https://github.com/peterk/swebert-summarize

A tiny NLP demo on how to summarize swedish texts using the BERT model from the National Library of Sweden.
https://github.com/peterk/swebert-summarize

bert-model docker nlp-machine-learning python

Last synced: 6 months ago
JSON representation

A tiny NLP demo on how to summarize swedish texts using the BERT model from the National Library of Sweden.

• Host: GitHub
• URL: https://github.com/peterk/swebert-summarize
• Owner: peterk
• License: apache-2.0
• Created: 2020-05-20T13:27:06.000Z (about 6 years ago)
• Default Branch: master
• Last Pushed: 2024-07-27T15:51:10.000Z (almost 2 years ago)
• Last Synced: 2025-01-20T13:35:32.110Z (over 1 year ago)
• Topics: bert-model, docker, nlp-machine-learning, python
• Language: HTML
• Homepage:
• Size: 148 KB
• Stars: 0
• Watchers: 2
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# swebert-summarize

A tiny demo on how to summarize swedish texts using the [BERT model from the National Library of Sweden](https://github.com/Kungbib/swedish-bert-models). It uses the [Bert Extractive Summarizer library](https://pypi.org/project/bert-extractive-summarizer/).

Install requirements:

```pip install -r requirements.txt```

Run it with 

```python summarize.py my_long_text.txt```

# Simple web gui

A simple web gui in Flask is provided for testing. To run it, install Docker and docker-compose. 

To run it:

```docker-compose up -d```

On the first run the BERT-model is downloaded (this may take several minutes). The ser ver will be available on 0.0.0.0:80.