CAP 4601 Project 2

This project uses natural language processing to create a knowledge database from unstructured text.

Try out this model in Google Colab

Dev setup

This project requires a hell of dependencies. Mostly due to various deep learning frameworks used in the PURE model. There is a docker container provided to deal with those issues.

Install all required packages with

pip install -r requirementsGPU.txt

Update the pre-commit hooks

pre-commit install

for proper code styling. Check style with

python -m isort .
python -m black .
python -m flake8

Train the models

Run

python -m src.semevalModel.trainEntity
python -m src.semevalModel.trainRelation

Run main

Adjust the paths for the models directory. Run

python -m src.cap_4601_project_2.main