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.
Course Outline
Foundations of TinyML Pipelines
- Overview of TinyML workflow stages.
- Characteristics of edge hardware.
- Pipeline design considerations.
Data Collection and Preprocessing
- Collecting structured and sensor data.
- Data labeling and augmentation strategies.
- Preparing datasets for constrained environments.
Model Development for TinyML
- Selecting model architectures for microcontrollers.
- Training workflows using standard ML frameworks.
- Evaluating model performance indicators.
Model Optimization and Compression
- Quantization techniques.
- Pruning and weight sharing.
- Balancing accuracy and resource limits.
Model Conversion and Packaging
- Exporting models to TensorFlow Lite.
- Integrating models into embedded toolchains.
- Managing model size and memory constraints.
Deployment on Microcontrollers
- Flashing models onto hardware targets.
- Configuring run-time environments.
- Real-time inference testing.
Monitoring, Testing, and Validation
- Testing strategies for deployed TinyML systems.
- Debugging model behaviour on hardware.
- Performance validation in field conditions.
Integrating the Full End-to-End Pipeline
- Building automated workflows.
- Versioning data, models, and firmware.
- Managing updates and iterations.
Summary and Next Steps
Requirements
- A foundational understanding of machine learning concepts.
- Experience in embedded programming.
- Familiarity with Python-based data workflows.
Target Audience
- AI engineers.
- Software developers.
- Embedded systems experts.
21 Hours
