Node Embeddings

Generate node embeddings using Node2Vec and visualize them.

from py3plex.core import multinet
from py3plex.wrappers import train_node2vec_embedding
from py3plex.visualization.embedding_visualization import embedding_visualization

# Load network
network = multinet.multi_layer_network().load_network(
    "../datasets/imdb_gml.gml", directed=True, input_type="gml")

# Save as edgelist for Node2Vec
network.save_network("../datasets/test.edgelist")

# Generate embedding
train_node2vec_embedding.call_node2vec_binary(
    "../datasets/test.edgelist",
    "../datasets/test_embedding.emb",
    binary="../bin/node2vec",
    weighted=False)

# Load and visualize
network.load_embedding("../datasets/test_embedding.emb")
embedding_visualization.visualize_embedding(network)

Examples

See:

example_n2v_embedding.py - Node2Vec embeddings
example_embedding_visualization.py - Embedding visualization
example_embedding_construction.py - Custom embeddings

Repository: https://github.com/SkBlaz/Py3Plex/tree/master/examples