.cost(55.5)
assert result == -cost
likelihood.assert_called()
prior.assert_called()
def test_posterior_initial_guess():
likelihood = MagicMock()
likelihood.side_effect = lambda x: x
prior = MagicMock()
prior.side_effect = lambda x: x
prior.sample.side_effect = range(100)
posterior = Posterior(likelihood, prior)
samples = posterior.generate_initial_guesses(n_guesses=2, prior_samples=10)
assert samples == [9, 8]
prior.sample.side_effect = range(100)
samples = posterior.generate_initial_guesses(n_guesses=1, prior_samples=1)
assert samples == [0]
def test_posterior_bad_initial_guess_types():
likelihood = MagicMock(return_value=4)
prior = MagicMock(return_value=5)
posterior = Posterior(likelihood, prior)
with pytest.raises(TypeError):
posterior.generate_initial_guesses(2.2)
with pytest.raises(TypeError):
posterior.generate_initial_guesses(2, 3.3)
UniformPrior(lower=0., upper=12., variable_indices=[2]) ] # Now we use the JointPrior class to combine the various components into a single prior # distribution which covers all the model parameters. prior = JointPrior(components=prior_components, n_variables=4) # As with the likelihood, prior objects can also be called as function to return a # log-probability value when passed a vector of model parameters. We can also draw # samples from the prior directly using the sample() method: prior_sample = prior.sample() print(prior_sample) # The likelihood and prior can be easily combined into a posterior distribution # using the Posterior class: from inference.posterior import Posterior posterior = Posterior(likelihood=likelihood, prior=prior) # Now we have constructed a posterior distribution, we can sample from it # using Markov-chain Monte-Carlo (MCMC). # The inference.mcmc module contains implementations of various MCMC sampling algorithms. # Here we import the PcaChain class and use it to create a Markov-chain object: from inference.mcmc import PcaChain chain = PcaChain(posterior=posterior, start=initial_guess) # We generate samples by advancing the chain by a chosen number of steps using the advance method: chain.advance(25000) # we can check the status of the chain using the plot_diagnostics method: chain.plot_diagnostics()
def test_posterior_bad_initial_guess_values():
likelihood = MagicMock(return_value=4)
prior = MagicMock(return_value=5)
posterior = Posterior(likelihood, prior)
with pytest.raises(ValueError):
posterior.generate_initial_guesses(-1)
with pytest.raises(ValueError):
posterior.generate_initial_guesses(0)
with pytest.raises(ValueError):
posterior.generate_initial_guesses(1, -3)
with pytest.raises(ValueError):
posterior.generate_initial_guesses(1, 0)
with pytest.raises(ValueError):
posterior.generate_initial_guesses(2, 1)