NeuralNet Visualizer

An interactive web app to visualize neural network architectures and their activations.
Built with Python + PyTorch for model building, and a lightweight web front-end for visualization.

Live demo (GitHub Pages)

Tips: Run local for best performance. Thanks

---

Features & Usage

Feature	Description
Custom Model Support	Define any PyTorch model (MLP, CNN, Transformer, etc.) and visualize it directly.
Interactive Play Mode	Animate activations flowing through the network with a single click.
Layer & Neuron Mapping	Shows layers, neurons, and weighted connections clearly.
Load .pth Files	Import and visualize pretrained models saved in PyTorch.
Deploy Anywhere	Works locally or on GitHub Pages for free hosting.

• Provide Python Model code + load .pth model file.
• Displays layers, neurons, and weighted connections by use Build button
• Interactive Run button to animate activations

---

Getting Started (Run Your Own Locally)

1. Clone the repo

git clone https://github.com/Iro96/NeuralNet-Visualizer.git
cd neural-net-visualizer

2. Start the backend server

python server.py

3. Experience the tool

• Just open file index.html in root directory ( ./ )
• Experience the tool your own way.

---

The End

Thanks a lot for the help from @ZhenYu15