Metadata-Version: 2.4
Name: photokan
Version: 0.0.1
Summary: Photonic KAN simulation for PyTorch
Author-email: Koneti <koneti@gmail.com>
License: MIT
Project-URL: Homepage, https://photokan.dev
Project-URL: Repository, https://github.com/koneti/photokan
Keywords: photonic,kan,kolmogorov-arnold,pytorch,neuromorphic,optical
Classifier: Development Status :: 1 - Planning
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.9
Description-Content-Type: text/markdown

# photokan

Photonic KAN simulation for PyTorch — an open-source bridge between Kolmogorov-Arnold Networks and photonic hardware constraints.

**Status:** Coming soon. Follow development at [photokan.dev](https://photokan.dev).

## What's Coming

- `PhotonicKANLayer` — KAN layer with photonic hardware constraints
- Photonic simulation backend with realistic TFLN noise models
- Energy cost estimation per operation
- MNIST / CIFAR-10 benchmarks: GPU vs simulated photonic execution
- Model profiler for photonic offload candidate detection

## Install

```bash
pip install photokan
```

## License

MIT
