Zhongdao/gcn_clustering
Code for CVPR'19 paper Linkage-based Face Clustering via GCN
Jupyter NotebookMIT
Issues
- 1
- 3
about the features.zip
#32 opened by KosukeHao - 1
计算边分数时, score_dict[e[0], e[1]] = 0.5*(score_dict[e[0], e[1]] + score[i]),这样会导致后遍历到的边分数权重更大,为什么不用真正的平均呢
#28 opened by zhufq00 - 1
- 3
无监督还是有监督?
#35 opened by littlejiajia - 0
bcubed函数里面,根据我的验证,precision和recall两个变量可能搞反了
#43 opened by FDuiv - 19
模型测试效果不理想
#11 opened by lichanglin3445667 - 0
face images of features
#42 opened by mensaochun - 2
Extracting features from ArcFace
#33 opened by saedr - 1
Images of features.zip
#24 opened by ZPdesu - 0
IJB-B list
#41 opened by houguanqun - 0
关于原始数据集
#40 opened by StalkerDAs - 0
- 0
usinng mean aggregation can not satisfy the requriment that G = g(X, A) is an aggregation matrix of size N × N and each row is summed up to 1
#37 opened by fenglv12345 - 1
- 0
About the dataset
#36 opened by KimKeepLearning - 0
请问为什么训练的时候没有softmax,而测试的时候使用
#31 opened by KimKeepLearning - 0
关于labels
#34 opened by csqmaurine - 3
features.zip
#3 opened by xxx2974 - 7
- 1
请问,GCN 如何使用 batch 训练?这里的batch 是指的 对 node的 分批训练吗?
#26 opened by Emma0118 - 0
- 0
- 8
某些batch速度比较慢,有没有类似问题的啊?
#10 opened by CaiQiuYu - 0
- 0
- 14
About features of CASIA
#8 opened by LZDSJTU - 0
- 1
the idea about GCN
#22 opened by ltm920716 - 0
is this a clustering model or classification model? why the code "self.labels = np.load(label_path)" is in feeder.py
#21 opened by fusionchan - 4
how to generate the KNN graph?
#2 opened by whxhngns200 - 0
- 0
关于邻接矩阵的变换
#19 opened by yingjianling - 1
关于idea
#18 opened by BluebirdStory - 4
pred will contain nan if k_at_hop is large
#6 opened by luzai - 0
测试的一个小 bug
#17 opened by ludics - 0
Ms-Celeb-1M的测试结果是多少,不知道作者有单独测试过吗
#16 opened by Weizhongjin - 1
关于Pression和Recall的问题,代码中应当是计算反了
#15 opened by CaiQiuYu - 1
About acknowledgement
#14 opened by XiaohangZhan - 2
predicting edges are all zeros during training
#7 opened by jxyecn - 0
about training error.
#9 opened by xxx2974 - 3
How to generate "attention aggregation" feature?
#5 opened by jxyecn - 1
Why use the random subset of CASIA?
#4 opened by LZDSJTU - 1