3v324v23's picture
Initial commit of Neural Network Playground with improved node display and Linear Regression support
f02caca
|
raw
history blame
1.55 kB

Neural Network Playground

An interactive web-based application for visualizing and experimenting with neural network architectures.

Features

  • Drag-and-Drop Interface: Easily create neural network architectures by dragging and dropping different layer types
  • Multiple Layer Types: Support for Input, Hidden, Output, Convolutional, and Pooling layers
  • Dynamic Connections: Create connections between layers to define your network topology
  • Visual Styling: Beautiful gradient-based styling for different layer types with animations
  • Layer Properties: View and edit detailed properties for each layer
  • Network Validation: Automatic validation of network architectures
  • Training Simulation: Visual simulation of the training process
  • Responsive Design: Works on desktop and mobile devices

Getting Started

  1. Clone this repository
  2. Open index.html in your browser or use a local server:
    python -m http.server
    
  3. Visit http://localhost:8000 in your browser

How to Use

  1. Drag layer components from the left panel onto the canvas
  2. Connect layers by dragging from output ports (right side) to input ports (left side)
  3. Click on a layer to view its properties
  4. Edit layer properties by clicking the edit button
  5. Click "Run Network" to simulate training

Technologies Used

  • HTML5
  • CSS3 (with animations and gradients)
  • JavaScript (vanilla)
  • No external libraries required!

License

MIT

Contributing

Contributions, issues, and feature requests are welcome!