def plot(self, **kwargs):
r"""Plot the graph.
See :func:`pygsp.plotting.plot_graph`.
"""
from pygsp import plotting
plotting.plot_graph(self, **kwargs)
def plot(self, **kwargs):
r"""
Plot the graph.
See plotting doc.
"""
from pygsp import plotting
plotting.plot_graph(self, show_plot=True, **kwargs)
def plot(self, **kwargs):
r"""
Plot the graph.
See plotting doc.
"""
from pygsp import plotting
plotting.plot_graph(self, **kwargs)
L_reg = Gs[i].L + reg_eps * sparse.eye(Gs[i].N)
Gs[i].mr['K_reg'] = kronReduction(L_reg, ind)
Gs[i].mr['green_kernel'] = filters.Filter(Gs[i], lambda x: 1. /
(reg_eps + x))
return Gs
G = graphs.SwissRoll(N=1000, seed=42)
levels = 5
Gs = multiresolution(G, levels, sparsify=True)
fig = plt.figure(figsize=(10, 2.5))
for i in range(4):
ax = fig.add_subplot(1, 4, i + 1, projection='3d')
plotting.plot_graph(Gs[i + 1], ax=ax)
_ = ax.set_title(
'Pyramid Level: {} \n Number of nodes: {} \n Number of edges: {}'.
format(i + 1, Gs[i + 1].N, Gs[i + 1].Ne))
ax.set_axis_off()
fig.tight_layout()
plt.show()
G = graphs.Sensor(1200, distribute=True)
levels = 5
Gs = multiresolution(G, levels, sparsify=True)
fig = plt.figure(figsize=(10, 2.5))
for i in range(4):
ax = fig.add_subplot(1, 4, i + 1) # , projection='3d'
plotting.plot_graph(Gs[i + 1], ax=ax)
import numpy as np from pycgsp.graph import cvxhull_graph from pygsp.plotting import plot_graph theta, phi = np.linspace(0,np.pi,6, endpoint=False)[1:], np.linspace(0,2*np.pi,9, endpoint=False) theta, phi = np.meshgrid(theta, phi) x,y,z = np.cos(phi)*np.sin(theta), np.sin(phi)*np.sin(theta), np.cos(theta) R = np.stack((x.flatten(), y.flatten(), z.flatten()), axis=-1) G, _ = cvxhull_graph(R) plot_graph(G)
ch_names=montage,
unit="m")
C = np.matmul(X, X.T) / X.shape[1]
C = C - np.eye(C.shape[0])
C[C < threshold] = 0
G = graphs.Graph(C, lap_type="normalized", coords=montage.get_pos2d())
print("- Nodes:", G.N)
print("- Edges:", G.Ne)
return G
if __name__ == '__main__':
from pygsp import plotting
from applications.eeg.eye_dataset import MONTAGE
s = load_subject(8, True)
r = load_run_from_subject(s, 5)
X, y = get_trial(r, 47)
data, labels = get_subject_dataset(1, False)
print(data.shape)
q = 0.1
k = 0.15
plot_montage(MONTAGE)
G = create_spatial_eeg_graph(MONTAGE, q, k)
plotting.plot_graph(G, "matplotlib", save_as="graph")
fig = plt.figure()
plt.imshow(G.W.todense())
plt.colorbar()
fig.savefig("W.jpg")
# plot_spectrum(3, 2, q, k)
l, v = np.linalg.eig(C)
l_0, v_0 = np.linalg.eig(C_0)
l_1, v_1 = np.linalg.eig(C_1)
P_0 = np.matmul(np.matmul(v_0.T, C_0), v_0)
P_1 = np.matmul(np.matmul(v_1.T, C_1), v_1)
P = np.matmul(np.matmul(v.T, C), v)
print("P_0: %1.3f, P_1: %1.3f" %
(np.matrix.trace(P_0), np.matrix.trace(P_1)))
G_rbf = create_spatial_eeg_graph(MONTAGE, q=0.1, k=0.1)
G_rbf.compute_laplacian("normalized")
G_rbf.compute_fourier_basis()
U_rbf = G_rbf.U
plotting.plot_graph(G_rbf, "matplotlib", save_as="grap")
fig = plt.figure()
plt.imshow(G_rbf.W.todense())
plt.colorbar()
fig.savefig("W.png")
P_0_rbf = np.matmul(np.matmul(U_rbf.T, C_0), U_rbf)
P_1_rbf = np.matmul(np.matmul(U_rbf.T, C_1), U_rbf)
P_rbf = np.matmul(np.matmul(U_rbf.T, C), U_rbf)
fig = plt.figure()
plt.imshow(P_0_rbf)
plt.colorbar()
fig.savefig("P_0_rbf.png")
