def test_importance_sampling(
model_approx,
y_,
z_,
):
sampler = mp.ImportanceSampler(n_samples=500)
print_factor = lambda *args: print(args[0])
print_status = lambda *args: print(args[2])
callback = mp.expectation_propagation.EPHistory(callbacks=(print_factor,
print_status))
opt = mp.EPOptimiser(model_approx.factor_graph,
default_optimiser=sampler,
callback=callback)
new_approx = opt.run(model_approx, max_steps=5)
y = new_approx.mean_field[y_].mean
z_pred = new_approx(new_approx.mean_field.mean)[z_]
y_pred = z_pred > 0
(tpr, fpr), (fnr, tnr) = np.dot(
np.array([y, 1 - y]).reshape(2, -1),
np.array([y_pred, 1 - y_pred]).reshape(2, -1).T)
accuracy = (tpr + tnr) / (tpr + fpr + fnr + tnr)
assert 0.95 > accuracy > 0.7
def test_simple(model_approx, centres):
laplace = graph.LaplaceOptimiser()
ep_opt = graph.EPOptimiser(model_approx, default_optimiser=laplace)
new_approx = ep_opt.run(model_approx, max_steps=20)
mu_ = new_approx.factor_graph.name_variable_dict["mu"]
logt_ = new_approx.factor_graph.name_variable_dict["logt"]
assert new_approx.mean_field[mu_].mean == pytest.approx(np.mean(centres),
rel=1.0)
assert new_approx.mean_field[logt_].mean == pytest.approx(np.log(
np.std(centres)**-2),
rel=1.0)
def test_laplace(
model_approx,
a_,
b_,
y_,
z_,
):
laplace = mp.LaplaceFactorOptimiser()
opt = mp.EPOptimiser(model_approx.factor_graph, default_optimiser=laplace)
model_approx = opt.run(model_approx)
y = model_approx.mean_field[y_].mean
y_pred = model_approx.mean_field[z_].mean
assert mp.utils.r2_score(y, y_pred) > 0.95
def test_path_prefix(output_directory, factor_graph_model):
paths = DirectoryPaths(
path_prefix="path_prefix",
name="name",
is_identifier_in_paths=False,
)
optimiser = g.EPOptimiser(
factor_graph=factor_graph_model.graph,
paths=paths,
default_optimiser=MockSearch(),
)
assert optimiser.output_path == output_directory / "path_prefix/name"
_run_optimisation(factor_graph_model, paths=paths)
assert (output_directory / "path_prefix/name/AnalysisFactor0").exists()
assert (output_directory / "path_prefix/name/AnalysisFactor1").exists()
def test_importance_sampling(
model,
model_approx,
linear_factor,
y_,
z_,
):
laplace = mp.LaplaceFactorOptimiser()
sampler = mp.ImportanceSampler(n_samples=500, force_sample=True, delta=0.8)
ep_opt = mp.EPOptimiser(model,
default_optimiser=laplace,
factor_optimisers={linear_factor: sampler})
model_approx = ep_opt.run(model_approx, max_steps=3)
y = model_approx.mean_field[y_].mean
y_pred = model_approx.mean_field[z_].mean
assert mp.utils.r2_score(y, y_pred) > 0.90
def test_laplace(
model_approx,
y_,
z_,
):
laplace = mp.LaplaceFactorOptimiser(default_opt_kws={'jac': True})
opt = mp.EPOptimiser(model_approx.factor_graph, default_optimiser=laplace)
new_approx = opt.run(model_approx)
y = new_approx.mean_field[y_].mean
z_pred = new_approx(new_approx.mean_field.mean)[z_]
y_pred = z_pred > 0
(tpr, fpr), (fnr, tnr) = np.dot(
np.array([y, 1 - y]).reshape(2, -1),
np.array([y_pred, 1 - y_pred]).reshape(2, -1).T)
accuracy = (tpr + tnr) / (tpr + fpr + fnr + tnr)
assert 0.95 > accuracy > 0.7
def test_hierarchical(centres, widths):
centres_ = [Variable(f"x_{i}") for i in range(n)]
mu_ = Variable("mu")
logt_ = Variable("logt")
centre_likelihoods = [
messages.NormalMessage(c, w).as_factor(x)
for c, w, x in zip(centres, widths, centres_)
]
hierarchical_factor = graph.Factor(
hierarchical_loglike_t,
mu_,
logt_,
