def test_plot_points_h2_poincare_disk(self):
points = self.H2.random_uniform(self.n_samples)
visualization.plot(points, space='H2_poincare_disk')
def test_plot_points_s2(self):
points = self.S2.random_uniform(self.n_samples)
visualization.plot(points, space='S2')
x = torch.from_numpy(x).reshape((1, 3, 32, 32)).float()
with torch.no_grad():
x = net(x)
return entropy_loss(x, torch.zeros(1, dtype=torch.long))
fig = plt.figure(figsize=(10, 10))
ax = fig.add_subplot(111, projection='3d')
sphere = Sphere()
# f_points = sphere.fibonnaci_points(10000).swapaxes(0, 1)
# tf_points = transformer.transform(f_points, base_point)
#
# print(tf_points[:, 0].min(), tf_points[:, 0].max())
# print(tf_points[:, 1].min(), tf_points[:, 1].max())
# print(tf_points[:, 2].min(), tf_points[:, 2].max())
sphere.plot_heatmap(ax=ax, n_points=10000, scalar_function=loss_f)
correct_points = points[labels == 0]
correct_labels = labels[labels == 0]
wrong_points = points[labels != 0]
wrong_labels = labels[labels != 0]
plot_points = np.concatenate((correct_points[:100], wrong_points[:100]))
plot_labels = np.concatenate((correct_labels[:100], wrong_labels[:100]))
visualization.plot(plot_points,
ax=ax,
space='S2',
c=plot_labels == 0,
s=80,
alpha=0.5)
plt.show()
def test_plot_points_se3(self):
points = self.SE3_GROUP.random_uniform(self.n_samples)
visualization.plot(points, space='SE3_GROUP')
def main():
"""Perform tangent PCA at the mean."""
fig = plt.figure(figsize=(15, 5))
hyperbolic_plane = Hyperbolic(dimension=2)
data = hyperbolic_plane.random_uniform(n_samples=140)
mean = FrechetMean(metric=hyperbolic_plane.metric)
mean.fit(data)
mean_estimate = mean.estimate_
tpca = TangentPCA(metric=hyperbolic_plane.metric, n_components=2)
tpca = tpca.fit(data, base_point=mean_estimate)
tangent_projected_data = tpca.transform(data)
geodesic_0 = hyperbolic_plane.metric.geodesic(
initial_point=mean_estimate, initial_tangent_vec=tpca.components_[0])
geodesic_1 = hyperbolic_plane.metric.geodesic(
initial_point=mean_estimate, initial_tangent_vec=tpca.components_[1])
n_steps = 100
t = np.linspace(-1, 1, n_steps)
geodesic_points_0 = geodesic_0(t)
geodesic_points_1 = geodesic_1(t)
print('Coordinates of the Log of the first 5 data points at the mean, '
'projected on the principal components:')
print(tangent_projected_data[:5])
ax_var = fig.add_subplot(121)
xticks = np.arange(1, 2 + 1, 1)
ax_var.xaxis.set_ticks(xticks)
ax_var.set_title('Explained variance')
ax_var.set_xlabel('Number of Principal Components')
ax_var.set_ylim((0, 1))
ax_var.plot(xticks, tpca.explained_variance_ratio_)
ax = fig.add_subplot(122)
visualization.plot(mean_estimate,
ax,
space='H2_poincare_disk',
color='darkgreen',
s=10)
visualization.plot(geodesic_points_0,
ax,
space='H2_poincare_disk',
linewidth=2)
visualization.plot(geodesic_points_1,
ax,
space='H2_poincare_disk',
linewidth=2)
visualization.plot(data,
ax,
space='H2_poincare_disk',
color='black',
alpha=0.7)
plt.show()