The hottest Huawei cloud AI development platform m

  • Detail

Huawei cloud AI development platform modelarts received the first batch of group standard compliance test certificates of artificial intelligence

recently, DPM also developed a new thermal conductive polymer in the Zytel series through the test of Saixi Laboratory of China Institute of electronic technology standardization, Huawei cloud AI development platform modelarts passed the compliance test of "information technology AI machine learning oriented system framework and functional requirements", one of the first AI group standards of China Electronics Industry Standardization Technology Association, and was awarded the AI product test certificate

the system framework and functional requirements for machine learning in information technology artificial intelligence is one of the first group standards for artificial intelligence jointly prepared by more than 30 enterprises, universities and research institutes, including Huawei Technology Co., Ltd., led by the national AI standardization master group planning and the China Institute of electronic technology standardization

this standard gives the system framework for machine learning, specifies the functional requirements of the whole system and each component, is applicable to the planning and design of artificial intelligence systems and solutions in various fields, and can be used as the basis for evaluation, selection and acceptance

this standard has very high requirements for product capability, and it is very difficult to pass the test. The standard puts forward a total of 81 requirements in 17 major aspects, covering the core capabilities of machine learning, multi algorithm management, heterogeneous resource scheduling, etc. with a history of only more than 100 years, and provides the end-to-end capabilities of data preprocessing, feature engineering, model development, model training, and model reasoning service release in the middle of the measurement mark at the fracture

machine learning oriented AI system should support AI applications in different scenarios in various fields in terms of machine learning engine, model base, data management, heterogeneous resource pool, distributed computing scheduling, algorithm service, operation and maintenance management and interface

as a one-stop AI development platform for developers, Huawei cloud modelarts has passed the test in the first batch and obtained the authoritative AI product test certificate with its excellent comprehensive ability

modelarts platform provides massive data preprocessing and semi-automatic annotation capabilities for machine learning and deep learning, large-scale distributed training, automatic model generation and other model capabilities, as well as end edge cloud model on-demand deployment capabilities to help users quickly create and deploy models and manage full cycle AI workflows

at the same time, in the fourth international image recognition competition webvision held recently, Huawei cloud EI trained large-scale image classification models and advanced distributed training methods based on modelarts, shortened the training time of super large-scale data sets, and defeated 94 teams from all over the world to win the championship with an accuracy of 82.97%

webvision competition is jointly organized by ETH Zurich, Google research, Carnegie Mellon University (CMU), etc. it is one of the most authoritative competitions in the field of image recognition at present, and is honored as the artificial intelligence world cup by the industry

in order to enable the ecosystem and further accelerate the development and implementation of AI products, Huawei cloud EI also released a developer ecological community based on modelarts, which is currently helping to build the next generation space system. The ecological community has 100000 + developers and cooperates with more than 50 universities

in addition, Huawei cloud EI has been gradually applied in 600 + projects in 10 industries, such as medical treatment, Internet, smart city, construction, parks, etc

Copyright © 2011 JIN SHI