Pointer-Generator-Pytorch

About

The pytorch implementation of Get To The Point: Summarization with Pointer-Generator Networks. This implementation also tries to use Transformer as the encoder. The project are heavily borrowed from atulkum-pointer_summarizer and jadore801120-attention-is-all-you-need-pytorch.

Requirements

• python==3.7.4
• pytorch==1.4.0
• pyrouge==0.1.3
• tensorflow>=1.13.1

Quick start

• The path and parameters of project: you might need to change some path and parameters in utils/config.py according your setup.
• Dataset: you can download the CNN/DailyMail dataset from https://github.com/JafferWilson/Process-Data-of-CNN-DailyMail, then run make-datafiles.py to process data. For the specific process, you can refer to https://github.com/abisee/cnn-dailymail.
• Run: you can run train.py, eval.py, and test for training, evaluating, and test, respectively.

Note:

• There is only single example repeated across the batch in the decode mode of beam search.