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
Introductory Insights into the Huawei Ascend Platform
- Perspective on Ascend architecture and its broader ecosystem
- Summary of MindSpore and CANN functionalities
- Practical applications and sector-specific importance
Preparing the Development Setup
- Deployment of the CANN toolkit and MindSpore
- Leveraging ModelArts and CloudMatrix for project management
- Validating the environment using example models
Creating Models with MindSpore
- Defining and training models within MindSpore
- Managing data pipelines and structuring datasets
- Converting models into Ascend-compatible formats
Enhancing Performance on Ascend
- Applying operator fusion and developing custom kernels
- Managing tiling strategies and AI Core allocation
- Utilizing benchmarking and profiling utilities
Deployment Methodologies
- Balancing the pros and cons of edge versus cloud implementation
- Employing the MindX SDK for rollout
- Aligning with CloudMatrix operational flows
Troubleshooting and Oversight
- Applying Profiler and AiD for trace analysis
- Resolving runtime errors
- Tracking resource consumption and data throughput
Case Studies and Lab Exercises
- End-to-end pipeline creation using MindSpore
- Practical session: Construct, refine, and implement a model on Ascend
- Performance analysis compared against alternative platforms
Conclusion and Forward-Looking Actions
Requirements
- Proficiency in neural networks and AI processing flows
- Proficiency in Python coding
- Awareness of model training and implementation pipelines
Target Audience
- AI engineers
- Data scientists engaged with the Huawei AI ecosystem
- ML developers utilizing 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
