- PyTorch implementation of the paper QANET: Combining Local Convolution with Global Self-Attention for Reading Comprehension by Adams Wei Yu, David Dohan, Minh-Thang Luong, Rui Zhao, Kai Chen, Mohammad Norouzi, Quoc V. Le
Here are our 📚 Notes for the implementation
git clone https://github.com/abhirajtiwari/SQuAD2.git
pip3 install -r requirements.txt
- This downloads SQuAD 2.0 training and dev sets, as well as the GloVe 300-dimensional word vectors (840B)
- This also pre-processes the dataset for efficient data loading
python3 setup.py
CLI args training args.py
python3 train.py -n train_run_1 --num_workers 4 --num_epochs 30 --eval_steps 50000 --batch_size 64 --hidden_size 128
To load the tensorboard
tensorboard --logdir save
If you find QANet useful in your research, please consider citing:
@article{DBLP:journals/corr/abs-1804-09541,
author = {Adams Wei Yu and
David Dohan and
Minh{-}Thang Luong and
Rui Zhao and
Kai Chen and
Mohammad Norouzi and
Quoc V. Le},
title = {QANet: Combining Local Convolution with Global Self-Attention for
Reading Comprehension},
journal = {CoRR},
volume = {abs/1804.09541},
year = {2018},
url = {http://arxiv.org/abs/1804.09541},
archivePrefix = {arXiv},
eprint = {1804.09541},
timestamp = {Mon, 13 Aug 2018 16:48:18 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-1804-09541.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
If you liked our work, please leave a 🌟