intfloat/SimKGC

Issues

• 关于模型性能极度提升的问题

  #47 opened 3 months ago by hcfun
  0

• I need the response about eval-model-path. Thanks！

  #45 opened 3 months ago by yuhuixie683
  0

• Hi, I would like to confirm exactly what part of the code you are describing in your experimental procedure for conducting complex relationships? Or can you tell me which part of the code you are referring to?

  #44 opened 5 months ago by wangkaibo12138
  0

• CUDA out of memory

  #43 opened 6 months ago by lijun-1999
  2

• Can't run the program

  #27 opened a year ago by dqs12345
  3

• question about contrastive learning

  #42 opened 7 months ago by yw3l
  0

• Re-Ranking On FB15K237

  #41 opened 7 months ago by yw3l
  0

• Pre-trained Language Model with bert-base-uncased and bert-large-uncased.

  #37 opened 9 months ago by yw3l
  0

• How to use other type of data

  #40 opened 10 months ago by PoisonousBromineChan
  0

• How to use model file locally instead of linking to Huggingface? How to modify the code?

  #39 opened 10 months ago by PoisonousBromineChan
  2

• Pretrained Language Models (like BERT) with embedding-based models (like TransE)

  #2 opened 3 years ago by sm354
  5

• Regarding the model: The first encoder is used to calculate the relationship aware embedding of the head entity h hh, while the input of the second encoder BERTt only contains the textual description of entity t.

  #38 opened 10 months ago by whistle9
  3

• negative samplings

  #24 opened 2 years ago by Wangchuan0
  5

• Question about cosine similarity

  #36 opened a year ago by yezhyzh
  2

• 您好，我有个问题关于best_mode和last_model?

  #35 opened a year ago by YongqingQI
  2

• Questions about the inverse relations

  #34 opened a year ago by ShaoZhangHao
  1

• Is the reproduction of the machine demanding?

  #33 opened a year ago by SYuan03
  1

• prediction

  #32 opened a year ago by yys-1
  2

• predictions

  #31 opened a year ago by yys-1
  0

• the number of testing targets

  #30 opened a year ago by harlen1212
  0

• Additional description information during testing(inference)

  #29 opened a year ago by meaningful96
  3

• How to use a trained model to obtain entity embeddings

  #28 opened a year ago by xga0
  1

• 为什么在训练过程中标签是生成的，而不是使用真实的标签

  #26 opened a year ago by YongqingQI
  1

• 您好，这两行代码不是很明白，可以请教一下吗？

  #25 opened a year ago by YongqingQI
  2

• Negative sampling

  #21 opened 2 years ago by whistle9
  6

• Hello, when applying re-ranking, is the loss corrected?

  #23 opened 2 years ago by GitHubBoys
  4

• Questions about the loss function

  #22 opened 2 years ago by handsomelys
  9

• About Hits@K

  #19 opened 2 years ago by EurusNotes
  2

• 5.5 Entity Visualization

  #20 opened 2 years ago by 789456123fcf
  2

• Question about the mask in code

  #18 opened 2 years ago by handsomelys
  2

• Questions about metrics

  #17 opened 2 years ago by handsomelys
  2

• 请问支持中文吗？

  #16 opened 2 years ago by WatsonWangZh
  1

• Wikidata5M失效问题

  #15 opened 2 years ago by cyjie429
  3

• How to get(save) the tensor of the entity (or hr) with the trained SimKGC model?

  #14 opened 2 years ago by EurusNotes
  3

• 关于关闭 Re-ranking 的设定

  #13 opened 2 years ago by EurusNotes
  3

• The performance about infoNCEloss？

  #12 opened 2 years ago by maoulee
  2

• Code Problem

  #10 opened 2 years ago by qiushuigongchang
  8

• Negative Sample

  #11 opened 2 years ago by qiushuigongchang
  2

• 关于wikidata5m的text 的问题

  #9 opened 2 years ago by EurusNotes
  2

• Negative sampling

  #7 opened 3 years ago by NoviceCrom
  1

• Accuracy of Wiki-trans

  #6 opened 3 years ago by EurusNotes
  2

• How to run the different model variants decribed in the paper?

  #4 opened 3 years ago by lena-schwert
  4

• About PyTorch Ver

  #5 opened 3 years ago by EurusNotes
  2

• Do you use the alias files of Wikidata5M?

  #3 opened 3 years ago by lena-schwert
  3

• Did you try without LM pretrained weights?

  #1 opened 3 years ago by apoorvumang
  5