Resources & Extra
TinyML & Edge AI
TinyML is a rapidly evolving field. Here are the core tools and books to help you go deeper into deploying AI on constrained hardware.
Recommended Books
- TinyML: Machine Learning with TensorFlow Lite on Arduino and Ultra-Low-Power Microcontrollers by Pete Warden & Daniel Situnayake. This is the foundational book for the field, written by the engineers who built TensorFlow Lite Micro.
Core Frameworks
- TensorFlow Lite for Microcontrollers: The official guide for TFLite Micro, the runtime used to execute
.tflitemodels on ESP32 and other MCUs. - Edge Impulse: A fantastic, user-friendly platform that automates much of the data collection, DSP (Digital Signal Processing), model training, and deployment pipeline for embedded devices. Highly recommended for rapid prototyping.
- Espressif ESP-NN: Espressif's optimized neural network functions. TFLite Micro uses these under the hood to accelerate math operations using the ESP32-S3's vector instructions.
Datasets and Models
- Kaggle: The premier platform for finding massive datasets to train your models.
- TensorFlow Hub / Kaggle Models: A repository of pre-trained models. While many are too large for microcontrollers, you can find lightweight models (like MobileNet) suitable for the edge.