This model implements the work in the following paper:

Jonathan Raiman and John Miller. Globally Normalized Reader. Empirical Methods in Natural Language Processing (EMNLP), 2017.

If you use the dataset/code in your research, please cite the above paper:

@inproceedings{raiman2015gnr,
    author={Raiman, Jonathan and Miller, John},
    booktitle={Empirical Methods in Natural Language Processing (EMNLP)},
    title={Globally Normalized Reader},
    year={2017},
}

You can also visit https://github.com/baidu-research/GloballyNormalizedReader to get more information.

1. Please use docker image to install the latest PaddlePaddle, by running:

   docker pull paddledev/paddle

2. Download all necessary data by running:

   cd data && ./download.sh && cd ..

3. Preprocess and featurizer data:

   python featurize.py --datadir data --outdir data/featurized  --glove-path data/glove.840B.300d.txt

• Configurate the model by modifying config.py if needed, and then run:

  python train.py 2>&1 | tee train.log

• Infer by a trained model by running:

  python infer.py \
    --model_path models/pass_00000.tar.gz \
    --data_dir data/featurized/ \
    --batch_size 2 \
    --use_gpu 0 \
    --trainer_count 1 \
    2>&1 | tee infer.log