This is a open source TensorFlow implementation for "Multi-Granularity Hierarchical Attention Fusion Networks for Reading Comprehension and Question Answering" paper by Alibaba for Squad dataset.

1. To train the network run the following command python main.py --experiment_name=baseline --mode=train
2. For inference run the following command python main.py --experiment_name=baseline --mode=show_examples
3. For running in Manual question and answer mode run the following command python main.py --experiment_name=baseline --mode=compare

Note: this code is adapted in part from the Final Project (SQuAD) for CS224n, Winter 2018