malore350/nltk

Text Summarization of Research Papers with Python

Introduction

This repo serves the purpose of summarizing texts, and contains two models for that: K-means clustering and BART. It had been trained on papers from arXiv, specifically on the files from Computer Vision category. One can run a demo to interactively utilize it. The tutorial for deploying and using it is given in the next article.

Usage

1. Clone this repo to your favourite destination
2. Download weights for BART model from the following link, and add to the './nltk/model/' folder: https://mega.nz/file/tlQkyIwT#9f17-IRJ8ZsErBTzBMiigSzSVco9m0MG5UwDJxdi6uk
3. Either run the 'demo test.py' file, or run the following snippet in your terminal: python demofile.py.
4. Connect to the generated address.