PyWebNN v0.1.0 - Experimental Release

⚠️ EXPERIMENTAL RELEASE - NOT FOR PRODUCTION

DO NOT USE IN PRODUCTION. This is a proof-of-concept for research and experimentation only.

This is the first experimental release of PyWebNN, a Python implementation of the W3C WebNN API with Rust bindings.

---

🎯 Features

• 🐍 Python WebNN API - Complete W3C WebNN specification implementation
• 📊 ONNX Conversion - Full operation support, cross-platform
• 🍎 CoreML Conversion - Basic operations (add, matmul) for macOS
• ✅ Graph Validation - Comprehensive validation matching Chromium's WebNN
• 🔍 Visualization - Generate Graphviz diagrams
• 📦 NumPy Integration - Seamless tensor data conversion
• 🌐 Cross-platform - Linux, macOS, Windows wheels

📥 Installation

pip install pywebnn

🚀 Quick Start

import webnn
import numpy as np

# Create context and builder
ml = webnn.ML()
context = ml.create_context(device_type="cpu")
builder = context.create_graph_builder()

# Build a simple neural network layer
input_tensor = builder.input("input", [1, 4], "float32")
weights = np.random.randn(4, 3).astype(np.float32)
weights_const = builder.constant(weights)
output = builder.matmul(input_tensor, weights_const)

# Compile and convert
graph = builder.build({"output": output})
context.convert_to_onnx(graph, "model.onnx")
print("✓ Model saved to model.onnx")

📚 Documentation

• Getting Started Guide
• API Reference
• Examples
• Full Documentation

⚠️ Known Limitations

• CoreML: Only supports add and matmul operations (activations coming soon)
• Execution: No actual tensor computation in compute() yet - only conversion
• Operations: Subset of WebNN spec implemented (see TODO.txt)
• Testing: Limited test coverage, use for experimentation only

🛠️ What's Included

This release includes pre-built wheels for:

• Linux: x86_64, aarch64 (manylinux)
• macOS: x86_64 (Intel), aarch64 (Apple Silicon)
• Windows: x64, x86
• Python: 3.8, 3.9, 3.10, 3.11, 3.12

🔗 Links

• PyPI: https://pypi.org/project/pywebnn/
• Repository: https://github.com/tarekziade/rustnn
• Issues: https://github.com/tarekziade/rustnn/issues
• W3C WebNN Spec: https://www.w3.org/TR/webnn/

🙏 Acknowledgments

Built with PyO3, Maturin, and inspired by the W3C WebNN Community Group.

---

Made with ❤️ by Tarek Ziadé