def main(graph_file_name, path_file_name):
g = Graph()
print("Reading...")
g.read_edgelist(filename=graph_file_name)
walker_class = walker.BasicWalker(g, workers=8)
sentences = walker_class.simulate_walks(
num_walks=10, walk_length=10)
fw = codecs.open(path_file_name, "w", "utf-8")
for path in sentences:
for i in range(0, len(path)):
if i < len(path) - 1:
fw.write(str(path[i] + " "))
else:
fw.write(str(path[i] + "\n"))
fw.close()
def __init__(self,
graph,
path_length,
num_paths,
dim,
p=1.0,
q=1.0,
dw=False,
**kwargs):
kwargs["workers"] = kwargs.get("workers", 1)
if dw:
kwargs["hs"] = 1
p = 1.0
q = 1.0
self.graph = graph
if dw:
self.walker = walker.BasicWalker(graph, workers=kwargs["workers"])
else:
self.walker = walker.Walker(graph,
p=p,
q=q,
workers=kwargs["workers"])
# self.walker = walker.QuantumWalker(graph, p=p, q=q, workers=kwargs["workers"])
print("Preprocess transition probs...")
self.walker.preprocess_transition_probs()
sentences = self.walker.simulate_walks(num_walks=num_paths,
walk_length=path_length)
kwargs["sentences"] = sentences
kwargs["min_count"] = kwargs.get("min_count", 0)
kwargs["size"] = kwargs.get("size", dim)
kwargs["sg"] = 1
self.size = kwargs["size"]
print("Learning representation...")
word2vec = Word2Vec(**kwargs)
self.vectors = {}
self.embeddings = np.zeros((graph.node_size, dim))
for word in graph.G.nodes():
self.vectors[word] = word2vec[word]
self.embeddings[graph.look_up_dict[word]] = word2vec[word]
del word2vec
def __init__(self,
graph,
path_length,
num_paths,
dim,
p=1.0,
q=1.0,
dw=0,
**kwargs):
kwargs["workers"] = kwargs.get("workers", 1)
if dw:
kwargs["hs"] = 1
p = 1.0
q = 1.0
self.graph = graph
if dw == 1:
self.walker = walker.BasicWalker(graph, workers=kwargs["workers"])
elif dw == 2:
self.walker = walker.MHWalker(graph, workers=kwargs["workers"])
else:
self.walker = walker.Walker(graph,
p=p,
q=q,
workers=kwargs["workers"])
print("Preprocess transition probs...")
self.walker.preprocess_transition_probs()
sentences = self.walker.simulate_walks(num_walks=num_paths,
walk_length=path_length)
kwargs["sentences"] = sentences
kwargs["min_count"] = kwargs.get("min_count", 0)
kwargs["size"] = kwargs.get("size", dim)
kwargs["sg"] = 1
self.size = kwargs["size"]
print("Learning representation...")
word2vec = Word2Vec(**kwargs)
self.vectors = {}
for word in graph.G.nodes():
self.vectors[word] = word2vec.wv[word]
del word2vec
def __init__(self, graph, path_length, num_paths, dim, p=1.0, q=1.0, dw=False, **kwargs):
kwargs["workers"] = kwargs.get("workers", 5)
if dw:
kwargs["hs"] = 0
p = 1.0
q = 1.0
t1 = time.time()
self.graph = graph
if dw:
self.walker = walker.BasicWalker(graph, workers=kwargs["workers"],walk_length=path_length)
t2 = time.time()
self.walker.preprocess_transition_probs()
t3 = time.time()
print("prob time: "+str(t3-t2))
sentences = self.walker.simulate_walks(num_walks=num_paths, walk_length=path_length)
else:
self.walker = walker.Walker(
graph, p=p, q=q, workers=kwargs["workers"])
t2 = time.time()
self.walker.preprocess_transition_probs()
t3 = time.time()
print("prob time: "+str(t3-t2))
sentences = self.walker.simulate_walks(num_walks=num_paths, walk_length=path_length)
t2 = time.time()
print("walking time: "+str(t2-t1))
kwargs["sentences"] = sentences
kwargs["min_count"] = kwargs.get("min_count", 0)
kwargs["size"] = kwargs.get("size", dim)
kwargs["sg"] = 1
self.size = kwargs["size"]
# print("Learning representation...")
word2vec = Word2Vec(**kwargs)
self.vectors = {}
for word in graph.G.nodes():
self.vectors[word] = word2vec.wv[word]
del word2vec