https://github.com/AIKevin/Pointer_Generator_Summarizer
Pointer Generator Network: Seq2Seq with attention, pointing and coverage mechanism for abstractive summarization.
https://github.com/AIKevin/Pointer_Generator_Summarizer
abstractive-summarization artificial-intelligence attention-mechanism deep-learning encoder-decoder natural-language-generation natural-language-processing neural-networks pointer-generator python3 rnn rnn-tensorflow seq2seq summarization tensorflow
Last synced: 16 days ago
JSON representation
Pointer Generator Network: Seq2Seq with attention, pointing and coverage mechanism for abstractive summarization.
- Host: GitHub
- URL: https://github.com/AIKevin/Pointer_Generator_Summarizer
- Owner: AIKevin
- License: other
- Created: 2019-07-25T20:03:07.000Z (about 6 years ago)
- Default Branch: master
- Last Pushed: 2019-07-26T10:52:15.000Z (about 6 years ago)
- Last Synced: 2025-01-13T05:33:11.158Z (9 months ago)
- Topics: abstractive-summarization, artificial-intelligence, attention-mechanism, deep-learning, encoder-decoder, natural-language-generation, natural-language-processing, neural-networks, pointer-generator, python3, rnn, rnn-tensorflow, seq2seq, summarization, tensorflow
- Language: Python
- Homepage: http://www.abigailsee.com/2017/04/16/taming-rnns-for-better-summarization.html
- Size: 50.8 KB
- Stars: 17
- Watchers: 2
- Forks: 8
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- StarryDivineSky - AIKevin/Pointer_Generator_Summarizer
README
Pointer_Generator_Summarizer
The pointer generator is a deep neural network built for abstractive summarizations.
For more informations on this model, you can check out the scientific article here: https://arxiv.org/pdf/1704.04368
You can also see this blog article from the author: http://www.abigailsee.com/2017/04/16/taming-rnns-for-better-summarization.html
With my collaborator Stephane Belemkoabga (https://github.com/steph1793) , we re-made this model in tensorflow for our research project. This neural net will be our baseline model.
We will do some experiments with this model, and propose a new architecture based on this one.
In this project, you can:
- train models
- test *
- evaluate *
* : for the test and evaluation, the main methods are not done yet, but we will release them very soon.
This project reads .bin format files. For our experiments, we will be working on the ccn and dailymail datasets.
You can download the preprocessed files with this link :
https://github.com/JafferWilson/Process-Data-of-CNN-DailyMail
Or do the pre-processing by yourself with this link :
https://github.com/abisee/cnn-dailymail