传播

此笔记本展示了使用标签传播对图进行聚类。

[1]:

from IPython.display import SVG

[2]:

import numpy as np

[3]:

from sknetwork.data import karate_club, painters, movie_actor
from sknetwork.clustering import PropagationClustering, get_modularity
from sknetwork.linalg import normalize
from sknetwork.utils import get_membership
from sknetwork.visualization import visualize_graph, visualize_bigraph

图

[4]:

graph = karate_club(metadata=True)
adjacency = graph.adjacency
position = graph.position

[5]:

propagation = PropagationClustering()
labels = propagation.fit_predict(adjacency)

[6]:

labels_unique, counts = np.unique(labels, return_counts=True)
print(labels_unique, counts)

[0 1] [19 15]

[7]:

image = visualize_graph(adjacency, position, labels=labels)
SVG(image)

[7]:

../../_images/tutorials_clustering_propagation_9_0.svg

[8]:

# metric
get_modularity(adjacency, labels)

[8]:

0.35231755424063116

[9]:

# aggregate graph
adjacency_aggregate = propagation.aggregate_

[10]:

average = normalize(get_membership(labels).T)
position_aggregate = average.dot(position)
labels_unique, counts = np.unique(labels, return_counts=True)

[11]:

image = visualize_graph(adjacency_aggregate, position_aggregate, counts, labels=labels_unique,
                  display_node_weight=True, node_weights=counts)
SVG(image)

[11]:

../../_images/tutorials_clustering_propagation_13_0.svg

[12]:

# probability distribution over clusters
probs = propagation.predict_proba()

[13]:

# scores for cluster 1
scores = probs[:,1]

[14]:

image = visualize_graph(adjacency, position, scores=scores)
SVG(image)

[14]:

../../_images/tutorials_clustering_propagation_16_0.svg

有向图

[15]:

graph = painters(metadata=True)
adjacency = graph.adjacency
names = graph.names
position = graph.position

[16]:

propagation = PropagationClustering()
labels = propagation.fit_predict(adjacency)

[17]:

labels_unique, counts = np.unique(labels, return_counts=True)
print(labels_unique, counts)

[0 1] [10  4]

[18]:

image = visualize_graph(adjacency, position, names=names, labels=labels)
SVG(image)

[18]:

../../_images/tutorials_clustering_propagation_21_0.svg

[19]:

get_modularity(adjacency, labels)

[19]:

0.256

[20]:

# aggregate graph
adjacency_aggregate = propagation.aggregate_

[21]:

average = normalize(get_membership(labels).T)
position_aggregate = average.dot(position)
labels_unique, counts = np.unique(labels, return_counts=True)

[22]:

image = visualize_graph(adjacency_aggregate, position_aggregate, counts, labels=labels_unique,
                    display_node_weight=True, node_weights=counts)
SVG(image)

[22]:

../../_images/tutorials_clustering_propagation_25_0.svg

[23]:

# probability distribution over clusters
probs = propagation.predict_proba()

[24]:

# scores for cluster 1
scores = probs[:,1]

[25]:

image = visualize_graph(adjacency, position, scores=scores)
SVG(image)

[25]:

../../_images/tutorials_clustering_propagation_28_0.svg

二部图

[26]:

graph = movie_actor(metadata=True)
biadjacency = graph.biadjacency
names_row = graph.names_row
names_col = graph.names_col

[27]:

propagation = PropagationClustering()
propagation.fit(biadjacency)
labels_row = propagation.labels_row_
labels_col = propagation.labels_col_

[28]:

image = visualize_bigraph(biadjacency, names_row, names_col, labels_row, labels_col)
SVG(image)

[28]:

../../_images/tutorials_clustering_propagation_32_0.svg

[29]:

get_modularity(biadjacency, labels_row, labels_col)

[29]:

0.0

[30]:

# aggregate graph
biadjacency_aggregate = propagation.aggregate_

[31]:

labels_unique_row, counts_row = np.unique(labels_row, return_counts=True)
labels_unique_col, counts_col = np.unique(labels_col, return_counts=True)

[32]:

image = visualize_bigraph(biadjacency_aggregate, counts_row, counts_col, labels_unique_row, labels_unique_col,
                    display_node_weight=True, node_weights_row=counts_row, node_weights_col=counts_col)
SVG(image)

[32]:

../../_images/tutorials_clustering_propagation_36_0.svg

[33]:

# probability distribution over clusters
probs_row = propagation.predict_proba()
probs_col = propagation.predict_proba(columns=True)