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
Core Principles of TinyML Workflows
- Summary of TinyML process phases
- Attributes of edge computing hardware
- Strategic considerations for workflow architecture
Data Acquisition and Refinement
- Gathering organized and sensor-derived data
- Methods for data annotation and expansion
- Formatting datasets for restricted environments
TinyML Model Construction
- Choosing architectural designs for microcontrollers
- Training processes utilizing conventional ML frameworks
- Assessing key performance metrics
Model Refinement and Reduction
- Quantization methods
- Pruning strategies and weight sharing
- Reconciling precision with resource limitations
Model Transformation and Packaging
- Exporting models to TensorFlow Lite
- Incorporating models into embedded development toolchains
- Handling model dimensions and memory restrictions
Implementation on Microcontrollers
- Writing models to hardware targets
- Setting up runtime environments
- Conducting real-time inference assessments
Surveillance, Testing, and Verification
- Validation approaches for deployed TinyML systems
- Troubleshooting model performance on hardware
- Verifying performance in operational field conditions
Integrating the Complete End-to-End Workflow
- Establishing automated processing streams
- Version control for data, models, and firmware
- Oversight of updates and iterative improvements
Conclusion and Subsequent Actions
Requirements
- Comprehension of core machine learning principles
- Proficiency in embedded coding
- Acquaintance with Python-centric data processing pipelines
Target Audience
- Artificial Intelligence Engineers
- Software Developers
- Embedded Systems Specialists
