THUDM/GraphMAE

Issues

• 无法运行代码

  #66 opened 6 months ago by YuantingMaSC
  0

• 请问您的CUDA版本是多少啊

  #65 opened 6 months ago by jweden33
  0

• Results on graph classification NCI1 dataset

  #64 opened 7 months ago by AliothF
  0

• Graph Classification Accuracy Calculation?

  #63 opened 7 months ago by gdreiman-insitro
  0

• 图1(b)

  #62 opened 8 months ago by 1090h2400
  1

• 关于NCI1的结果

  #42 opened a year ago by songshuhan
  3

• Node embedding with feature reconstruction

  #61 opened 9 months ago by jpfn2407
  0

• transfer learning on the clintox dataset

  #60 opened 10 months ago by shoyua
  0

• Clarification on Inductive Setting in Graph Neural Networks

  #59 opened 10 months ago by SangwookBaek
  0

• how to customize dataset and use pretrain model

  #58 opened 10 months ago by wml1993
  0

• 重构Loss设计

  #57 opened a year ago by StefanIsSmart
  1

• Accuracy calculation problem

  #56 opened a year ago by openchao
  0

• closed

  #55 opened a year ago
  0

• Question about dataset of transfer learning

  #54 opened a year ago by StylesZhang
  1

• how to set hyperparameters（e.g. num_hidden, num_heads）

  #53 opened a year ago by Lipxiao
  5

• How to use for graph classification

  #52 opened a year ago by PeterDeSOM
  1

• 你好，请问你们遇见过这个问题吗

  #51 opened a year ago by youshengithub
  1

• inductive node classification

  #50 opened a year ago by ghjksnsi
  1

• link prediction

  #49 opened a year ago by Lipxiao
  1

• Find implement issue in PyG version

  #47 opened a year ago by fs302
  1

• Great work! Have you tried heterogeneous graphs?

  #48 opened a year ago by squareRoot3
  1

• Cannot Reproduce Reported Results with Code in PyG Branch

  #45 opened a year ago by RocccYan
  1

• 关于VGAE用于节点分类问题

  #44 opened a year ago by songshuhan
  1

• Code about Reddit Dataset

  #43 opened a year ago by yolandalalala
  1

• 关于使用gcn作为encoder的问题

  #41 opened a year ago by zhenshuozhang
  4

• Graph Dataloader

  #19 opened 2 years ago by DuanhaoranCC
  1

• reproduction issue

  #46 opened a year ago by AaronNing
  1

• 实验结果的一些问题

  #40 opened a year ago by ZewLin
  3

• 想咨询为什么把损失函数换成mse，loss一直是0，为什么普通的mse就不能做这个重建任务呢？还有就是如果使用GCN做decoder的效果也只有65%左右

  #38 opened a year ago by songshuhan
  13

• 你好，我运行以下命令，无法复现论文的结果，能帮我解释下为什么吗？非常感谢

  #39 opened a year ago by zzzsh2000
  1

• GraphMAE for unsupervised learning

  #37 opened a year ago by Hesham-Aliy
  1

• I cannot run this method based on the current torch_geometric version (2.2.0)

  #33 opened 2 years ago by HelloWorldLTY
  6

• A question about node classification

  #36 opened 2 years ago by jweihe
  1

• About random seed settings

  #32 opened 2 years ago by AzureLeon1
  1

• self.enc_mask_token = nn.Parameter(torch.zeros(1, in_dim))

  #34 opened 2 years ago by songshuhan
  7

• I have a question

  #29 opened 2 years ago by RocZKUNP
  3

• keyword type error in GraphMAE torch_geometric

  #28 opened 2 years ago by voladorlu
  2

• Question about transfer learning time cost.

  #35 opened 2 years ago by Data-reindeer
  1

• Pls provide the detailed hyperparameters

  #31 opened 2 years ago by Newiz430
  1

• Dataset split is not the standard, like in GCN.

  #27 opened 2 years ago by rllm-project
  3

• about the CCA-SSG implementation

  #30 opened 2 years ago by MR9812
  2

• 有能调用的API让我们调用算法吗？

  #26 opened 2 years ago by YangxinCs
  1

• when I download the dataset of IMDB-BINARY,allways failed.So I download local and put it in the package,it still download...and faild

  #22 opened 2 years ago by danfeng-1
  3

• RuntimeError: Class values must be smaller than num_classes.

  #23 opened 2 years ago by Mah-Zam
  5

• The setting of loss_fn

  #15 opened 2 years ago by stuzhangqiqi
  3

• Encoding Mask Token

  #21 opened 2 years ago by cuongth95
  4

• Transfer Learning

  #20 opened 2 years ago by DuanhaoranCC
  1

• The inconsistent performance on COLLAB dataset

  #16 opened 2 years ago by ZhuYun97
  7

• 请问能否提供一下损失函数为mse时的参数设置？

  #17 opened 2 years ago by yifanQi98
  2

• The hyper-parameters of transfer learning

  #14 opened 2 years ago by jiaruHithub
  5