def test_adjacency_equiv(shape):
"""Test adjacency equivalence for lattice adjacency."""
from sklearn.feature_extraction import grid_to_graph
# sklearn requires at least two dimensions
sk_shape = shape if len(shape) > 1 else (shape + (1, ))
conn_sk = grid_to_graph(*sk_shape).toarray()
conn = combine_adjacency(*shape)
want_shape = (np.prod(shape), ) * 2
assert conn.shape == conn_sk.shape == want_shape
assert (conn.data == 1.).all()
conn = conn.toarray()
# we end up with some duplicates that can turn into 2's and 3's,
# eventually we might want to keep these as 1's but it's easy enough
# with a .astype(bool) (also matches sklearn output) so let's leave it
# for now
assert np.in1d(conn, [0, 1, 2, 3]).all()
assert conn.shape == conn_sk.shape
assert_array_equal(conn, conn_sk)
def test_output_equiv(shape, out_type, adjacency):
"""Test equivalence of output types."""
rng = np.random.RandomState(0)
n_subjects = 10
data = rng.randn(n_subjects, *shape)
data -= data.mean(axis=0, keepdims=True)
data[:, 2:4] += 2
data[:, 6:9] += 2
want_mask = np.zeros(shape, int)
want_mask[2:4] = 1
want_mask[6:9] = 2
if adjacency is not None:
assert adjacency == 'sparse'
adjacency = combine_adjacency(*shape)
clusters = permutation_cluster_1samp_test(X=data,
n_permutations=1,
adjacency=adjacency,
out_type=out_type)[1]
got_mask = np.zeros_like(want_mask)
for n, clu in enumerate(clusters, 1):
if out_type == 'mask':
if len(shape) == 1 and adjacency is None:
assert isinstance(clu, tuple)
assert len(clu) == 1
assert isinstance(clu[0], slice)
else:
assert isinstance(clu, np.ndarray)
assert clu.dtype == bool
assert clu.shape == shape
got_mask[clu] = n
else:
assert isinstance(clu, tuple)
for c in clu:
assert isinstance(c, np.ndarray)
assert c.dtype.kind == 'i'
assert out_type == 'indices'
got_mask[np.ix_(*clu)] = n
assert_array_equal(got_mask, want_mask)
def permutation_correlation(diff_data, info, n_permutations, p_value):
sensor_adjacency, ch_names = find_ch_adjacency(info, "eeg")
adjacency = combine_adjacency(sensor_adjacency, diff_data.shape[2],
diff_data.shape[3])
T_obs, clusters, cluster_p_values, H0 = permutation_cluster_1samp_test(
diff_data,
n_permutations=n_permutations,
threshold=None,
tail=0,
adjacency=adjacency,
out_type="mask",
verbose=True,
)
# Create new stats image with only significant clusters for plotting
T_obs_plot = np.nan * np.ones_like(T_obs)
for c, p_val in zip(clusters, cluster_p_values):
if p_val <= p_value:
print(f"Significant cluster with p-value {p_val}")
T_obs_plot[c] = T_obs[c]
return T_obs_plot, cluster_p_values[cluster_p_values <= p_value]
def test_cluster_permutation_with_adjacency(numba_conditional):
"""Test cluster level permutations with adjacency matrix."""
from sklearn.feature_extraction.image import grid_to_graph
condition1_1d, condition2_1d, condition1_2d, condition2_2d = \
_get_conditions()
n_pts = condition1_1d.shape[1]
# we don't care about p-values in any of these, so do fewer permutations
args = dict(seed=None,
max_step=1,
exclude=None,
out_type='mask',
step_down_p=0,
t_power=1,
threshold=1.67,
check_disjoint=False,
n_permutations=50)
did_warn = False
for X1d, X2d, func, spatio_temporal_func in \
[(condition1_1d, condition1_2d,
permutation_cluster_1samp_test,
spatio_temporal_cluster_1samp_test),
([condition1_1d, condition2_1d],
[condition1_2d, condition2_2d],
permutation_cluster_test,
spatio_temporal_cluster_test)]:
out = func(X1d, **args)
adjacency = grid_to_graph(1, n_pts)
out_adjacency = func(X1d, adjacency=adjacency, **args)
assert_array_equal(out[0], out_adjacency[0])
for a, b in zip(out_adjacency[1], out[1]):
assert_array_equal(out[0][a], out[0][b])
assert np.all(a[b])
# test spatio-temporal w/o time adjacency (repeat spatial pattern)
adjacency_2 = sparse.coo_matrix(
linalg.block_diag(adjacency.asfptype().todense(),
adjacency.asfptype().todense()))
# nesting here is time then space:
adjacency_2a = combine_adjacency(np.eye(2), adjacency)
assert_array_equal(adjacency_2.toarray().astype(bool),
adjacency_2a.toarray().astype(bool))
if isinstance(X1d, list):
X1d_2 = [np.concatenate((x, x), axis=1) for x in X1d]
else:
X1d_2 = np.concatenate((X1d, X1d), axis=1)
out_adjacency_2 = func(X1d_2, adjacency=adjacency_2, **args)
# make sure we were operating on the same values
split = len(out[0])
assert_array_equal(out[0], out_adjacency_2[0][:split])
assert_array_equal(out[0], out_adjacency_2[0][split:])
# make sure we really got 2x the number of original clusters
n_clust_orig = len(out[1])
assert len(out_adjacency_2[1]) == 2 * n_clust_orig
# Make sure that we got the old ones back
data_1 = {np.sum(out[0][b[:n_pts]]) for b in out[1]}
data_2 = {np.sum(out_adjacency_2[0][a]) for a in out_adjacency_2[1][:]}
assert len(data_1.intersection(data_2)) == len(data_1)
