def test_group_api_params(three_var_model, raises, params, type_, kw, formula):
with three_var_model, raises:
g = Group([three_var_model.one], params=params, **kw)
assert isinstance(g, type_)
if g.has_logq:
# should work as well
logq = g.logq
logq = g.set_size_and_deterministic(logq, 1, 0)
logq.eval()
def test_group_api_vfam(three_var_model, raises, vfam, type_, kw):
with three_var_model, raises:
g = Group([three_var_model.one], vfam, **kw)
assert isinstance(g, type_)
assert not hasattr(g, "_kwargs")
if isinstance(g, NormalizingFlowGroup):
assert isinstance(g.flow, pm.flows.AbstractFlow)
assert g.flow.formula == vfam
def __init__(self, *args, **kwargs):
local_rv = kwargs.get("local_rv")
groups = [self._group_class(None, *args, **kwargs)]
if local_rv is not None:
groups.extend([
Group([v], params=p, local=True, model=kwargs.get("model"))
for v, p in local_rv.items()
])
super().__init__(groups, model=kwargs.get("model"))
def test_rowwise_approx(three_var_model, parametric_grouped_approxes):
# add to inference that supports aevb
cls, kw = parametric_grouped_approxes
with three_var_model:
try:
approx = Approximation([
cls([three_var_model.one], rowwise=True, **kw),
Group(None, vfam="mf")
])
inference = pm.KLqp(approx)
approx = inference.fit(3, obj_n_mc=2)
approx.sample(10)
approx.sample_node(three_var_model.one).eval()
except pm.opvi.BatchedGroupError:
pytest.skip("Does not support rowwise grouping")
def test_group_api_vfam(three_var_model, raises, vfam, type_, kw):
with three_var_model, raises:
g = Group([three_var_model.one], vfam, **kw)
assert isinstance(g, type_)
assert not hasattr(g, "_kwargs")