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.
Duration 21 hours
Course Outline
Foundations of Cambricon and MLU Architecture
- Introduction to Cambricon’s AI chip ecosystem.
- Deep dive into MLU architecture and instruction pipelines.
- Review of supported model types and applicable use cases.
Establishing the Development Toolchain
- Installation of BANGPy and the Neuware SDK.
- Configuring environments for Python and C++ development.
- Understanding model compatibility and preprocessing requirements.
Model Construction with BANGPy
- Managing tensor structures and shapes.
- Building computation graphs.
- Implementing custom operations within BANGPy.
Deployment via Neuware Runtime
- Processes for converting and loading models.
- Controlling execution and inference.
- Best practices for deploying to edge and data center devices.
Enhancing Performance
- Tuning memory mapping and layers.
- Utilizing execution tracing and profiling tools.
- Identifying and resolving common performance bottlenecks.
Integrating MLUs into Applications
- Leveraging Neuware APIs for seamless application integration.
- Supporting streaming and multi-model architectures.
- Handling hybrid CPU-MLU inference scenarios.
Capstone Project and Real-World Application
- Practical lab: Deploying a computer vision or NLP model.
- Implementing edge inference with BANGPy integration.
- Validating model accuracy and throughput.
Course Wrap-up and Path Forward
Requirements
- A solid grasp of machine learning model architectures.
- Proficiency in Python and/or C++ programming.
- Knowledge of model deployment strategies and hardware acceleration concepts.
Target Audience
- Embedded AI developers.
- ML engineers focused on edge or data center deployment.
- Developers working with Chinese AI infrastructure ecosystems.
Testimonials (1)
That we can cover advance topic and work with real-life example
