This code is based on the code for paper Discrete-State Variational Autoencoders for Joint Discovery and Factorization of Relations by Diego Marcheggiani and Ivan Titov.

Create a python 3.6 environment using pyenv or conda. Then install python packages with pip: pip install -r requirements.txt

Extract data to ./data directory.

• Baseline model: python -m main oie
• RegDVAE:
  • First get the KB embeddings: python -m main ext_kb -out model/ext_kb/m_001
  • python -m main oie_reg -lekd m001

• lexicalized dependency path between arguments (entities) of the relation,
• first entity
• second entity
• entity types of the first and second entity
• trigger word
• id of the sentence
• raw sentence
• pos tags of the entire sentence
• relation between the two entities if any (used only for evaluation)

• python3.6
• numpy>=1.4
• scipy
• nltk
• tensorflow>=r1.7
• recordclass
• pandas