Welcome to Linear Regression with Gradient Descent, a Rust-based project designed to make linear regression approachable and visually stunning! Whether you’re a data science enthusiast or a Rustacean, this project has something for you.

This repository implements a linear regression model using gradient descent. It includes:

1. Trainer: Train the model with your dataset.
2. Predictor: Use the trained model to predict values.
3. GUI: Visualize the data, regression line, and gradient descent process interactively.
4. Linear Regression: library to handle linear regression stuff.

• Gradient Descent: Train the model iteratively.
• Interactive GUI: Adjust settings, visualize plots, and watch gradient descent in action.
• Makefile Integration: Simplify build and execution workflows.
• Environment Configuration: Store paths and configurations in a .env file.

Here’s the project layout:

. ├── Cargo.lock # Dependency lock file ├── Cargo.toml # Workspace configuration ├── data # Data folder │ ├── data.csv # Dataset for training │ └── theta.txt # Model parameters (theta0, theta1) ├── gui # GUI application │ ├── Cargo.toml │ └── src │ ├── app.rs # Main application logic │ ├── components # Reusable UI components │ │ ├── mod.rs │ │ ├── navbar.rs │ │ ├── plane.rs │ │ └── sidebar.rs │ ├── main.rs # GUI entry point │ ├── settings # Settings modules │ │ ├── grid_settings.rs │ │ ├── mod.rs │ │ ├── plot_settings.rs │ │ └── sidebar_settings.rs │ └── utils.rs # Helper functions ├── linear_regression # Core library for regression logic │ ├── Cargo.toml │ └── src │ ├── lib.rs │ ├── linear_regression.rs │ └── utils.rs ├── Makefile # Build and execution shortcuts ├── predictor # Predictor application │ ├── Cargo.toml │ └── src │ └── main.rs # Entry point for predictions ├── README.md # This file! ├── trainer # Trainer application │ ├── Cargo.toml │ └── src │ └── main.rs # Entry point for training 

To keep things clean and configurable, the project uses a .env file. Make sure to create it in the root directory:

THETA_PATH=data/theta.txt DATASET_PATH=data/data.csv 

• THETA_PATH: Path to the file where model parameters (theta0, theta1) are stored.
• DATASET_PATH: Path to the CSV file containing your dataset.

The Makefile simplifies building and running the project. Here’s what you can do:

make build

make predictor

make trainer

make gui

make web

make clean

1. Ensure data/data.csv contains your dataset (e.g., mileage vs. price).
2. Run the trainer:

   make trainer

3. The trained parameters will be saved to data/theta.txt.

1. Use the predictor to estimate values based on the trained model:

   make predictor

2. Enter a mileage value when prompted, and the program will output the estimated price.

1. Run the GUI:

   make gui # or make web

2. Explore your dataset, adjust parameters, and watch the regression line and gradient descent come to life!

Your data/data.csv should look like this:

mileage,price 5000,20000 10000,18000 15000,16000 20000,14000 25000,12000 

Contributions are welcome! Feel free to open issues or submit pull requests to enhance the project.

Enjoy using Linear Regression with Gradient Descent! 🚀 Let me know if you have any feedback or feature requests.