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
- Examination of Ascend architectural components and ecosystem structure
- Fundamentals of MindSpore and the CANN software stack
- Applicable use cases and their significance in sectoral contexts
Establishing the Development Environment
- Deployment of the CANN toolkit and integration of MindSpore
- Utilizing ModelArts and CloudMatrix for the orchestration of projects
- Verification of environment stability using standard sample models
Model Development via MindSpore
- Defining architectures and executing training processes within MindSpore
- Configuration of data pipelines and formatting of datasets
- Exporting trained models to Ascend-compatible formats
Performance Optimization for Ascend
- Implementation of operator fusion and development of custom kernels
- Application of tiling strategies and management of AI Core scheduling
- Utilization of benchmarking and profiling instruments
Deployment Methodologies
- Assessment of tradeoffs between edge and cloud deployment models
- Execution of deployment tasks using the MindX SDK
- Integration with CloudMatrix operational workflows
Debugging and System Monitoring
- Application of Profiler and AiD tools for trace analysis
- Identification and resolution of runtime failures
- Oversight of resource consumption and throughput metrics
Case Studies and Laboratory Integration
- Development of a complete pipeline leveraging MindSpore capabilities
- Laboratory exercise: construction, optimization, and deployment of a model on Ascend
- Comparative performance analysis across alternative platforms
Summary and Subsequent Actions
Requirements
- Foundational knowledge 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
- Machine learning developers working with Ascend and MindSpore for government operations
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