-
This project is a step by step process of how to build a car selling price prediction website. We will first build a model using sklearn and regression algorithms using CarDekho India dataset from kaggle.com.
-
Second step would be to write a python streamlit script that uses the saved model for making predictions.
(Streamlit is open-source app framework is the easiest way for data scientists and machine learning engineers to create beautiful, performant apps.) -
During model building we will cover almost all data science concepts such as data load and cleaning, feature engineering, randomsearchcv for hyperparameter tunning, cross validation etc. Technology and tools wise this project covers,
- Python
- Numpy and Pandas for data cleaning
- Matplotlib and Seaborn for data visualization
- Sklearn for model building
- Jupyter notebook, visual studio code as IDE
- Streamlit to built web app
- Heroku to deploy and manage web apps