The Subject Chatbot helps the student to ask some basic questions about the given subject(I have considered AI and ML books for traning it).
The project consists of both frontend react and backend flask components. Below, you'll find explanations of the working of each folder, along with the commands to run the Python (backend) and React (frontend) applications.
npm install
npm run build
npm start
pip install -r requirements.txt
python app.py
Note: Before executing the above commands, be sure you are in the respective folder.
- The
frontend
folder contains the user interface components of the chatbot which is created using react and react-three/fiber
- It contains an input field where user can enter the question and the chatbot returns the answer.
- it also contains 4 buttons as conversation starters
- The
backend
folder contains the flask server which takes the user questions and sends the answers to the React UI.
The backend is designed using Flask.
- We take the input from the user and use langchain library and we get the answer from the pre-trained openAI network and then send back the answer to the user.
- Currently, we trained the model using only two books data b4 and b5 which are inside the backend folder.
- The working prototype can be checked in this URL- https://3-d-subject-chatbot.vercel.app/
- Note that the answer to your question may be delayed because of the free hosting for the backend flask app by 'Render'
- Need to serve different sessions for different users
- Reset functionality for the existing user and erase the conversation to start a new convo.
- Deploy in a much faster hosting
- Train the model with more data. Currently, it is trained on only 2 books.