Car Selling Price Prediction 🚙

A simple ML app to predict the prices of used cars. It's build using traditional ML algorithm and deployed using Flask on Heroku.

Demo

Link: https://car-selling1price.herokuapp.com/

Data on Mongodb Atlas

💻 Local Installation

Clone this Repo.
Create a virtual environment using venv or conda to keep your project packages separately.
After navigating to the project base folder and activating the enviroment, run this command to install the required packages:

pip install -r requirements.txt

Now if you are using conda, you can directly create the env and install all packages using the file 'requirements.yml'. Just run:

conda env create -f requirements.yml

After installation, run 'app.py' which will use Flask's development server:

python app.py

📂 Directory Tree

.
├── app.py
├── data
│   ├── Car_details.csv
│   └── demo.gif
├── database.py
├── LICENSE
├── models
│   └── model.pkl
├── notebooks
│   └── Car_Selling_Price_Prediction.ipynb
├── preprocess.py
├── Procfile
├── README.md
├── requirements.txt
├── requirements.yml
├── static
│   └── index.css
└── templates
    ├── index.html
    └── output.html

5 directories, 15 files

Deployment on Heroku

We will be using gunicorn as our web server.

Create a Procfile, a text file in the root directory of application, to explicitly declare what command should be executed by heroku to start your app.

web: gunicorn <app_file_name>:app

Login or signup to Heroku in order to create virtual app. You can either connect your github profile or download Heroku CLI to manually deploy this project.

Database

To store new data inputted by user, we have used a NoSQL MongoDB database and hosted it on MongoDB Atlas. This stored data can be further used for training a better model.

SankalpM-29/car_selling_price