A web application for extractive summarization based on SummaRuNNer.
Paper: https://adrienguille.github.io/publications/ecir2024.pdf
Model: https://github.com/Baragouine/radsum
git clone https://github.com/Baragouine/radsum_app.git
conda create --name RADSUM_APP python=3.9
conda activate RADSUM_APP
cd radsum_app/back/
pip install -r requirements.txt
conda install pytorch-geometric -c rusty1s -c conda-forge
To install nltk data:
- Open a python console.
- Type
import nltk; nltk.download()
. - Download all data.
- Close the python console.
pip install django
pip install djangorestframework
pip install django-cors-headers
cd back/ # location: radsum_app/back/back
Download https://drive.google.com/file/d/1NfVJa21NIGyX9USrnS-hpqnaWitORqP0/view?usp=sharing, extract and copy it to radsum_app/back/back
.
The path must be radsum_app/back/back/data/
.
python manage.py migrate
python manage.py runserver
Open a new terminal.
npm must be installed (version 9.8.0 recommended).
Enter into cd radsum_app/front/radsum/
.
Run npm install
.
Run npm install react-scripts
.
Finally, run npm start
.
© 2023 Raoufdine SAID & Adrien GUILLE - All rights reserved. This software is distributed under the MIT License.