QANet-pytorch

Introduction

An implementation of QANet with PyTorch.
refer to hengruo/QANet-pytorch
Any contributions are welcome!

cowork-rules

• branch name: <type>/<brief>

  type: feature, bug, etc.
  brief: brief content

  e.g., feature/init

• commit message: <[type]><description>

  type: FEATURE, BUG, CLEANUP, etc.
  description: description the changes in this commit

  e.g., [FEATURE]init

Usage

1. Install pytorch 1.0+ for Python 3.6+
2. Run pip install -r requirements.txt to install python dependencies.
3. Run download.sh to download the dataset.
4. Start Coding

Todo

• typing lint
• tuning
• squad2.0
• augmentation
• In residual network, add raw value first or dropout first