Real-Time Neural Network Visualization & Diagnostic Library in Python
NeuroScope is a Python-based library that enables real-time visualization of neural networks.
It helps developers, researchers, and students observe, debug, and interpret deep learning models as they train or run — bringing transparency to the black box of neural networks.
Visualize activations, neurons, and weights dynamically — making it easier to understand your models and communicate them visually.
- 🧩 Visualize neural networks in real time
- 🔍 Observe neuron activations and weight connections
- ⚙️ Detect vanishing or dead neurons
- 🧠 Understand signal flow through forward/backpropagation
- 🎓 Ideal for teaching, research, and debugging
- 🔧 Built with PyTorch (TensorFlow support planned)
from neuroscope import visualize
import torch
import torch.nn as nn
# Example model: 3 inputs → 5 hidden → 2 outputs
class TinyNet(nn.Module):
def __init__(self):
super().__init__()
self.fc1 = nn.Linear(in_features=3, out_features=5)
self.fc2 = nn.Linear(in_features=5, out_features=2)
def forward(self, x):
x = self.fc1(x)
x = torch.relu(x)
x = self.fc2(x)
return x
# Visualize model
if __name__ == "__main__":
model = TinyNet()
visualize(model)
Run the example:
bash
Copy code
python examples/demo_pytorch.py
🖼️ Visualization Preview
Neural Network Code Real-Time Visualization
🧰 Installation
Once available on PyPI:
bash
Copy code
pip install neuroscope
Or install manually:
bash
Copy code
git clone https://github.com/Hassan-xl/NeuroScope.git
cd NeuroScope
pip install -e .
📁 Project Structure
bash
Copy code
NeuroScope/
│
├── neuroscope/ # Core library
├── examples/ # Example demos
├── neuroscope.egg-info/ # Package info
├── README.md # Documentation
├── pyproject.toml # Build configuration
├── setup.py # Setup script
├── Demo-pytorch.jpg # Code screenshot
└── visual-output.jpg # Visualization screenshot
📹 Demo Video
🎥 Watch the live demonstration here:
🔗 https://lnkd.in/eCYQjgkj
🧑💻 Author
Hassan
AI Developer | Deep Learning Enthusiast
📬 [Add your email or LinkedIn link here]
🌐 [Add your personal site or portfolio link here]
📜 License
This project is released under the MIT License — free to use, modify, and distribute.
“You can’t improve what you can’t see.”
NeuroScope turns deep learning into a visible, interactive experience.