# About this Poject
reimplement TransE Mode
The original paper is [Translating Embeddings for Modeling Multi-relational Data](http://papers.nips.cc/paper/5071-translating-embeddings-for-modeling-multi-rela "With a Title")
# The Test Result
Dataset ：WN18
raw entity hits@10: 0.651
raw entity mean_rank: 284
filter entity hits@10: 0.728
filter entity mean_rank: 278
FB15K
raw entity hits@10: 0.332
raw entity mean_rank: 389
filter entity hits@10: 0.458
filter entity mean_rank: 297
基于TransE构建多关系知识图谱模型（代码）.rar
需积分: 3 85 浏览量
2024-02-29
16:49:05
上传
评论
收藏 18MB RAR 举报
基于TransE构建多关系知识图谱模型（代码）.rar （27个子文件） 
基于TransE构建多关系知识图谱模型（代码） WN18 
relation2id.txt 360B entity2id.txt 589KB train_ids.txt 1.76MB wordnet-mlj12-test.txt 165KB wordnet-mlj12-train.txt 4.54MB wordnet-mlj12-valid.txt 165KB .idea 
other.xml 233B vcs.xml 180B TransE.iml 576B misc.xml 288B inspectionProfiles profiles_settings.xml 174B modules.xml 264B .gitignore 38B FB15k freebase_mtr100_mte100-valid.txt 3.72MB freebase_mtr100_mte100-test.txt 4.39MB valid.txt 3.72MB relation2id.txt 88KB entity2id.txt 220KB test.txt 4.39MB freebase_mtr100_mte100-train.txt 35.86MB train.txt 35.86MB transE.py 10KB TransE_pytoch.py 24KB test.py 12KB entity_vector_50dim 15.41MB README.md 510B relation_vector_50dim 1.38MB资源评论
我悟了-
- 粉丝: 1907
- 资源: 133
