- Python3.7
- virtualenv
Download glove.twitter.27B.zip
from stanfordnlp/GloVe and unzip to .vector_cache
# Setup python virtual environment
$ virtualenv venv --python=python3
$ source venv/bin/activate
# Install python dependencies
$ pip install -r requirements.txt
You may either run the ipynb
jupyter notebook on Colab with GPU acclerator (Recommend)
Or do it locally (DON'T since it will take forever unless you have a CUDA-enabled GPU)
$ python semeval2019_task5_hateval_final.py
You may modify the TrainParameters
class to train classifier for different labels and different weight intilization presets. See comment for detail
- Yonael Bekele
- Michael Lin
- Helen Zhao