jforjohn/drug_embeddings

The corpus is from the SemEval-2013 Task 9 and here we want to try different embeddings for identifying drugs (NER use case).

                __________
         ______/ ________ \______
       _/      ____________      \_
     _/____________    ____________\_
    /  ___________ \  / ___________  \
   /  /XXXXXXXXXXX\ \/ /XXXXXXXXXXX\  \
  /  /############/    \############\  \
  |  \XXXXXXXXXXX/ _  _ \XXXXXXXXXXX/  |
__|\_____   ___   //  \\   ___   _____/|__
[_       \     \  X    X  /     /       _]
__|     \ \                    / /     |__
[____  \ \ \   ____________   / / /  ____]
     \  \ \ \/||.||.||.||.||\/ / /  /
      \_ \ \  ||.||.||.||.||  / / _/
        \ \   ||.||.||.||.||   / /
         \_   ||_||_||_||_||   _/
           \     ........     /
            \________________/

This project is about training an embedding or use pretrained ones for a Named Entity Recognition task. As an example corpus is used the one from SemEval 2013 task9.

In the repository you need to specify the parameters that you want in the config.json file and run the run.py file in the src directory to run the experiments.