sna_clustering.py 文件源码

def create_graph(self):
        path = ""
        data = pd.read_csv(path + '../../worldoss:ocean/Web_Crawler/generated_repo_topic_data.csv', error_bad_lines=False, header=None,
                           sep=",", delimiter='\n')  # pandas ?????? ???? SNA ?? csv ?? ????
        # Creating node list
        node = []
        for i in data.values:
            for j in i[0].split(',')[1:]:
                node.append(j)

        node = list(set(node))

        # Creating edge list
        self.edges = []

        for i in data.values:
            l = i[0].split(',')[1:]
            for j in range(len(l)):
                for k in range(j + 1, len(l)):
                    self.edges.append((l[j], l[k]))

        self.G = nx.Graph()
        self.G.add_nodes_from(node)
        self.G.add_edges_from(self.edges)

        print nx.number_of_nodes(self.G)
        print nx.number_of_edges(self.G)