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
Overview of CANN and Ascend AI Processors
- Defining CANN and its position within Huawei’s AI computing ecosystem
- Introduction to Ascend processor architecture (including models 310, 910, etc.)
- Review of supported AI frameworks and the associated toolchain
Model Conversion and Compilation Processes
- Utilizing the ATC tool for converting models from TensorFlow, PyTorch, and ONNX
- Generating and verifying OM model files
- Addressing unsupported operators and typical conversion challenges
Deployment via MindSpore and Other Frameworks
- Implementing model deployment using MindSpore Lite
- Incorporating OM models into applications via Python APIs or C++ SDKs
- Managing components with the Ascend Model Manager
Performance Tuning and Profiling
- Exploring optimizations for AI Cores, memory management, and tiling
- Analyzing model execution using CANN profiling tools
- Applying best practices to enhance inference speed and resource efficiency
Error Management and Debugging Techniques
- Identifying and resolving frequent deployment errors
- Interpreting logs and utilizing the error diagnosis utility
- Conducting unit tests and functional verification for deployed models
Edge and Cloud Deployment Cases
- Deploying to Ascend 310 for edge-case applications
- Connecting with cloud-based APIs and microservice architectures
- Examining real-world case studies in computer vision and NLP
Recap and Future Directions
Requirements
- Proficiency with Python-based deep learning frameworks, such as TensorFlow or PyTorch
- Solid understanding of neural network architectures and model training workflows
- Foundational knowledge of the Linux command-line interface and scripting
Target Audience
- AI engineers focused on model deployment
- Machine learning professionals aiming to leverage hardware acceleration
- Deep learning developers creating inference solutions
14 Hours