*centres_,
factor_jacobian=hierarchical_loglike_t_jac,
)
model = graph.utils.prod(centre_likelihoods) * hierarchical_factor
model_approx = graph.EPMeanField.from_approx_dists(
model,
{
mu_: messages.NormalMessage(0.0, 10.0),
logt_: messages.NormalMessage(0.0, 10.0),
**{x_: messages.NormalMessage(0.0, 10.0)
for x_ in centres_},
},
)
laplace = graph.LaplaceOptimiser()
ep_opt = graph.EPOptimiser(model_approx, default_optimiser=laplace)
new_approx = ep_opt.run(model_approx, max_steps=10)
print(new_approx)
mu_ = new_approx.factor_graph.name_variable_dict["mu"]
logt_ = new_approx.factor_graph.name_variable_dict["logt"]
assert new_approx.mean_field[mu_].mean == pytest.approx(np.mean(centres),
rel=0.2)
assert new_approx.mean_field[logt_].mean == pytest.approx(np.log(
np.std(centres)**-2),
rel=0.2)
def test_laplace(
model,
start_approx,
y_,
z_,
):
model_approx = graph.EPMeanField.from_approx_dists(model, start_approx)
laplace = graph.LaplaceOptimiser()
opt = graph.EPOptimiser(model_approx.factor_graph, default_optimiser=laplace)
new_approx = opt.run(model_approx, max_steps=10)
y = new_approx.mean_field[y_].mean
z_pred = new_approx(new_approx.mean_field.mean)[z_]
y_pred = z_pred > 0
(tpr, fpr), (fnr, tnr) = np.dot(
np.array([y, 1 - y]).reshape(2, -1),
np.array([y_pred, 1 - y_pred]).reshape(2, -1).T,
)
accuracy = (tpr + tnr) / (tpr + fpr + fnr + tnr)
assert 0.95 > accuracy > 0.75
def test_full_hierachical(data):
samples = {
Variable(f"samples_{i}"): sample
for i, sample in enumerate(data)
}
x_i_ = [Variable(f"x_{i}") for i in range(n)]
logt_i_ = [Variable(f"logt_{i}") for i in range(n)]
mu_x_ = Variable("mu_x")
logt_x_ = Variable("logt_x")
mu_logt_ = Variable("mu_logt")
logt_logt_ = Variable("logt_logt")
hierarchical_params = (mu_x_, logt_x_, mu_logt_, logt_logt_)
# Setting up model
data_loglikes = [
graph.Factor(
normal_loglike_t,
s_,
x_,
logt_,
factor_jacobian=normal_loglike_t_jacobian,
name=f"normal_{i}",
) for i, (s_, x_, logt_) in enumerate(zip(samples, x_i_, logt_i_))
]
centre_loglike = graph.Factor(
hierarchical_loglike_t,
mu_x_,
logt_x_,
*x_i_,
name="mean_loglike",
factor_jacobian=hierarchical_loglike_t_jac,
)
logt_loglike = graph.Factor(
hierarchical_loglike_t,
mu_logt_,
logt_logt_,
*logt_i_,
name="logt_loglike",
factor_jacobian=hierarchical_loglike_t_jac,
)
priors = [
messages.NormalMessage(0.0, 10.0).as_factor(v, name=f"prior_{v.name}")
for v in hierarchical_params
]
model = graph.utils.prod(data_loglikes + priors +
[centre_loglike, logt_loglike])
model_approx = graph.EPMeanField.from_approx_dists(
model,
{
**{v: messages.NormalMessage(0.0, 10.0)
for v in model.variables},
**{
v: messages.FixedMessage(sample)
for v, sample in samples.items()
},
},
)
# Mean field approximation
model_approx = graph.EPMeanField.from_approx_dists(
model,
{
**{v: messages.NormalMessage(0.0, 10.0)
for v in model.variables},
**{
v: messages.FixedMessage(sample)
for v, sample in samples.items()
},
},
)
laplace = graph.LaplaceOptimiser()
ep_opt = graph.EPOptimiser(model, default_optimiser=laplace)
new_approx = ep_opt.run(model_approx, max_steps=20)
new_approx.mean_field.subset(hierarchical_params)
m = np.mean([np.mean(sample) for sample in data])
logt = np.mean([np.log(np.std(sample)**-2) for sample in data])
assert new_approx.mean_field[mu_x_].mean == pytest.approx(m, rel=1.0)
assert new_approx.mean_field[mu_logt_].mean == pytest.approx(logt, rel=1.0)