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
- Overview of Edge AI: current trends, limitations, and use cases
- Architecture of the Huawei Ascend 310 chip and its compatible toolchain
- The role of CANN in the edge AI deployment ecosystem
Preparing and Converting Models
- Exporting trained models from TensorFlow, PyTorch, and MindSpore
- Leveraging ATC to transform models into OM format for Ascend hardware
- Addressing unsupported operations and employing lightweight conversion tactics
Creating Inference Pipelines with AscendCL
- Executing OM models on Ascend 310 via the AscendCL API
- Managing input/output preprocessing, memory, and device control
- Integration within embedded containers or lightweight runtime settings
Optimizing for Edge Limitations
- Minimizing model footprint and adjusting precision (FP16, INT8)
- Identifying performance bottlenecks using the CANN profiler
- Overseeing memory arrangement and data streaming for optimal performance
Deployment using MindSpore Lite
- Utilizing the MindSpore Lite runtime for mobile and embedded platforms
- Contrasting MindSpore Lite against the raw AscendCL pipeline
- Bundling inference models for device-specific deployment
Edge Deployment Scenarios and Real-World Cases
- Case study: implementing an object detection model on a smart camera with Ascend 310
- Case study: real-time classification within an IoT sensor hub
- Monitoring and updating models deployed at the edge
Recap and Future Directions
Requirements
- Background in AI model creation or deployment processes
- Foundational understanding of embedded systems, Linux, and Python
- Working knowledge of deep learning frameworks like TensorFlow or PyTorch
Target Audience
- Developers building IoT solutions
- Engineers specialized in embedded AI
- Edge system integrators and specialists in AI deployment
Testimonials (1)
That we can cover advance topic and work with real-life example