Welcome to the ML Algorithms Visualizer! This project is designed to help you visualize machine learning algorithms in action. Currently, it supports Linear Regression, and more algorithms will be added in the future.
This project uses:
- React for building the user interface.
- TensorFlow.js for implementing machine learning algorithms.
- Recharts.js for graph visualization. (Note: We plan to move to D3.js for better customization and performance.)
- Linear Regression: Visualize the training process of a linear regression model.
- Interactive Graph: Draw arbitrary functions on the graph.
- Customizable Learning Rate: Adjust the learning rate to see its effect on the training process.
- Additional Features: Choose extra features like x², x³, x⁴, and sin(x) to see how they affect the predicted line's curvature.
To start using the ML Algorithms Visualizer, follow these steps:
- Visit the Site: Go to ML Algorithms Visualizer.
- Draw a Function: Draw an arbitrary function on the graph.
- Select Learning Rate: Choose a learning rate that suits your needs.
- Choose Extra Features: Select additional features like x², x³, x⁴, and sin(x) to determine how curvy the predicted line will be.
- Train the Model: Press the Train button to visualize the training process.
- Implement more machine learning algorithms.
- Transition to D3.js for enhanced graph visualization and performance improvements.