def quic_graph_lasso_ebic(X, gamma=0):
'''Run QuicGraphLassoEBIC with gamma.
QuicGraphLassoEBIC is a convenience class. Results should be identical to
those obtained via quic_graph_lasso_ebic_manual.
'''
print 'QuicGraphLassoEBIC with:'
print ' mode: path'
print ' gamma: {}'.format(gamma)
model = QuicGraphLassoEBIC(lam=1.0, init_method='cov', gamma=gamma)
model.fit(X)
print ' len(path lams): {}'.format(len(model.path_))
print ' lam_scale_: {}'.format(model.lam_scale_)
print ' lam_: {}'.format(model.lam_)
return model.covariance_, model.precision_, model.lam_
def test_integration_quic_graph_lasso_ebic(self, params_in, expected):
'''
Just tests inputs/outputs (not validity of result).
'''
X = datasets.load_diabetes().data
ic = QuicGraphLassoEBIC(**params_in)
ic.fit(X)
result_vec = [
np.linalg.norm(ic.covariance_),
np.linalg.norm(ic.precision_),
]
if isinstance(ic.lam_, float):
result_vec.append(ic.lam_)
elif isinstance(ic.lam_, np.ndarray):
assert ic.lam_.shape == params_in['lam'].shape
print result_vec
assert_allclose(expected, result_vec, rtol=1e-1)
#%%
# run analysis on all subjects
# build timeseries for first scan (before)
totalSub_series_1 = timeSeries(func_files=fileList(totalSub,'1'), confound_files=confList(totalSub, '1'))
totalSub_series_2 = timeSeries(func_files=fileList(totalSub,'2'), confound_files=confList(totalSub, '2'))
# In[skggm]:
# run EBIC LASSO on each subject
from inverse_covariance import QuicGraphLassoEBIC
# call estimator EBIC
estimator = QuicGraphLassoEBIC(init_method="cov", verbose=1)
# create emtpy array
ebicMat = []
ebicMat2 = []
# run loop over timeseries and create precision matrix for each subject, append to a list of matrices
for i in totalSub_series_1:
estimator.fit(i)
ebicMat.append(-estimator.precision_)
for i in totalSub_series_2:
estimator.fit(i)
ebicMat2.append(-estimator.precision_)
ket1_forLASSO.shape
#%%
import matplotlib.pyplot as plt
%matplotlib qt
# Display the sparse inverse covariance
plt.figure(figsize=(7.5, 7.5))
plt.imshow(
np.triu(-estimator.precision_, 1), interpolation="nearest", cmap=plt.cm.RdBu_r