Thank you for sending your enquiry! One of our team members will contact you shortly.
Thank you for sending your booking! One of our team members will contact you shortly.
Duration 21 hours
Course Outline
Foundations of TinyML
- Analyzing the constraints and potential of TinyML
- Surveying prevalent microcontroller architectures
- Evaluating Raspberry Pi against Arduino and alternative boards
Hardware Configuration and Initial Setup
- Setting up the Raspberry Pi OS environment
- Setting up Arduino boards for operation
- Linking sensors and auxiliary peripherals
Strategies for Data Acquisition
- Recording sensor inputs
- Managing audio, motion, and environmental metrics
- Constructing annotated datasets
Developing Models for Edge Hardware
- Choosing appropriate model structures
- Training TinyML models using TensorFlow Lite
- Assessing performance for embedded applications
Optimizing and Converting Models
- Applying quantization techniques
- Transforming models for microcontroller implementation
- Enhancing memory usage and computational efficiency
Implementation on Raspberry Pi
- Executing TensorFlow Lite inference
- Incorporating model results into software applications
- Resolving performance-related challenges
Implementation on Arduino
- Utilizing the Arduino TensorFlow Lite Micro library
- Installing models onto microcontrollers
- Verifying accuracy and runtime behavior
Constructing Complete TinyML Systems
- Architecting comprehensive embedded AI processes
- Deploying interactive, real-world prototypes
- Testing and iterating on project features
Wrap-up and Future Directions
Requirements
- A solid grasp of fundamental programming principles.
- Practical experience in utilizing microcontrollers.
- Working knowledge of Python or C/C++.
Intended Participants
- Makers.
- Hobbyists.
- Developers specializing in Embedded AI.
Testimonials (1)
The exercises
