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 14 hours
Course Outline
Introduction to Edge AI and the Ascend 310
- Understanding Edge AI: key trends, hardware constraints, and practical applications
- Exploring the Huawei Ascend 310 chip architecture and its supported toolchain
- Understanding the role of CANN within the broader edge AI deployment ecosystem
Model Preparation and Conversion
- Exporting trained models from frameworks such as TensorFlow, PyTorch, and MindSpore
- Utilizing ATC to transform models into the OM format compatible with Ascend devices
- Addressing unsupported operators and implementing lightweight conversion strategies
Developing Inference Pipelines with AscendCL
- Executing OM models on the Ascend 310 using the AscendCL API
- Managing input/output preprocessing, memory allocation, and device control
- Integrating solutions within embedded containers or lightweight runtime environments
Optimization for Edge Constraints
- Minimizing model size and adjusting precision (FP16, INT8) for efficiency
- Employing the CANN profiler to pinpoint and resolve performance bottlenecks
- Optimizing memory layout and data streaming to enhance overall performance
Deploying with MindSpore Lite
- Leveraging the MindSpore Lite runtime for mobile and embedded targets
- Evaluating MindSpore Lite against raw AscendCL pipelines
- Packaging inference models tailored for specific device deployments
Edge Deployment Scenarios and Case Studies
- Case study: Implementing object detection on a smart camera using the Ascend 310
- Case study: Achieving real-time classification within an IoT sensor hub
- Strategies for monitoring and updating models deployed at the edge
Summary and Next Steps
Requirements
- Prior experience with AI model development or deployment workflows
- Basic understanding of embedded systems, Linux, and Python
- Familiarity with deep learning frameworks such as TensorFlow or PyTorch
Audience
- IoT solution developers
- Embedded AI engineers
- Edge system integrators and AI deployment specialists
Testimonials (1)
That we can cover advance topic and work with real-life example
