yxgeee / SpCL

[NeurIPS-2020] Self-paced Contrastive Learning with Hybrid Memory for Domain Adaptive Object Re-ID.
https://yxgeee.github.io/projects/spcl
MIT License
318 stars 67 forks source link

聚类紧密度计算 #40

Closed colinzhaoxp closed 3 years ago

colinzhaoxp commented 3 years ago

作者您好,感谢您卓越工作的分享! 在看代码时,这里有一些疑惑,为何每个类的紧密度要取min值呢?这里面的值难道不都一样吗? cluster_R_comp = [min(cluster_R_comp[i]) for i in sorted(cluster_R_comp.keys())]

colinzhaoxp commented 3 years ago

想了好久没想明白,结果今天提交完issue就想明白了:)