/Pointer-Generator

Tutorial: Using pytorch implementation of Get To The Point: Summarization with Pointer-Generator Networks for easier to understand the algorithms

Primary LanguagePythonMIT LicenseMIT

Point to Generator

Using pytorch implementation of Get To The Point: Summarization with Pointer-Generator Networks

Enabled

1. Point Network
2. Coverage

Required

python 3.5+ (anaconda)
pytorch

Train

python main.py

e.g.
Epoch [2/10000], Step [0/1], Loss: 548.0956, Coverage Loss: 2.000000
Test Accuracy of the model on the test: 1.0 %

Examples

近期，联想投票门事件成为舆论焦点日，联想控股董事长联想集团创始人柳传志两次针对此事件表态，称这是真正有策划有布置，动机极为恶劣的阴谋与此同时，更多的民营企业家也站出来

License

MIT © Sohone Guo