# now use the other algorithm
if isinstance(X1d, list):
X1d_3 = [np.reshape(x, (-1, 2, n_space)) for x in X1d_2]
else:
X1d_3 = np.reshape(X1d_2, (-1, 2, n_space))
out_adjacency_3 = spatio_temporal_func(X1d_3,
n_permutations=50,
adjacency=adjacency,
max_step=0,
threshold=1.67,
check_disjoint=True)
# make sure we were operating on the same values
split = len(out[0])
assert_array_equal(out[0], out_adjacency_3[0][0])
assert_array_equal(out[0], out_adjacency_3[0][1])
# make sure we really got 2x the number of original clusters
assert len(out_adjacency_3[1]) == 2 * n_clust_orig
# Make sure that we got the old ones back
data_1 = {np.sum(out[0][b[:n_pts]]) for b in out[1]}
data_2 = {
np.sum(out_adjacency_3[0][a[0], a[1]])
for a in out_adjacency_3[1]
}
assert len(data_1.intersection(data_2)) == len(data_1)
# test new versus old method
out_adjacency_4 = spatio_temporal_func(X1d_3,
n_permutations=50,
adjacency=adjacency,
max_step=2,
threshold=1.67)
out_adjacency_5 = spatio_temporal_func(X1d_3,
n_permutations=50,
adjacency=adjacency,
max_step=1,
threshold=1.67)
# clusters could be in a different order
sums_4 = [np.sum(out_adjacency_4[0][a]) for a in out_adjacency_4[1]]
sums_5 = [np.sum(out_adjacency_4[0][a]) for a in out_adjacency_5[1]]
sums_4 = np.sort(sums_4)
sums_5 = np.sort(sums_5)
assert_array_almost_equal(sums_4, sums_5)
if not _force_serial:
pytest.raises(ValueError,
spatio_temporal_func,
X1d_3,
n_permutations=1,
adjacency=adjacency,
max_step=1,
threshold=1.67,
n_jobs=-1000)
# not enough TFCE params
with pytest.raises(KeyError, match='threshold, if dict, must have'):
spatio_temporal_func(X1d_3,
adjacency=adjacency,
threshold=dict(me='hello'))
# too extreme a start threshold
with pytest.warns(None) as w:
spatio_temporal_func(X1d_3,
adjacency=adjacency,
threshold=dict(start=10, step=1))
if not did_warn:
assert len(w) == 1
did_warn = True
with pytest.raises(ValueError, match='threshold.*<= 0 for tail == -1'):
spatio_temporal_func(X1d_3,
adjacency=adjacency,
tail=-1,
threshold=dict(start=1, step=-1))
with pytest.warns(RuntimeWarning, match='threshold.* is more extreme'):
spatio_temporal_func(X1d_3,
adjacency=adjacency,
tail=1,
threshold=dict(start=100, step=1))
bad_con = adjacency.todense()
with pytest.raises(ValueError, match='must be a SciPy sparse matrix'):
spatio_temporal_func(X1d_3,
n_permutations=50,
adjacency=bad_con,
max_step=1,
threshold=1.67)
bad_con = adjacency.tocsr()[:-1, :-1].tocoo()
with pytest.raises(ValueError, match='adjacency.*the correct size'):
spatio_temporal_func(X1d_3,
n_permutations=50,
adjacency=bad_con,
max_step=1,
threshold=1.67)
with pytest.raises(TypeError, match='must be a'):
spatio_temporal_func(X1d_3, adjacency=adjacency, threshold=[])
with pytest.raises(ValueError, match='Invalid value for the \'tail\''):
with pytest.warns(None): # sometimes ignoring tail
spatio_temporal_func(X1d_3, adjacency=adjacency, tail=2)
# make sure it actually found a significant point
out_adjacency_6 = spatio_temporal_func(X1d_3,
n_permutations=50,
adjacency=adjacency,
max_step=1,
threshold=dict(start=1, step=1))
assert np.min(out_adjacency_6[2]) < 0.05
with pytest.raises(ValueError, match='not compatible'):
with pytest.warns(RuntimeWarning, match='No clusters'):
spatio_temporal_func(X1d_3,
n_permutations=50,
adjacency=adjacency,
threshold=1e-3,
stat_fun=lambda *x: f_oneway(*x)[:-1],
buffer_size=None)
# transpose again to (epochs, frequencies, times, channels)
X = [np.transpose(x, (0, 2, 3, 1)) for x in epochs_power]
# %%
# Remember the note on the adjacency matrix from above: For 3D data, as here,
# we must use :func:`mne.stats.combine_adjacency` to extend the
# sensor-based adjacency to incorporate the time-frequency plane as well.
#
# Here, the integer inputs are converted into a lattice and
# combined with the sensor adjacency matrix so that data at similar
# times and with similar frequencies and at close sensor locations are
# clustered together.
# our data at each observation is of shape frequencies × times × channels
tfr_adjacency = combine_adjacency(
len(freqs), len(this_tfr.times), adjacency)
# %%
# Now we can run the cluster permutation test, but first we have to set a
# threshold. This example decimates in time and uses few frequencies so we need
# to increase the threshold from the default value in order to have
# differentiated clusters (i.e., so that our algorithm doesn't just find one
# large cluster). For a more principled method of setting this parameter,
# threshold-free cluster enhancement may be used.
# See :ref:`disc-stats` for a discussion.
# This time we don't calculate a threshold based on the F distribution.
# We might as well select an arbitrary threshold for cluster forming
tfr_threshold = 15.0
# run cluster based permutation analysis
