def test_tangent_and_untangent_space(get_covmats):
"""Test tangent space projection and retro-projection should be the same"""
n_matrices, n_channels = 10, 3
covmats = get_covmats(n_matrices, n_channels)
Xts = tangent_space(covmats, np.eye(n_channels))
covmats_ut = untangent_space(Xts, np.eye(n_channels))
assert covmats_ut == approx(covmats)
def test_untangent_space(rndstate):
"""Test untangent space projection"""
n_matrices, n_channels = 10, 3
n_ts = (n_channels * (n_channels + 1)) // 2
T = rndstate.randn(n_matrices, n_ts)
covmats = untangent_space(T, np.eye(n_channels))
assert covmats.shape == (n_matrices, n_channels, n_channels)
def inverse_transform(self, X, y=None):
"""Inverse transform.
Project back a set of tangent space vector in the manifold.
Parameters
----------
X : ndarray, shape (n_trials, n_ts)
ndarray of SPD matrices.
y : ndarray | None (default None)
Not used, here for compatibility with sklearn API.
Returns
-------
cov : ndarray, shape (n_trials, n_channels, n_channels)
the covariance matrices corresponding to each of tangent vector.
"""
self._check_reference_points(X)
return untangent_space(X, self.reference_)
def test_tangent_and_untangent_space():
"""Test tangent space projection and retro-projection should be the same"""
C = generate_cov(10,3)
T = tangent_space(C,np.eye(3))
covmats = untangent_space(T,np.eye(3))
assert_array_equal(C,covmats)
def test_untangent_space():
"""Test untangent space projection"""
T = np.random.randn(10,6)
covmats = untangent_space(T,np.eye(3))
