stavrostheocharis/quotera

Text paraphrasing tool

Deploy a Streamlit app, or a FastAPI API to paraphrase text

Requirements

Python version

• Main supported version : 3.9
  
• Other supported versions : 3.7 & 3.8

To use the scripts on your computer, please make sure you have one of these versions installed.

Install environment & dependencies

In order to run the needed scripts you need to have python installed and run the command below.

python3 -m venv /path/to/new/virtual/environment

To activate your venv run source "env_name"/bin/activate.

To install dependencies run the command:

pip install -r requirements.txt

Also for the parrot library run the command:

pip install git+https://github.com/PrithivirajDamodaran/Parrot_Paraphraser.git

Paraphrase models

This implementation has taken into consideration AI models coming from Hugging Face (specifically the "prithivida/parrot_paraphraser_on_T5" in combination with the Parrot_Paraphraser package, but you can use other variations too). In order to run everything, an account and a token from Hugging Face are mandatory. In order to get everything ready, please read this guide.

Once you have a token from Hugging Face, create a token.txt file in your quotera folder and paste it inside your token (it will automatically be used by the app).

How to use it

As stated there are 2 ways. You can use the Streamlit app, or the API.

Once installed the dependencies

• run the following command from CLI to open the app in the default web browser:

  python quotera_streamlit_cli.py 
  

• run the following command from CLI to open the API in the default web browser:

  python quotera_cli.py serve
  

How to contribute?

We welcome any suggestions, problem reports, and contributions! For any changes you would like to make to this project, we invite you to submit an issue.

For more information, see CONTRIBUTING instructions.

If you wish to containerize the app, see DOCKER instructions.

References

1. How To Create an End-2-End Text Paraphrase App