Scikitlearn:用于聚类评估的ARI分数

2024-03-29 02:16:57 发布

您现在位置:Python中文网/ 问答频道 /正文

我正在计算evaluating the cluster performance的调整后兰德指数分数。假设真实集群和预测集群如下所示。格式{i, "x"}表示元素"x"ith集群中

>>> labels_true = [{0,"a"}, {0,"b"}, {0,"c"}, {1,"d"}, {1,"e"}, {1,"f"}]
>>> labels_pred = [{0,"a"}, {0,"b"}, {1,"c"}, {1,"d"}, {2,"e"}, {2,"f"}]
>>> metrics.adjusted_rand_score(labels_true, labels_pred)

ARI得分为1.0,但似乎不应为1.0,因为预测的聚类与真实聚类不同

我想知道这是否是计算ARI分数的有效方法


Tags: thetrue元素labels格式performance集群聚类