The RCZoo project is a toolkit for reading comprehension model. It contains the PyTorch reimplement of multiple reading comprehension models
- run
sh download.shto download the dataset and the glove embeddings. - run
sh runs/train_squad.sh [bidaf|drqa|slqa|fusionnet|docqa]to start the train process. (Check the xxx.sh scripts before run, as the preprocessing only need to be executed once)
python 3.5
Pytorch 0.4
tqdm
We train each model on train set for 40 epoch, and report the best performance on dev set.
| Model | Exact Match | F1 | EM(+ELMo) | F1(+ELMo) |
|---|---|---|---|---|
| Rnet | 69.25 | 78.97 | ||
| BiDAF | 70.47 | 79.90 | 73.04 | 81.48 |
| documentqa | 71.47 | 80.84 | ||
| DrQA | 68.39 | 77.90 | ||
| QAnet | ... | ... | ||
| SLQA | 67.09 | 76.67 | ||
| FusionNet | 68.27 | 77.79 |
some code are borrowed from DrQA, a cool project about reading comprehension.