Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/bjornmelin/edge-ai-engineering
📱 Optimized ML for edge devices. Showcasing efficient model deployment, GPU-CPU memory transfer optimization, and real-world edge AI applications. 🤖
https://github.com/bjornmelin/edge-ai-engineering
cuda edge-computing embedded-systems gpu-optimization iot mobile-ml model-optimization python tflite
Last synced: 11 days ago
JSON representation
📱 Optimized ML for edge devices. Showcasing efficient model deployment, GPU-CPU memory transfer optimization, and real-world edge AI applications. 🤖
- Host: GitHub
- URL: https://github.com/bjornmelin/edge-ai-engineering
- Owner: BjornMelin
- License: mit
- Created: 2025-01-24T14:49:12.000Z (20 days ago)
- Default Branch: main
- Last Pushed: 2025-01-24T14:51:18.000Z (20 days ago)
- Last Synced: 2025-01-24T15:33:25.854Z (20 days ago)
- Topics: cuda, edge-computing, embedded-systems, gpu-optimization, iot, mobile-ml, model-optimization, python, tflite
- Homepage:
- Size: 0 Bytes
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Edge AI Engineering 📱
[](https://www.python.org/downloads/)
[](https://www.tensorflow.org/lite)
[](https://pytorch.org/mobile)
[](https://developer.nvidia.com/cuda-toolkit)
[](LICENSE)
> Optimized machine learning models for edge and mobile devices. Showcasing efficient model deployment, optimization techniques, and real-world edge AI applications.
[Features](#features) • [Installation](#installation) • [Quick Start](#quick-start) • [Documentation](#documentation) • [Contributing](#contributing)
## 📑 Table of Contents
- [Features](#features)
- [Project Structure](#project-structure)
- [Prerequisites](#prerequisites)
- [Installation](#installation)
- [Quick Start](#quick-start)
- [Documentation](#documentation)
- [Models](#models)
- [Optimization](#optimization)
- [Benchmarks](#benchmarks)
- [Contributing](#contributing)
- [Versioning](#versioning)
- [Authors](#authors)
- [Citation](#citation)
- [License](#license)
- [Acknowledgments](#acknowledgments)
## ✨ Features
- Model quantization and optimization
- Mobile-first architectures
- Battery-efficient inference
- Cross-platform deployment
- Edge-optimized pipelines
## 📁 Project Structure
```mermaid
graph TD
A[edge-ai-engineering] --> B[models]
A --> C[optimization]
A --> D[deployment]
A --> E[benchmarks]
B --> F[tflite]
B --> G[pytorch-mobile]
C --> H[quantization]
C --> I[compression]
D --> J[android]
D --> K[ios]
E --> L[performance]
E --> M[battery]
```
Click to expand full directory structure
```plaintext
edge-ai-engineering/
├── models/ # Model implementations
│ ├── tflite/ # TensorFlow Lite models
│ └── pytorch/ # PyTorch Mobile models
├── optimization/ # Optimization tools
│ ├── quantization/ # Model quantization
│ └── compression/ # Model compression
├── deployment/ # Platform-specific deployment
│ ├── android/ # Android deployment
│ └── ios/ # iOS deployment
├── benchmarks/ # Performance testing
└── README.md # Documentation
```
## 🔧 Prerequisites
- Python 3.8+
- TensorFlow Lite 2.14+
- PyTorch Mobile 2.2+
- Android SDK/NDK
- Xcode (for iOS)
## 📦 Installation
```bash
# Clone repository
git clone https://github.com/BjornMelin/edge-ai-engineering.git
cd edge-ai-engineering
# Create environment
python -m venv venv
source venv/bin/activate
# Install dependencies
pip install -r requirements.txt
```
## 🚀 Quick Start
```python
from edge_ai import optimization, deployment
# Optimize model for mobile
optimized_model = optimization.quantize_for_mobile(
model,
target_platform="android",
quantization="int8"
)
# Deploy to device
deployment = deployment.MobileDeployment(
model=optimized_model,
platform="android",
optimize_battery=True
)
# Generate deployment package
deployment.export()
```
## 📚 Documentation
### Models
| Model | Task | Size | Latency (ms) |
|-------|------|------|--------------|
| MobileNetV3 | Classification | 4MB | 15 |
| TinyYOLO | Detection | 8MB | 25 |
| MobileViT | Vision | 6MB | 20 |
### Optimization
- Int8 quantization
- Model pruning
- Architecture optimization
- Memory footprint reduction
### Benchmarks
Performance on different devices:
| Device | Model | Battery Impact | FPS | Memory |
|--------|-------|----------------|-----|---------|
| Pixel 6 | MobileNet | 2%/hr | 30 | 120MB |
| iPhone 13 | TinyYOLO | 3%/hr | 25 | 150MB |
| RPi 4 | MobileViT | N/A | 15 | 200MB |
## 🤝 Contributing
- [Contributing Guidelines](CONTRIBUTING.md)
- [Code of Conduct](CODE_OF_CONDUCT.md)
- [Development Guide](DEVELOPMENT.md)
## 📌 Versioning
We use [SemVer](http://semver.org/) for versioning. For available versions, see the [tags on this repository](https://github.com/BjornMelin/edge-ai-engineering/tags).
## ✍️ Authors
**Bjorn Melin**
- GitHub: [@BjornMelin](https://github.com/BjornMelin)
- LinkedIn: [Bjorn Melin](https://linkedin.com/in/bjorn-melin)
## 📝 Citation
```bibtex
@misc{melin2024edgeaiengineering,
author = {Melin, Bjorn},
title = {Edge AI Engineering: Optimized Mobile Machine Learning},
year = {2024},
publisher = {GitHub},
url = {https://github.com/BjornMelin/edge-ai-engineering}
}
```
## 📄 License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
## 🙏 Acknowledgments
- TensorFlow Lite team
- PyTorch Mobile developers
- Mobile ML community
- Edge computing researchers
---
Made with 📱 and ❤️ by Bjorn Melin