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
Introduction to CANN and Ascend AI Processors
- What is CANN? Its role in Huawei’s AI compute stack.
- An overview of the Ascend processor architecture (including models like 310 and 910).
- Overview of supported AI frameworks and the toolchain.
Model Conversion and Compilation
- Using the ATC tool for model conversion (covering TensorFlow, PyTorch, ONNX).
- Creating and validating OM model files.
- Addressing unsupported operators and common conversion challenges.
Deploying with MindSpore and Other Frameworks
- Deploying models using MindSpore Lite.
- Integrating OM models with Python APIs or C++ SDKs.
- Working with the Ascend Model Manager.
Performance Optimization and Profiling
- Understanding AI Core, memory, and tiling optimizations.
- Profiling model execution using CANN tools.
- Best practices for enhancing inference speed and resource efficiency.
Error Handling and Debugging
- Common deployment errors and their resolutions.
- Reading logs and utilizing the error diagnosis tool.
- Performing unit testing and functional validation of deployed models.
Edge and Cloud Deployment Scenarios
- Deploying to Ascend 310 for edge applications.
- Integrating with cloud-based APIs and microservices.
- Real-world case studies in computer vision and NLP.
Summary and Next Steps
Requirements
- Experience with Python-based deep learning frameworks, such as TensorFlow or PyTorch.
- A solid understanding of neural network architectures and model training workflows.
- Basic familiarity with the Linux command-line interface and scripting.
Audience
- AI engineers focused on model deployment.
- Machine learning practitioners aiming to leverage hardware acceleration.
- Deep learning developers constructing inference solutions.
14 Hours
