Get in Touch

Course Outline

Intro to the Huawei AI Ecosystem

  • Ascend AI processors: An overview of models 310, 910, and 910B
  • Key components: MindSpore, CANN, and AscendCL
  • Market positioning and core architectural tenets

The Function of CANN within Huawei's AI Stack

  • Defining CANN: Purpose of the SDK and its internal hierarchy
  • ATC, TBE, and AscendCL: Processes for compiling and running models
  • Supporting inference efficiency and deployment via CANN

Introduction to MindSpore and its Design

  • Workflows for training and inference within MindSpore
  • Graph mode, PyNative, and hardware abstraction layers
  • Connecting to Ascend NPUs through the CANN backend

AI Lifecycle on Ascend: From Training to Deployment

  • Building models in MindSpore or migrating from other frameworks
  • Exporting and compiling models utilizing ATC
  • Deploying on Ascend hardware via OM models and AscendCL

Benchmarking Against Other AI Ecosystems

  • MindSpore compared to PyTorch and TensorFlow: Objectives and market niche
  • Deployment processes on Ascend versus GPU-centric stacks
  • Prospects and constraints for enterprise adoption

Enterprise Integration Cases

  • Applications in smart manufacturing, government AI, and telecommunications
  • Considerations regarding scalability, regulatory compliance, and ecosystem health
  • Hybrid cloud/on-prem deployments leveraging the Huawei stack

Recap and Future Directions

Requirements

  • General knowledge of AI workflows or platform design
  • Foundational grasp of model training and deployment processes
  • No specific prior experience with CANN or MindSpore is necessary

Target Audience

  • Professionals evaluating AI platforms and infrastructure architects
  • AI/ML DevOps specialists and pipeline integration experts
  • Technology leaders and strategic decision-makers
 14 Hours

Number of participants


Price per participant

Upcoming Courses

Related Categories