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 7 hours
Course Outline
Overview of Huawei’s AI Ecosystem
- Ascend AI hardware specifications: 310, 910, and 910B chips
- MindSpore, CANN, and their supporting toolset
- The AI development workflow: progressing from training to deployment
Comprehending the CANN Toolkit
- Definition of CANN and its significance in the AI stack
- Summary of core components, including ATC, AscendCL, and operator libraries
- The specific role CANN plays within AI inference pipelines
Initial Steps with MindSpore and CANN
- Environment setup procedures (integrating MindSpore, CANN, and Python)
- Training a fundamental model using MindSpore
- Model export and conversion processes utilizing ATC
Executing Inference on Ascend Devices
- Utilizing OM models via AscendCL or Python APIs
- Handling basic input and output preprocessing
- Verification and validation of model outputs
Integration with Other Frameworks
- Support overview for TensorFlow, PyTorch, and ONNX
- Analysis of supported operators and existing limitations
- Demonstration of simple model conversion (e.g., from ONNX to OM)
Exploring the CANN and MindSpore Developer Community
- Essential resources: official documentation, GitHub repositories, and sample code
- Introduction to MindSpore Hub and the available model zoo
- Access to community forums, events, and support channels
Summary and Recommended Next Steps
Requirements
- A foundational understanding of core machine learning and deep learning principles
- Basic programming proficiency in Python
- No previous experience with CANN or Ascend hardware is necessary
Target Audience
- Machine learning developers looking to explore deployment workflows
- Students or researchers new to the Huawei AI ecosystem
- AI framework contributors and enthusiasts interested in model acceleration techniques
