This repository is a curated list of resources for edge machine learning, targeting researchers, developers, and practitioners interested in deploying ML models on resource-constrained devices. It provides a comprehensive overview of papers, tools, datasets, and hardware relevant to the field.

How It Works

The list is organized into categorized sections, covering key areas like papers on specific applications, AutoML techniques, efficient architectures, federated learning, ML algorithms optimized for edge, network pruning, quantization, datasets, inference engines, MCU/MPU software packages, AI chips, books, and challenges. This structured approach allows users to quickly navigate and find relevant information for their edge ML projects.

Quick Start & Requirements

This is a curated list, not a software package. No installation or execution is required. All resources are linked for direct access.

Highlighted Details

• Extensive list of inference engines including TensorFlow Lite, ONNX Runtime, Arm Compute Library, and more.
• Includes specific datasets like Visual Wake Words for benchmarking tiny vision models.
• Features software packages for MCU/MPU development, such as STM32Cube function packs and NXP's eIQ.
• Covers AI chips and relevant books for on-device ML.

Maintenance & Community

The project is licensed under CC0, indicating a public domain dedication. Contributions are welcome, with data managed in YAML format and generated via scripts, suggesting a structured contribution process.

Licensing & Compatibility

CC0: To the extent possible under law, Bisonai has waived all copyright and related or neighboring rights to this work. This allows for unrestricted use, modification, and distribution, including for commercial purposes.

Limitations & Caveats

As a curated list, the quality and up-to-dateness of the linked resources depend on the maintainers and the community. Some links may become outdated over time.