This is a simple dynamic single-page website developed using the MERN stack (MongoDB, Express.js, React.js, Node.js). The website showcases classical Arabic poetry generated using the GPT-J language model.
You can view the live website from here: https://apg-gptj-bgpp.onrender.com/
The AI-generated classical Arabic poetry showcased on this website was generated using the GPT-J language model. The research paper titled "Generating Classical Arabic Poetry using Pre-trained Models" was presented at the EANLP conference and is now published on the ACL Anthology research journal.
You can view the paper from here: https://aclanthology.org/2022.wanlp-1.6/
To run this project locally, follow the instructions below:
- Clone the repository to your local machine.
- Install the required dependencies by running the following command:
npm install - Start the development server by running the following command:
npm start - Open your browser and visit
http://localhost:3000to access the website.
The website allows you to interact with the AI-generated classical Arabic poetry. You can select various parameters to filter the poems generated by the GPT-J model. The available parameters include:
- Model: Select the desired model for generating the poetry.
- Meter: Choose a specific meter (بحر) for the poems.
- Topic: Specify a topic (موضوع) for the poems.
- Qafya: Select a qafya (قافية) for the poems.
Once you have selected the desired parameters, the website will display the generated poems based on your selection. You can view the poems and click the "View Another Poem" button to load a different poem.
Please note that the poems shown on the website are pre-generated, and the model is not live.
The client code of this project relies on the following dependencies:
- axios
- @mui/material
- react
- react-dom
- react-scripts
Make sure to install these dependencies using npm or any other package manager before running the project.
I would like to express our gratitude to the authors of the research paper "Generating Classical Arabic Poetry using Pre-trained Models" for their valuable contributions to the field of AI-generated poetry.
This project is licensed under the MIT License. Feel free to use, modify, and distribute this code for academic and non-commercial purposes.
If you have any questions, suggestions, or feedback regarding this project, please don't hesitate to contact us.