This project focuses on using machine learning algorithms to estimate car prices. Various regression algorithms were implemented, including:

• Linear Regression
• Lasso Regression
• Ridge Regression
• Decision Tree
• Random Forest
• XGBoost

Model evaluation, grid-search, and cross-validation were performed, resulting in the following scores:

The final models chosen were Random Forest and XGBoost. Feature importance was determined separately for each model to reduce feature counts. The models were saved using Pickle and converted into a Streamlit file for deployment outside the notebook environment. The Streamlit file was published on both AWS EC2 instances and the Streamlit website, enabling users to make predictions interactively.

• Streamlit Live
• AWS EC2
• GitHub Notebook Link
• Kaggle Notebook Link
• LinkedIn