UPDATED to 2025 - 2nd Semester
- Materials will be uploaded to this repo every week.
- Slides, Notebooks, Code, and Docs in English
- Videos in Portuguese
The students should be familiar with Embedded Machine Learning (TinyML).
For the students who have not previously attended the IESTI01 TinyML course, it is suggested to review the following classes before starting the IESTI05 course:
- Class 2 - Introduction to TinyML [Slides] [Docs] [Video]
- Class 3 - TinyML Challenges - Embedded Systems [Slides] [Docs] [Video]
- Class 4 - TinyML Challenges - Machine Learning [Slides] [Video]
- Class 9 - The Building Blocks of DL - DNN Recap, Datasets and Model Performance Metrics [Slides][Notebooks and Docs] [Video]
- Class 10 - Introducing Convolutions (CNN) [Slides] [Docs] [Video]
- Class 11 - Image Classification using CNN [Slides] [Notebooks and Docs] [Video]
- Class 12 - Introduction to Edge Impulse – CNN with Cifar-10 [Slides] [Notebooks] [Video]
- Class 25 - Image Classification Introduction [Slides] [Docs] [Notebooks] [Video]
- Class 26 - Image Classification using Edge Impulse Studio [Slides] [Video]
- Class 1 - Introduction to Edge Ai Course [Slides] [Video] [Quiz]
- Class 2 - Introduction to Linux [Slides] [Docs] [Video] [Quiz] [Book]
- Class 3 - Raspberry Pi Setup Walkthrough [Slides] [Docs] [Video] [Quiz]
- Class 4 - Setup Python Environment [Slides] [Docs] [Video] [Quiz] [Book]
- Class 5 - Image Classification: Introduction to TFLite [Slides] [Docs] [Video] [Quiz] [Book]
- Class 6 - Image Classification Project: Goal & Data [Slides] [Docs] [Video] [Quiz] [Book]
- Class 7 - Image Classification Project: Training, Test & Deploy - EI [Slides] [Docs] [Video] [Quiz] [Book]
- Class 8 - Object Detection: Fundamentals [Slides] [Docs] [Video] [Quiz] [Book]
- Class 9 - Object Detection Project: Goal, Data & Labeling [Slides] [Docs] [Video] [Quiz] [Book]
- Class 10 - Object Detection Project: Training, Test & Deploy - EI [Slides] [Docs] [Video] [Quiz] [Book]
- Class 11 - Object Detection with YOLO: Fundamentals [Slides] [Docs] [Video] [Quiz] [Book]
- Class 12 - Object Detection Project: Train & Deploy – YOLO [Slides] [Docs] [Video] [Quiz] [Book]
- Class 13 - Bee Counting Project: Dataset, Train & Deploy – YOLO (Anderson Reis) [Slides] [Video] [Book]
- Class 14 - Bee Counting Project: Embedded Linux implementation (Guilherme Fernandes) [Slides] [Docs] [Video]
- Class 15 - Text Generation – RNN: The VerneBot [Slides] [Docs] [Video] [Quiz] [Book]
- Class 16a - Raspberry Pi 5 Setup [Slides] [Docs] [Video] [Quiz] [Book]
- Class 16b - Generative AI at the Edge - SLM (Small Language Models) [Slides] [Video] [Quiz] [Book]
- Class 17 - SLM with Ollama Review [Slides] [Video] [Quiz] [Book]
- Class 18 - Ollama with Python [Slides] [Docs] [Video] [Quiz] [Book]
- Class 19 - SLM: Optimization Techniques - Function (tools) Calling [Slides] [Docs] [Video] [Quiz] [Book]
- Class 20 - SLM: Optimization Techniques - RAG [Slides] [Docs] [Video] [Quiz] [Book]
- Class 20a - SLM: Optimization Techniques - Agents [Slides] [Docs] [Book]
- Class 21 - GPIO: Physical Computing [Slides] [Docs] [Video] [Quiz] [Book]
- Class 22 - SLMs_for_IoT_Control [Slides] [Docs] [Video] [Quiz] [Book]
- Class 23 - Application of Natural Language Models in IoT Edge Devices - SLM View [Slides] [Video]
- Class 24 - Application of Natural Language Models in IoT Edge Devices - Hardware/System View [Slides] [Video]
-
Part 1 [Assignment]
- Detecção de Câncer Cervical [Report] [Video]
- Coin Counting [Report] [GitHub] [Video]
- Fruit Quality Assessments [Report] [GitHub] [Video]
- Jokenpo [Report] [Video]
- Plant Disease Classification [Report] [Video]
- MediTrack [Report] [Video] [Demo]
- Detecção e Classificação de Carros e Placas de Trânsito [Report] [GitHub] [Video]
-
Part 2 [Assignment] [Video]
- To be delivered on December 7th
Edge AI Engineering with Raspberry Pi is a 15-week undergraduate course designed to teach students how to implement AI systems on edge devices, specifically using Raspberry Pi platforms.
The course is based on the e-book: "Edge AI Engineering" by Prof. Marcelo Rovai, UNIFEI 2025
The course is divided into two main parts:
- Part 1 (Weeks 1-7): Fixed Function AI - Focus on image classification and object detection
- Part 2 (Weeks 8-15): Generative AI - Focus on Small Language Models (SLMs) and RAG systems
Technical Skills:
- Raspberry Pi setup, configuration, and optimization
- Computer vision with OpenCV and TensorFlow Lite
- Image classification and object detection implementation
- Small Language Model deployment and integration
- Retrieval-Augmented Generation (RAG) systems
- Physical computing integration with sensors and actuators
Practical Applications:
- Real-time image processing and object detection
- Custom model training using Edge Impulse
- Building conversational AI systems on edge devices
- Creating intelligent IoT monitoring systems
- Natural language control of physical devices
- 40% - Weekly hands-on labs, quizzes and surveys
- 20% - Midterm project (Fixed Function AI system)
- 30% - Final project (Generative AI application)
- 10% - Participation and documentation
Students need basic knowledge of Python programming, Linux systems, Deep Learning, and electronics fundamentals.
The course emphasizes practical, hands-on learning with students building real AI applications that run efficiently on resource-constrained edge devices, bridging the gap between traditional computer vision and modern generative AI technologies.