learning_science/car_price_guesser

Car Price Guesser

Purpose

This project is a learning exercise for building machine learning models. The goal is to predict the price of a car based on its attributes (e.g., Year, Mileage, Brand, Model, etc.) using a dataset of New York cars.

Getting Started

This project uses uv for dependency management.

Prerequisites

• Python 3.14+
• uv installed

Installation

1. Clone the repository (if applicable).
2. Install dependencies:

   uv sync
   

Usage

1. Train the Model

To train the machine learning model (Linear Regression baseline):

uv run train_model.py

This script will:

• Load the New_York_cars.csv dataset.
• Preprocess the data (handle missing values, encode categorical features).
• Train the model.
• Save the trained model to car_price_model.joblib.
• Output performance metrics (MSE, R2 Score).

2. Make Predictions

You can use the trained model to guess the price of a car.

Interactive Mode:

uv run predict.py

Follow the prompts to enter car details.

Command Line Arguments:

uv run predict.py <Year> <Mileage> <Brand>

Example:

uv run predict.py 2020 50000 Toyota

Random Row Verification: To pick a random car from the dataset and compare the model's prediction against the actual price:

uv run predict.py --random