Skip to content

zhupengjia/simple_kbqa

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

simple_kbqa

A simple kbqa based on knowledge graph embedding

node and relation extracted via keyword based entity recognition multi relations extracted via part of speech with spacy graph search use pretrained graph embedding via rotatE model

  • Usage:

    • Train knowledge graph embedding

      git clone [email protected]:zhupengjia/KnowledgeGraphEmbedding.git
      cd KnowledgeGraphEmbedding/codes
      ./run2.py --data_path=$DATA_PATH --model=RotatE --save_path=$SAVE_PATH --double_entity_embedding --cuda --hidden_dim=50
    • Run

      ./interact.py --checkpoint=$CHECKPOINTFILE --w2v_word2idx=$WORDEMBEDDINGLOOKUPFILE --w2v_idx2vec=$WORDEMBEDDINGWEIGHTFILE --backend restful

      word2vec lookup and weight file can be generated from script in example directory:

      cd example
      ./bert_to_wordvec.py

About

A simple kbqa based on knowledge graph embedding

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published