sparse-coding 1.1.0

Biological vision principles for efficient representation learning from natural images

💰 Support This Research - Please Donate!

🙏 If this library helps your research or project, please consider donating to support continued development:

💳 DONATE VIA PAYPAL - CLICK HERE

---

Sparse Coding

💰 Biological vision principles for representation learning

Olshausen, B. A., & Field, D. J. (1996) - "Emergence of simple-cell receptive field properties"

📦 Installation

pip install sparse-coding

🚀 Quick Start

import sparse_coding
import numpy as np

# Create sample image patches (8x8 patches)
patches = np.random.randn(1000, 64)

# Initialize sparse coder
coder = sparse_coding.SparseCoder(
    dictionary_size=128,
    sparsity_lambda=0.1
)

# Learn sparse dictionary
coder.fit(patches)

# Encode new patches sparsely
test_patch = np.random.randn(1, 64)
sparse_code = coder.encode(test_patch)
reconstructed = coder.decode(sparse_code)

print(f"✅ Sparse coding: {np.sum(sparse_code != 0)} active out of {len(sparse_code)} atoms")
print(f"✅ Reconstruction error: {np.mean((test_patch - reconstructed)**2):.4f}")

🎓 About the Implementation

Implemented by Benedict Chen - bringing foundational AI research to modern Python.

📧 Contact: [e-mail]

📖 Citation

If you use this implementation in your research, please cite the original paper:

Olshausen, B. A., & Field, D. J. (1996) - "Emergence of simple-cell receptive field properties"

📜 License

Custom Non-Commercial License with Donation Requirements - See LICENSE file for details.

---

💰 Support This Work - Donation Appreciated!

This implementation represents hundreds of hours of research and development. If you find it valuable, please consider donating:

💳 DONATE VIA PAYPAL - CLICK HERE

Your support helps maintain and expand these research implementations! 🙏