snowkylin/line

Issues

• if not self.g.has_edge(self.node_index_reversed[negative_node], self.node_index_reversed[edge[0]])， can you get negative samples for second-order ?

  #19 opened 4 years ago by stockholm0101
  0

• Some errors occur

  #18 opened 4 years ago by wangxiaobaobao
  1

• Loss only decrease from 2.2 to 1.8

  #14 opened 4 years ago by yqNLP
  2

• sorry, this is not an issue. just my mistake.

  #17 opened 5 years ago by phucdoitoan
  0

• data replace

  #10 opened 6 years ago by jack0513
  6

• How to get co-authorship_graph.pkl?

  #1 opened 7 years ago by ynuwm
  6

• 数据集

  #15 opened 5 years ago by linyunanit
  1

• can i only maintain the second-order proximity?

  #16 opened 5 years ago by xhran2010
  0

• Run on CPU

  #13 opened 5 years ago by Zafar-southeast
  1

• Hello

  #12 opened 6 years ago by xiongweii
  0

• the program can't run

  #11 opened 6 years ago by heucoder
  1

• second order similarity issue

  #6 opened 6 years ago by yfyang86
  2

• About the embeddings of second-order similarity

  #9 opened 6 years ago by fukien
  2

• run on multiple cores?

  #8 opened 6 years ago by BlaBlaPer
  1

• hi，i am confused about the loss in your code.

  #7 opened 6 years ago by chandlervan
  1

• low efficiency of `tf.matmul`

  #5 opened 6 years ago by wizardforcel
  0

• Question about negative sampling in utils.py

  #4 opened 6 years ago by zhanxiuxiu
  5

• How can i get embedding between 0-1?

  #3 opened 6 years ago by blldd
  4

• loss function different from the paper

  #2 opened 7 years ago by dawnranger
  2