Get in Touch

Course Outline

Introduction to the Huawei Ascend Platform

  • Overview of Ascend architecture and ecosystem
  • Introduction to MindSpore and CANN
  • Real-world use cases and industry relevance

Setting Up the Development Environment

  • Installation of the CANN toolkit and MindSpore
  • Utilizing ModelArts and CloudMatrix for project orchestration
  • Validating the environment using sample models

Model Development with MindSpore

  • Defining and training models in MindSpore
  • Constructing data pipelines and formatting datasets
  • Exporting models to an Ascend-compatible format

Performance Optimization on Ascend

  • Implementing operator fusion and custom kernels
  • Employing tiling strategies and AI Core scheduling
  • Leveraging benchmarking and profiling tools

Deployment Strategies

  • Evaluating trade-offs between edge and cloud deployment
  • Utilizing the MindX SDK for deployment
  • Integrating with CloudMatrix workflows

Debugging and Monitoring

  • Using Profiler and AiD for tracing
  • Resolving runtime failures
  • Monitoring resource usage and throughput

Case Study and Lab Integration

  • Executing full pipeline development using MindSpore
  • Lab Activity: Build, optimize, and deploy a model on Ascend
  • Comparing performance against other platforms

Summary and Next Steps

Requirements

  • A solid understanding of neural networks and AI workflows
  • Proficiency in Python programming
  • Familiarity with model training and deployment pipelines

Target Audience

  • AI engineers
  • Data scientists utilizing the Huawei AI stack
  • ML developers employing Ascend and MindSpore
 21 Hours

Testimonials (1)

Related Categories