基于networkx可视化处聚类算法的结果

最新推荐文章于 2024-07-14 08:24:00 发布

Hy_home

最新推荐文章于 2024-07-14 08:24:00 发布

阅读量1.6k

点赞数

文章标签：数据可视化可视化

本文链接：https://blog.csdn.net/qq_35440510/article/details/105565917

版权

if __name__ == '__main__':
    import networkx as nx
    import matplotlib.pyplot as plt
    rpath4 = r'E:\community_detcttion\Dataset\football.gml'
    G =nx.read_gml(rpath4,'id')
    adjacency = nx.to_numpy_array(G)
    #将数据集进行处理转换为矩阵
    for i in range(adjacency.shape[0]):
        for j in range(adjacency.shape[1]):
            if i == j and adjacency[i, j] == 1:
                adjacency[i, j] = 0
            else:
                if adjacency[i, j] == adjacency[j, i]:
                    adjacency[j, i] = 0
    #处理完矩阵转换为图
    G2 =nx.from_numpy_matrix(adjacency)
    #算法最后的聚类结果其中列表索引代表节点，索引对应的数值为label,相同label的节点在同一社区（或者说聚类）
    community =[7, 25, 13, 81, 7, 81, 13, 7, 7, 7, 81, 7, 43, 13, 43, 13, 7, 96, 43, 91, 96, 7, 7, 7, 7, 25, 43, 96, 7, 91, 91, 43, 13, 25, 43, 91, 43, 25, 43, 13, 81, 7, 43, 43, 48, 25, 88, 13, 48, 88, 7, 7, 81, 88, 43, 91, 96, 48, 88, 113, 13, 43, 96, 48, 13, 96, 48, 88, 7, 7, 96, 43, 81, 88, 81, 48, 96, 7, 7, 91, 91, 81, 91, 88, 81, 43, 48, 96, 88, 25, 7, 48, 48, 7, 91, 96, 96, 48, 81, 43, 13, 91, 81, 25, 7, 25, 13, 81, 7, 25, 88, 7, 48, 96, 88]
    edge_list =G2.edges()
    pos =nx.spring_layout(G2)
    colors = community
    #先画点根据community来进行颜色的区分
    nx.draw_networkx_nodes(G2,pos,nodelist=G2.nodes(),node_color=colors)
    #再画边
    nx.draw_networkx_edges(G2,pos,nodelist=G2.edges())
    #最后显示点的节点号
    nx.draw_networkx_labels(G2,pos)
    plt.show()