Q & A Site With ML

An attempt to clone Quora's similar questions ML capability. A question-answer platform where users can ask questions and get answers from other users.

Screenshots

Similar Questions shown when adding a new question. (This section comes under dashboard and requires a user login)

Login as a user

Register an account

Dashboard that is visible after a successful login

Question format expectations when asking a question

Viewing an already asked question

Developer Setup

Requirements

• Python 3.7.9
• Node v14.16.0
• Docker & docker-compose
• vocab file. Download here. Then place it within Backend/model_api/model/

Pre-setup

Setup the basic backend db

• Go to Backend directory and run docker-compose

  docker-compose -f docker-compose.dev.yaml up -d

Backend - Common

• Move to Backend directory

• Add virtual environment using

  python -m venv env

• Activate virtual environment how-to-activate - this depends on OS

• Install project dependencies

  pip install -r requirements.txt

• Create a .env file at the root of the Backend folder using the contents of .env.sample file.

Backend - ML API (Flask)

• Move to Backend\model_api\ directory

• (Optional) To add hot-reloading, Add environment variable by running

  export FLASK_DEBUG=1

  export FLASK_APP=app.py

• Run the app

  python app.py

Backend - Django

• Move to Backend directory

• Run migrations

  python manage.py migrate

• Start the app

  python manage.py runserver

Frontend

• Move to frontend directory

• Install node dependencies

  npm -i

• Start the node application

  npm start