- clone the repo and make sure you have installed
tensorflowortensorflow-gpuon your local machine. - run following commands
python setup.py install cd examples python deepwalk_wiki.py 公众号：浅梦的学习笔记 | 微信：deepctrbot |
The design and implementation follows simple principles(graph in,embedding out) as much as possible.
we use networkxto create graphs.The input of networkx graph is as follows: node1 node2 <edge_weight>
G=nx.read_edgelist('../data/wiki/Wiki_edgelist.txt',create_using=nx.DiGraph(),nodetype=None,data=[('weight',int)])# Read graphmodel=DeepWalk(G,walk_length=10,num_walks=80,workers=1)#init modelmodel.train(window_size=5,iter=3)# train modelembeddings=model.get_embeddings()# get embedding vectorsG=nx.read_edgelist('../data/wiki/Wiki_edgelist.txt',create_using=nx.DiGraph(),nodetype=None,data=[('weight',int)])#read graphmodel=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 modelembeddings=model.get_embeddings()# get embedding vectorsG=nx.read_edgelist('../data/wiki/Wiki_edgelist.txt', create_using=nx.DiGraph(), nodetype=None, data= [('weight', int)])#read graphmodel=Node2Vec(G, walk_length=10, num_walks=80,p=0.25, q=4, workers=1)#init modelmodel.train(window_size=5, iter=3)# train modelembeddings=model.get_embeddings()# get embedding vectorsG=nx.read_edgelist('../data/wiki/Wiki_edgelist.txt',create_using=nx.DiGraph(),nodetype=None,data=[('weight',int)])#read graphmodel=SDNE(G,hidden_size=[256,128]) #init modelmodel.train(batch_size=3000,epochs=40,verbose=2)# train modelembeddings=model.get_embeddings()# get embedding vectorsG=nx.read_edgelist('../data/flight/brazil-airports.edgelist',create_using=nx.DiGraph(),nodetype=None,data=[('weight',int)])#read graphmodel=model=Struc2Vec(G, 10, 80, workers=4, verbose=40, ) #init modelmodel.train(window_size=5, iter=3)# train modelembeddings=model.get_embeddings()# get embedding vectors