thanhphucb2 / graphembedding

Implementation and experiments of graph embedding algorithms.

GraphEmbedding

Method

Model	Paper	Note
DeepWalk	[KDD 2014]DeepWalk: Online Learning of Social Representations	【Graph Embedding】DeepWalk：算法原理，实现和应用
LINE	[WWW 2015]LINE: Large-scale Information Network Embedding	【Graph Embedding】LINE：算法原理，实现和应用
Node2Vec	[KDD 2016]node2vec: Scalable Feature Learning for Networks	【Graph Embedding】Node2Vec：算法原理，实现和应用
SDNE	[KDD 2016]Structural Deep Network Embedding	【Graph Embedding】SDNE：算法原理，实现和应用
Struc2Vec	[KDD 2017]struc2vec: Learning Node Representations from Structural Identity	【Graph Embedding】Struc2Vec：算法原理，实现和应用

How to run examples

1. clone the repo and make sure you have installed tensorflow or tensorflow-gpu on your local machine.
2. run following commands

python setup.py install
cd examples
python deepwalk_wiki.py

DisscussionGroup & Related Projects

公众号：浅梦的学习笔记

微信：deepctrbot

AlgoNotes DeepCTR DeepMatch DeepCTR-Torch

Usage

The design and implementation follows simple principles(graph in,embedding out) as much as possible.

Input format

we use networkxto create graphs.The input of networkx graph is as follows: node1 node2 <edge_weight>

DeepWalk

G = nx.read_edgelist('../data/wiki/Wiki_edgelist.txt',create_using=nx.DiGraph(),nodetype=None,data=[('weight',int)])# Read graph

model = DeepWalk(G,walk_length=10,num_walks=80,workers=1)#init model
model.train(window_size=5,iter=3)# train model
embeddings = model.get_embeddings()# get embedding vectors

LINE

G = nx.read_edgelist('../data/wiki/Wiki_edgelist.txt',create_using=nx.DiGraph(),nodetype=None,data=[('weight',int)])#read graph

model = LINE(G,embedding_size=128,order='second') #init model,order can be ['first','second','all']
model.train(batch_size=1024,epochs=50,verbose=2)# train model
embeddings = model.get_embeddings()# get embedding vectors

Node2Vec

G=nx.read_edgelist('../data/wiki/Wiki_edgelist.txt',
                        create_using = nx.DiGraph(), nodetype = None, data = [('weight', int)])#read graph

model = Node2Vec(G, walk_length = 10, num_walks = 80,p = 0.25, q = 4, workers = 1)#init model
model.train(window_size = 5, iter = 3)# train model
embeddings = model.get_embeddings()# get embedding vectors

SDNE

G = nx.read_edgelist('../data/wiki/Wiki_edgelist.txt',create_using=nx.DiGraph(),nodetype=None,data=[('weight',int)])#read graph

model = SDNE(G,hidden_size=[256,128]) #init model
model.train(batch_size=3000,epochs=40,verbose=2)# train model
embeddings = model.get_embeddings()# get embedding vectors

Struc2Vec

G = nx.read_edgelist('../data/flight/brazil-airports.edgelist',create_using=nx.DiGraph(),nodetype=None,data=[('weight',int)])#read graph

model = model = Struc2Vec(G, 10, 80, workers=4, verbose=40, ) #init model
model.train(window_size = 5, iter = 3)# train model
embeddings = model.get_embeddings()# get embedding vectors