Zhongdao gcn_clustering issues

Zhongdao / gcn_clustering

Code for CVPR'19 paper Linkage-based Face Clustering via GCN

MIT License

358 stars 86 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

why first iteration not use thershold in test code

#45 yongqis opened 11 months ago
1
knn上界已经复现了，但我认为graph_propagation的max_sz不太合理，完全可以改成th逐渐递减的模式去划分clusters

#44 FDuiv closed 3 years ago
1
bcubed函数里面，根据我的验证，precision和recall两个变量可能搞反了

#43 FDuiv opened 3 years ago
0
face images of features

#42 mensaochun opened 3 years ago
0
IJB-B list

#41 houguanqun opened 3 years ago
0
关于原始数据集

#40 StalkerDAs opened 3 years ago
0
Using code in the new Gpu-s(Python Versions)

#39 ghost opened 3 years ago
0
Update test.py

#38 espectre closed 3 years ago
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 fenglv12345 opened 3 years ago
0
About the dataset

#36 KimKeepLearning opened 3 years ago
0
无监督还是有监督？

#35 littlejiajia opened 4 years ago
3
关于labels

#34 csqmaurine opened 4 years ago
0
Extracting features from ArcFace

#33 saedr opened 4 years ago
2
about the features.zip

#32 KosukeHao opened 4 years ago
3
请问为什么训练的时候没有softmax，而测试的时候使用

#31 KimKeepLearning closed 3 years ago
0
关于特征提取

#30 zhufq00 opened 4 years ago
1
我使用新数据集时，每个batch正样本比例只有1%左右，我使用resnet50提取特征，请问这有什么解决办法吗

#29 zhufq00 closed 4 years ago
0
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 zhufq00 closed 4 years ago
1
想请问下，我没有改动代码任何地方，但是训练2个batch后precison和recall就变为0了,请问该如何训练出best.ckpt的效果呢

#27 zhufq00 closed 4 years ago
0
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 Emma0118 opened 4 years ago
1
Can the author provide the data preprocessing script

#25 flowerinheart opened 4 years ago
0
Images of features.zip

#24 ZPdesu closed 4 years ago
1
运算矩阵大小

#23 ltm920716 closed 4 years ago
0
the idea about GCN

#22 ltm920716 opened 4 years ago
1
is this a clustering model or classification model？ why the code "self.labels = np.load(label_path)" is in feeder.py

#21 fusionchan opened 4 years ago
0
What is the ideal number of neighbour for any dataset?

#20 abhiksark opened 4 years ago
0
关于邻接矩阵的变换

#19 yingjianling opened 4 years ago
0
关于idea

#18 BluebirdStory closed 5 years ago
1
测试的一个小 bug

#17 ludics closed 5 years ago
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 Weizhongjin opened 5 years ago
0
关于Pression和Recall的问题，代码中应当是计算反了

#15 CaiQiuYu closed 5 years ago
1
About acknowledgement

#14 XiaohangZhan closed 5 years ago
1
索引越界问题！

#13 xxx2974 opened 5 years ago
0
关于梯度回传

#12 lycwgf opened 5 years ago
7
模型测试效果不理想

#11 lichanglin3445667 opened 5 years ago
19
某些batch速度比较慢，有没有类似问题的啊？

#10 CaiQiuYu opened 5 years ago
8
about training error.

#9 xxx2974 opened 5 years ago
0
About features of CASIA

#8 LZDSJTU opened 5 years ago
14
predicting edges are all zeros during training

#7 jxyecn opened 5 years ago
2
pred will contain nan if k_at_hop is large

#6 luzai opened 5 years ago
4
How to generate "attention aggregation" feature?

#5 jxyecn closed 5 years ago
3
Why use the random subset of CASIA?

#4 LZDSJTU closed 5 years ago
1
features.zip

#3 xxx2974 opened 5 years ago
3
how to generate the KNN graph？

#2 whxhngns200 closed 5 years ago
4
Could u pls provide the data preprocessing code?

#1 abcdvzz closed 5 years ago
1