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
\r\n Introduction to CV\/NLP Deployment with CANN\r\n<\/p>\r\n
- \r\n
- \r\n The lifecycle of AI models from training through to deployment\r\n <\/li>\r\n
- \r\n Critical performance factors for real-time CV and NLP applications\r\n <\/li>\r\n
- \r\n An overview of CANN SDK tools and their significance in model integration\r\n <\/li>\r\n<\/ul>\r\n
\r\n Preparing CV and NLP Models\r\n<\/p>\r\n
- \r\n
- \r\n Exporting models from PyTorch, TensorFlow, and MindSpore\r\n \r\n
- \r\n Managing model inputs\/outputs for image and text-based tasks\r\n \r\n
- \r\n Utilizing ATC to convert models into OM format\r\n \r\n<\/ul>\r\n
- \r\n Executing CV\/NLP inference via the AscendCL API\r\n \r\n
- \r\n Preprocessing workflows: image resizing, tokenization, and normalization\r\n \r\n
- \r\n Postprocessing tasks: bounding boxes, classification scores, and text output\r\n \r\n<\/ul>\r\n
- \r\n Profiling CV and NLP models using CANN utilities\r\n \r\n
- \r\n Minimizing latency through mixed-precision and batch tuning\r\n \r\n
- \r\n Managing memory and compute resources for streaming tasks\r\n \r\n<\/ul>\r\n
- \r\n Case study: object detection for intelligent surveillance\r\n \r\n
- \r\n Case study: visual quality inspection in manufacturing environments\r\n \r\n
- \r\n Creating live video analytics pipelines on Ascend 310\r\n \r\n<\/ul>\r\n
- \r\n Case study: sentiment analysis and intent detection\r\n \r\n
- \r\n Case study: document classification and summarization\r\n \r\n
- \r\n Real-time NLP integration with REST APIs and messaging systems\r\n \r\n<\/ul>\r\n
\r\n Deploying Inference Pipelines with AscendCL\r\n<\/p>\r\n
- \r\n
\r\n Performance Optimization Techniques\r\n<\/p>\r\n
- \r\n
\r\n Computer Vision Use Cases\r\n<\/p>\r\n
- \r\n
\r\n NLP Use Cases\r\n<\/p>\r\n
- \r\n
\r\n Summary and Next Steps\r\n<\/p>
Requirements
- \r\n
- \r\n Knowledge of deep learning principles in computer vision or NLP\r\n <\/li>\r\n
- \r\n Practical experience with Python and AI frameworks like TensorFlow, PyTorch, or MindSpore\r\n <\/li>\r\n
- \r\n Foundational understanding of model deployment or inference workflows\r\n <\/li>\r\n<\/ul>\r\n
\r\n Target Audience\r\n<\/p>\r\n
- \r\n
- \r\n Computer vision and NLP specialists working with Huawei’s Ascend platform\r\n <\/li>\r\n
- \r\n Data scientists and AI engineers developing real-time perception models\r\n <\/li>\r\n
- \r\n Developers integrating CANN pipelines in sectors such as manufacturing, surveillance, or media analytics\r\n <\/li>\r\n<\/ul>
14 Hours
