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 the Huawei Ascend Platform
- Comprehensive view of Ascend architecture and its ecosystem
- Brief introduction to MindSpore and CANN
- Practical use cases and their industry significance
Establishing the Development Environment
- Installation of the CANN toolkit and MindSpore frameworks
- Leveraging ModelArts and CloudMatrix for efficient project coordination
- Validating the setup using sample models
Model Development using MindSpore
- Defining and training models within the MindSpore framework
- Managing data pipelines and preparing datasets
- Converting models into Ascend-compatible formats
Performance Optimisation on Ascend
- Implementing operator fusion and custom kernels
- Applying tiling strategies and AI Core scheduling techniques
- Utilising benchmarking and profiling instruments
Deployment Methodologies
- Weighing the benefits and constraints of edge versus cloud deployment
- Executing deployments with the MindX SDK
- Integrating with CloudMatrix operational workflows
Debugging and Monitoring Practices
- Employing Profiler and AiD tools for detailed tracing
- Troubleshooting runtime issues and failures
- Tracking resource consumption and throughput metrics
Case Studies and Practical Integration
- End-to-end pipeline development leveraging MindSpore
- Practical session: Construct, optimise, and launch a model on Ascend hardware
- Comparative performance analysis against alternative platforms
Conclusions and Future Directions
Requirements
- A solid grasp of neural networks and AI operational workflows
- Proficiency in Python programming
- Knowledge of model training and deployment processes
Target Audience
- AI Engineers
- Data Scientists utilising the Huawei AI stack
- ML Developers working with Ascend and MindSpore
21 Hours
Testimonials (1)
That i gained a knowledge regarding streamlit library from python and for sure i'll try to use it to improve applications in my team which are made in R shiny
