Metadata-Version: 2.4
Name: bantai-kecia
Version: 1.0.0
Summary: A Photonic Deep Learning SDK with C++ Backend
Author: S Sardar
Author-email: sumansardar72362-lgtm <sumansardar72362@gmail.com>
Project-URL: Homepage, https://github.com/sumansardar72362-lgtm/BANTAI_Photonic_Sim
Project-URL: Bug Tracker, https://github.com/sumansardar72362-lgtm/BANTAI_Photonic_Sim/issues
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Intended Audience :: Developers
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.8
Description-Content-Type: text/markdown
Requires-Dist: numpy
Requires-Dist: matplotlib
Dynamic: author

<!-- # BANTAI Photonic Compute Stack
# BANTAI_Photonic_Sim -->
# BANTAI Photonic Compute Stack

**A cutting-edge Deep Learning framework simulating Diamond-Glass Hybrid Photonic Chips.**

BANTAI Kecia is a lightweight, high-performance neural network SDK that simulates mathematical interference of light waves for deep learning computations (Optical MAC Unit) and uses a custom 5D Glass Memory system (`.kecia` format) for zero-data-loss weight storage.

## ✨ Key Features
* **Optical MAC Simulation:** Performs Vector-Matrix Multiplication (VMM) at simulated speed-of-light using light intensity and phase.
* **5D Glass Memory System:** Securely saves and loads trained models using the proprietary `.kecia` format.
* **Hardware Fallback Logic:** 3-Tier automatic execution fallback (Photonic -> CUDA -> CPU).
* **Clean PyTorch-like API:** User-friendly and production-ready interface.
* **Built-in Autograd:** Supports custom Backpropagation and SGD Optimizer.

## 📦 Installation
You can easily install the BANTAI SDK via pip:
```bash
pip install bantai-kecia
