DeepGraphLearning/KnowledgeGraphEmbedding

Issues

• Three questions

  #55 opened 3 years ago by zulihit
  2

• "IndexError: index out of range in self" when I train on my own knowledge graph. Can anyone help me with this? Thanks a lot!

  #57 opened a year ago by YoannCheung
  0

• Why start = 4

  #56 opened 2 years ago by passermyh
  0

• How to designate dimension of embeddings

  #54 opened 3 years ago by yq-leo
  0

• Memory consumption issue

  #53 opened 3 years ago by p6jain
  1

• GPU Out Of Memory

  #38 opened 4 years ago by ms0510k
  2

• L3 regularization

  #52 opened 3 years ago by yancy826
  0

• CUDA out of memory (resolved) and method to make RotatE run faster

  #51 opened 3 years ago by Thuy-g
  2

• Init embedding random OR by bert encode from entity description text

  #50 opened 3 years ago by fridayL
  1

• regarding RELATION CATEGORY in FB15k

  #48 opened 3 years ago by seldce
  1

• confused about implementation of 1-1, 1-n, n-1, n-n

  #47 opened 3 years ago by junkangwu
  1

• Regarding the direction of the rotation

  #49 opened 3 years ago by seldce
  1

• hi，a priblem about multi GPU training

  #46 opened 4 years ago by Zhangs996
  1

• N-1 and 1-N relations

  #45 opened 4 years ago by Walker-ZD
  3

• How to split the train/valid/test

  #43 opened 4 years ago by SRL94
  1

• how do you generate the embedding? Is it a random walk or something? I mean what is the main algorithm for embedding.

  #42 opened 4 years ago by programmingboy
  1

• confused about implementation of DistMult

  #41 opened 4 years ago by junkangwu
  1

• When I train a more huge dataset by model RotatE, always show the question "CUDA out of memory" ?

  #44 opened 4 years ago by Dionysusssss
  1

• RuntimeError: The size of tensor a (2000) must match the size of tensor b (1000) at non-singleton dimension 2

  #40 opened 4 years ago by jweihe
  3

• Pretrained models

  #39 opened 4 years ago by Demirrr
  0

• How does RotatE deal with anti-symmetric relations?

  #37 opened 4 years ago by empty-id
  0

• Why do you separate negative head samples and negative tail samples?

  #34 opened 4 years ago by renli1024
  5

• Interpreting how does RotatE handle 1-N, N-N type relations...

  #36 opened 4 years ago by agoel00
  1

• A question about the data range of negative sampling

  #35 opened 4 years ago by renli1024
  1

• Can we initialize entity embeddings using GLoVE embeddings?

  #33 opened 4 years ago by kiranramnath007
  1

• Code early stop

  #32 opened 4 years ago by ToneLi
  0

• the early stop when setting the max step

  #31 opened 4 years ago by ToneLi
  0

• Hi,I want to

  #27 opened 5 years ago by Aristomd
  2

• Did you just use the first batch to train the model? Can you help me solve my problem?

  #29 opened 4 years ago by ToneLi
  2

• train error

  #28 opened 4 years ago by as3896001
  6

• Reproduce DistMult results

  #18 opened 5 years ago by jinxing94
  1

• Loss function of TransE and RotatE in the code

  #26 opened 5 years ago by ngl567
  2

• Can't find log files

  #25 opened 5 years ago by p6jain
  3

• why the ComplEx just get 0.74 Hits10 on FB15k which is lower than paper's 0.84 for 0.1

  #20 opened 5 years ago by WQQQQQQQQQQ
  1

• invalid syntax error

  #24 opened 5 years ago by Bishnukuet
  3

• Issue of the evaluation results on WN18RR

  #23 opened 5 years ago by ngl567
  1

• RuntimeError: CUDA out of memory.

  #22 opened 5 years ago by DonnieZhang586
  1

• About the seed setting.

  #21 opened 5 years ago by Zawo1995
  1

• why complEx

  #19 opened 5 years ago by WQQQQQQQQQQ
  0

• How to use embedding to compute triplet score

  #17 opened 5 years ago by zhongpeixiang
  1

• Compute Head and Tail Prediction on FB15K dataset

  #16 opened 5 years ago by MuhibKhan
  6

• Some question about the test process.

  #15 opened 5 years ago by zysforever
  2

• RotatE: this implementation vs GraphVite

  #14 opened 5 years ago by AndRossi
  2

• How to get the embeddings file of all the entities and relations?

  #13 opened 5 years ago by Shengqiang-Zhang
  1

• Question about relaxing RotatE

  #12 opened 5 years ago by Chen-Cai-OSU
  1

• Given two node embedding, how to predict the relation embedding between them?

  #11 opened 5 years ago by aqzheng
  1

• Scoring functions and Adversarial Loss Parameters

  #10 opened 5 years ago by sumitpai
  1

• Question about the embedding range (weight initialization and phase normalization)

  #9 opened 6 years ago by tabacof
  1

• Script for finding the best hyperparameters

  #8 opened 6 years ago by dschaehi
  6

• The dimension of entity vector and relation vector

  #7 opened 6 years ago by Joker-Song
  1