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
CANN and Ascend AI Chip Fundamentals
- Defining CANN and its function within Huawei’s AI compute ecosystem
- Examining Ascend processor designs (e.g., 310, 910)
- Survey of supported AI frameworks and toolchains
Model Translation and Compilation
- Applying the ATC utility for model conversion (TensorFlow, PyTorch, ONNX)
- Generating and verifying OM model files
- Managing unsupported operators and typical conversion hurdles
Deployment via MindSpore and Alternate Frameworks
- Implementing models with MindSpore Lite
- Integrating OM models using Python APIs or C++ SDKs
- Utilizing the Ascend Model Manager
Performance Tuning and Profiling
- Analyzing AI Core, memory, and tiling optimizations
- Monitoring model execution through CANN diagnostics
- Best practices for accelerating inference and optimizing resource consumption
Error Management and Debugging
- Identifying and resolving frequent deployment errors
- Interpreting logs and employing the error diagnostic utility
- Conducting unit tests and functional verification of deployed models
Edge and Cloud Deployment Use Cases
- Deploying to Ascend 310 for edge applications
- Connecting with cloud-based APIs and microservices
- Practical case studies in computer vision and NLP
Recap and Future Directions
Requirements
- Proficiency with Python-based deep learning frameworks, including TensorFlow or PyTorch.
- Solid knowledge of neural network structures and model training processes.
- Fundamental command of Linux CLI and scripting capabilities.
Target Audience
- AI engineers focused on model deployment strategies.
- Machine learning specialists aiming for hardware acceleration.
- Deep learning developers creating inference solutions.
