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
Introducing Huawei’s AI Landscape
- Overview of Ascend AI hardware: 310, 910, and 910B
- Key components: MindSpore, CANN, and AscendCL
- Strategic market positioning and core architecture principles
CANN’s Function within Huawei’s AI Ecosystem
- Defining CANN: SDK objectives and internal structure
- ATC, TBE, and AscendCL: Processes for compiling and running models
- CANN’s role in enhancing inference optimization and deployment
MindSpore Architecture and Overview
- Handling training and inference workflows in MindSpore
- Graph mode, PyNative, and hardware abstraction layers
- Connecting with the Ascend NPU through the CANN backend
Managing the AI Lifecycle on Ascend: From Training to Release
- Building models in MindSpore or converting them from alternative frameworks
- Exporting and compiling models via ATC
- Deploying on Ascend hardware using OM models and AscendCL
Benchmarking Against Other AI Stacks
- MindSpore versus PyTorch and TensorFlow: Distinct focuses and market positions
- Deployment processes on Ascend compared to GPU-based alternatives
- Potential benefits and constraints for enterprise application
Enterprise Application Scenarios
- Applications in smart manufacturing, governmental AI initiatives, and telecommunications
- Considerations regarding scalability, regulatory compliance, and ecosystem health
- Hybrid cloud and on-premise deployments utilizing the Huawei stack
Recap and Future Directions
Requirements
- General awareness of AI workflows or platform structures
- Fundamental knowledge of model training and deployment processes
- No previous practical experience with CANN or MindSpore is necessary
Target Audience
- AI platform assessors and infrastructure designers
- AI/ML DevOps specialists and pipeline integration experts
- Technology leaders and key decision-makers
