README.md

Code for our NAACL 2019 paper:

Sentence Embedding Alignment for Lifelong Relation Extraction

Paper link: https://arxiv.org/abs/1903.02588

Running the code

First prepare the GloVe embeddings

• download data from https://nlp.stanford.edu/projects/glove/, put glove.6B.300d.txt under data/

1. On Simple Question Dataset

cp config_SimQue.py config.py

python continue_train.py 100

2. On FewRel Dataset

cp config_FewRel.py config.py

python continue_train.py 100