Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / zhongdao/gcn_clustering issues and pull requests
#45 - why first iteration not use thershold in test code
Issue -
State: open - Opened by yongqis over 1 year ago
- 1 comment
#44 - 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
Issue -
State: closed - Opened by FDuiv almost 4 years ago
- 1 comment
#43 - bcubed函数里面,根据我的验证,precision和recall两个变量可能搞反了
Issue -
State: open - Opened by FDuiv almost 4 years ago
#42 - face images of features
Issue -
State: open - Opened by mensaochun about 4 years ago
#41 - IJB-B list
Issue -
State: open - Opened by houguanqun about 4 years ago
#40 - 关于原始数据集
Issue -
State: open - Opened by StalkerDAs about 4 years ago
#39 - Using code in the new Gpu-s(Python Versions)
Issue -
State: open - Opened by ghost over 4 years ago
#38 - Update test.py
Pull Request -
State: closed - Opened by espectre over 4 years ago
#37 - 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
Issue -
State: open - Opened by fenglv12345 over 4 years ago
#36 - About the dataset
Issue -
State: open - Opened by KimKeepLearning over 4 years ago
#35 - 无监督还是有监督?
Issue -
State: open - Opened by littlejiajia over 4 years ago
- 3 comments
#34 - 关于labels
Issue -
State: open - Opened by csqmaurine over 4 years ago
#34 - 关于labels
Issue -
State: open - Opened by csqmaurine over 4 years ago
#33 - Extracting features from ArcFace
Issue -
State: open - Opened by saedr almost 5 years ago
- 2 comments
#32 - about the features.zip
Issue -
State: open - Opened by KosukeHao almost 5 years ago
- 3 comments
#31 - 请问为什么训练的时候没有softmax,而测试的时候使用
Issue -
State: closed - Opened by KimKeepLearning almost 5 years ago
#31 - 请问为什么训练的时候没有softmax,而测试的时候使用
Issue -
State: closed - Opened by KimKeepLearning almost 5 years ago
#30 - 关于特征提取
Issue -
State: open - Opened by zhufq00 almost 5 years ago
- 1 comment
#29 - 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
Issue -
State: closed - Opened by zhufq00 about 5 years ago
#28 - 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
Issue -
State: closed - Opened by zhufq00 about 5 years ago
- 1 comment
#27 - 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
Issue -
State: closed - Opened by zhufq00 about 5 years ago
#26 - 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?
Issue -
State: open - Opened by Emma0118 about 5 years ago
- 1 comment
#25 - Can the author provide the data preprocessing script
Issue -
State: open - Opened by flowerinheart about 5 years ago
#24 - Images of features.zip
Issue -
State: closed - Opened by ZPdesu about 5 years ago
- 1 comment
#23 - 运算矩阵大小
Issue -
State: closed - Opened by ltm920716 about 5 years ago
#22 - the idea about GCN
Issue -
State: open - Opened by ltm920716 about 5 years ago
- 1 comment
#21 - is this a clustering model or classification model? why the code "self.labels = np.load(label_path)" is in feeder.py
Issue -
State: open - Opened by fusionchan over 5 years ago
#20 - What is the ideal number of neighbour for any dataset?
Issue -
State: open - Opened by abhiksark over 5 years ago
#19 - 关于邻接矩阵的变换
Issue -
State: open - Opened by yingjianling over 5 years ago
#18 - 关于idea
Issue -
State: closed - Opened by BluebirdStory over 5 years ago
- 1 comment
#17 - 测试的一个小 bug
Issue -
State: closed - Opened by ludics over 5 years ago
#16 - Ms-Celeb-1M的测试结果是多少,不知道作者有单独测试过吗
Issue -
State: open - Opened by Weizhongjin over 5 years ago
#15 - 关于Pression和Recall的问题,代码中应当是计算反了
Issue -
State: closed - Opened by CaiQiuYu over 5 years ago
- 1 comment
#14 - About acknowledgement
Issue -
State: closed - Opened by XiaohangZhan over 5 years ago
- 1 comment
#13 - 索引越界问题!
Issue -
State: open - Opened by xxx2974 over 5 years ago
#12 - 关于梯度回传
Issue -
State: open - Opened by lycwgf over 5 years ago
- 7 comments
#11 - 模型测试效果不理想
Issue -
State: open - Opened by lichanglin3445667 over 5 years ago
- 19 comments
#10 - 某些batch速度比较慢,有没有类似问题的啊?
Issue -
State: open - Opened by CaiQiuYu over 5 years ago
- 8 comments
#9 - about training error.
Issue -
State: open - Opened by xxx2974 over 5 years ago
#8 - About features of CASIA
Issue -
State: open - Opened by LZDSJTU over 5 years ago
- 14 comments
#7 - predicting edges are all zeros during training
Issue -
State: open - Opened by jxyecn over 5 years ago
- 2 comments
#6 - pred will contain nan if k_at_hop is large
Issue -
State: open - Opened by luzai over 5 years ago
- 4 comments
#5 - How to generate "attention aggregation" feature?
Issue -
State: closed - Opened by jxyecn over 5 years ago
- 3 comments
#4 - Why use the random subset of CASIA?
Issue -
State: closed - Opened by LZDSJTU over 5 years ago
- 1 comment
#3 - features.zip
Issue -
State: open - Opened by xxx2974 almost 6 years ago
- 3 comments
#2 - how to generate the KNN graph?
Issue -
State: closed - Opened by whxhngns200 almost 6 years ago
- 4 comments
#1 - Could u pls provide the data preprocessing code?
Issue -
State: closed - Opened by abcdvzz almost 6 years ago
- 1 comment