def _get_parent_graph(parents_neighbors_coeffs, exclude=None):
"""
Iterates through the input parent-neighghbour coefficient dictionary to
return only parent relations (i.e. where tau != 0)
"""
graph = defaultdict(list)
for j, i, tau, _ in pp._iter_coeffs(parents_neighbors_coeffs):
if tau != 0 and (i, tau) != exclude:
graph[j].append((i, tau))
return dict(graph)
def _get_parent_graph(parents_neighbors_coeffs, exclude=None):
"""
Iterates through the input parent-neighghbour coefficient dictionary to
return only parent relations (i.e. where tau != 0)
"""
graph = defaultdict(list)
for j, i, tau, _ in pp._iter_coeffs(parents_neighbors_coeffs):
if tau != 0 and (i, tau) != exclude:
graph[j].append((i, tau))
return dict(graph)
def test_linear_mediation_coeffs(data_frame_a):
# Build the dataframe and the model
(dataframe, true_parents), links_coeffs = data_frame_a
med = LinearMediation(dataframe=dataframe)
# Fit the model
med.fit_model(all_parents=true_parents, tau_max=3)
# Ensure the results make sense
for j, i, tau, coeff in pp._iter_coeffs(links_coeffs):
np.testing.assert_allclose(med.get_coeff(i=i, tau=tau, j=j),
coeff, rtol=1e-1)
def test_covariance_construction(covariance_parameters):
"""
Test the random noise covariance matrix construction from a set of
parameters
"""
# Unpack the covariance matrix and parameters
good_params, covar_matrix = covariance_parameters
# Check the values are passed correctly
for j, i, _, coeff in pp._iter_coeffs(good_params):
covar_coeff = covar_matrix[j, i]
err_message = "Node {} and parent node {} have".format(j, i)+\
" coefficient {} for tau == 0,\nbut the".format(coeff)+\
" coefficient in the covariance matrix"+\
" is {} ".format(covar_coeff)
np.testing.assert_approx_equal(coeff, covar_coeff, err_msg=err_message)
def test_covariance_construction(covariance_parameters):
"""
Test the random noise covariance matrix construction from a set of
parameters
"""
# Unpack the covariance matrix and parameters
good_params, covar_matrix = covariance_parameters
# Check the values are passed correctly
for j, i, _, coeff in pp._iter_coeffs(good_params):
covar_coeff = covar_matrix[j, i]
err_message = "Node {} and parent node {} have".format(j, i)+\
" coefficient {} for tau == 0,\nbut the".format(coeff)+\
" coefficient in the covariance matrix"+\
" is {} ".format(covar_coeff)
np.testing.assert_approx_equal(coeff, covar_coeff, err_msg=err_message)