Rindow

PHP libraries for practical computing.

Build more with Rindow — Machine learning, mathematics, and application frameworks designed to work together.

---

What we build

🧠 Neural Networks	📐 Mathematics	⚙️ Framework
Deep learning in pure PHP — DNN, CNN, RNN, Attention / Transformer, and RL agents. Keras-like API. CPU speed close to TensorFlow, GPU via OpenCL without NVIDIA.	Matrix operations on N-dimensional arrays, high-speed BLAS/LAPACK, plotting, and acceleration.	Modular PHP application architecture with DI Container, AOP / Interceptors, and annotation-based configuration.
Docs →	Docs →	Docs →

Quick start

composer require rindow/rindow-neuralnetworks
composer require rindow/rindow-math-plot
composer require rindow/rindow-math-matrix-matlibffi  # high-speed backend

Tutorials: Basic image classification · CNN · RNN seq2seq · NMT with Attention · Transformer

Why Rindow

• Off the chain — No dedicated ML environment, no expensive GPU machine. Run deep learning on your laptop and on ordinary PHP hosting.
• Decluttering — Small, modular, healthy components. Use only what you need, discard the rest.
• Interoperability — Open interfaces first. Built to be replaceable, aiming for standards anyone can join.

Links

• 🌐 Website: https://rindow.github.io/
• 📖 Blog: https://rindow.github.io/blog/index.html
• 💻 GitHub: https://github.com/rindow
• 📦 Packagist: https://packagist.org/packages/rindow/

Let's have fun together!