def test_sparse_and_regular_make_same_objective(self):
np.random.seed(123)
P, Y, L = generate_simple_label_matrix(
self.known_dimensions.num_examples,
self.known_dimensions.num_functions,
self.known_dimensions.num_classes,
)
sparse_event_occurence: List[EventCooccurence] = []
label_model = LabelModel(cardinality=self.known_dimensions.num_classes)
label_model._set_constants(L)
L_shift = L + 1
label_model_lind = label_model._create_L_ind(L_shift)
co_oc_matrix = label_model_lind.T @ label_model_lind
for a_id, cols in enumerate(co_oc_matrix):
for b_id, freq in enumerate(cols):
sparse_event_occurence.append(
EventCooccurence(a_id, b_id, frequency=freq))
sparse_model = SparseEventPairLabelModel()
sparse_model._set_constants(known_dimensions=self.known_dimensions)
sparse_model_objective = sparse_model._prepare_objective_from_sparse_event_cooccurence(
known_dimensions=self.known_dimensions,
sparse_event_occurence=sparse_event_occurence,
)
self.assertEqual(label_model.n, sparse_model.n)
self.assertEqual(label_model.m, sparse_model.m)
self.assertEqual(label_model.cardinality, sparse_model.cardinality)
label_model._generate_O(L_shift, )
label_model_O = label_model.O.detach().numpy()
np.testing.assert_almost_equal(label_model_O, sparse_model_objective)
def test_sparse_and_regular_make_same_l_ind_and_o(self):
np.random.seed(123)
P, Y, L = generate_simple_label_matrix(
self.known_dimensions.num_examples,
self.known_dimensions.num_functions,
self.known_dimensions.num_classes,
)
example_event_lists: List[ExampleEventListOccurence] = []
label_model = LabelModel(cardinality=self.known_dimensions.num_classes)
label_model._set_constants(L)
L_shift = L + 1
label_model_lind = label_model._create_L_ind(L_shift)
for example_num, example in enumerate(L):
event_list = []
for func_id, cls_id in enumerate(example):
if (cls_id) > -1:
event_id = func_id * self.known_dimensions.num_classes + cls_id
event_list.append(event_id)
example_event_lists.append((ExampleEventListOccurence(event_list)))
sparse_model = SparseExampleEventListLabelModel()
sparse_model._set_constants(known_dimensions=self.known_dimensions)
sparse_model_lind = sparse_model.get_l_ind(
known_dimensions=self.known_dimensions,
example_events_list=example_event_lists,
return_array=True,
)
sparse_model_objective = sparse_model._prepare_objective_from_sparse_example_eventlist(
known_dimensions=self.known_dimensions,
example_events_list=example_event_lists,
)
np.testing.assert_equal(label_model_lind, sparse_model_lind)
np.testing.assert_equal(label_model_lind, sparse_model_lind)
self.assertEqual(label_model.n, sparse_model.n)
self.assertEqual(label_model.m, sparse_model.m)
self.assertEqual(label_model.cardinality, sparse_model.cardinality)
label_model._generate_O(L_shift, )
label_model_O = label_model.O.detach().numpy()
np.testing.assert_almost_equal(label_model_O, sparse_model_objective)
def test_sparse_and_regular_make_same_probs(self) -> None:
"""Test the LabelModel's estimate of P and Y on a simple synthetic dataset."""
np.random.seed(123)
P, Y, L = generate_simple_label_matrix(
self.known_dimensions.num_examples,
self.known_dimensions.num_functions,
self.known_dimensions.num_classes,
)
sparse_event_occurence: List[EventCooccurence] = []
label_model = LabelModel(cardinality=self.known_dimensions.num_classes)
label_model._set_constants(L)
L_shift = L + 1
label_model_lind = label_model._create_L_ind(L_shift)
co_oc_matrix = label_model_lind.T @ label_model_lind
for a_id, cols in enumerate(co_oc_matrix):
for b_id, freq in enumerate(cols):
sparse_event_occurence.append(
EventCooccurence(a_id, b_id, frequency=freq))
sparse_model = SparseEventPairLabelModel()
sparse_model.fit_from_sparse_event_cooccurrence(
sparse_event_occurence=sparse_event_occurence,
known_dimensions=self.known_dimensions,
n_epochs=200,
lr=0.01,
seed=123,
)
label_model = LabelModel(cardinality=self.known_dimensions.num_classes)
label_model.fit(L, n_epochs=200, lr=0.01, seed=123)
P_lm = label_model.get_conditional_probs()
P_slm = sparse_model.get_conditional_probs()
np.testing.assert_array_almost_equal(
P_slm,
P_lm,
)
