def plot_cluster(data, mu, colors, ax):
"""
plot the cluster.
Note that the dimension of the column vector `colors`
should be the same as the number of clusters.
"""
# build distance matrix.
distance_matrix = build_distance_matrix(data, mu)
# get the assignments for each point.
assignments = np.argmin(distance_matrix, axis=1)
#
for k_th in range(mu.shape[0]):
#rows, cols = np.where(assignments == k_th)
rows = np.where(assignments == k_th)[0]
data_of_kth_cluster = data[rows, :]
ax.scatter(
data_of_kth_cluster[:, 0],
data_of_kth_cluster[:, 1],
# works for clusters more than 3
s=40,
c=colors[k_th % len(colors)])
ax.grid()
ax.set_xlabel("x")
ax.set_ylabel("y")
def plot_cluster(data, mu, colors, ax):
"""plot the cluster.
Note that the dimension of the column vector `colors`
should be the same as the number of clusters.
"""
# check if the dimension matches.
assert(len(colors) >= mu.shape[0])
# build distance matrix.
distance_matrix = build_distance_matrix(data, mu)
# get the assignments for each point.
assignments = np.argmin(distance_matrix, axis=1)
#
for k_th in range(mu.shape[0]):
rows, cols = np.where(assignments == k_th)
data_of_kth_cluster = data[rows, :]
ax.scatter(
data_of_kth_cluster[:, 0],
data_of_kth_cluster[:, 1],
s=40, c=colors[k_th])
ax.grid()
ax.set_xlabel("x")
ax.set_ylabel("y")
def plot_cluster(data, mu, colors, ax):
"""plot the cluster.
Note that the dimension of the column vector `colors`
should be the same as the number of clusters.
"""
# check if the dimension matches.
#assert(len(colors) >= mu.shape[0])
# build distance matrix.
distance_matrix = build_distance_matrix(data, mu)
# get the assignments for each point.
assignments = np.argmin(distance_matrix, axis=1)
#
for k_th in range(mu.shape[0]):
rows = np.where(assignments == k_th)
data_of_kth_cluster = data[rows, :]
ax.scatter(
np.squeeze(data_of_kth_cluster)[:, 0],
np.squeeze(data_of_kth_cluster)[:, 1],
s=40, c=colors[k_th % len(colors)])
ax.grid()
ax.set_xlabel("x")
ax.set_ylabel("y")
