| With the digital and intelligent transformation on a global scale.Emerging technologies and industries such as cloud computing and artificial intelligence have emerged at the historic moment,which has further ushered in a new round of inforrmation reform based on cloud computing,artificial intelligence as the core and big data as the key.Artificial intelligence has the characteristics of a high degree of integration of social and technological attributes.Artificial intelligence provides huge development potential to various industries such as agriculture,education,and medical care,and has a profound and extensive impact on the development and advancement of society.With the increasing reliance on artificial intelligence technology in various fields,society’s requirements for the core qualities of artificial intelligence talents are constantly increasing.Improving the training plan of artificial intelligence and increasing investment in artificial education is the key to training high-quality and high-quality artificial intelligence talents.In view of the shortage of high-quality artificial intelligence talents in China,universities should talk about the training of artificial intelligence talents as the core tasks and important future plans of universities.Artificial intelligence teaching has the characteristics of close integration of theory and practice.In order to solve the problems of low utilization of hardware resources and complex software installation in the implementation of artificial intelligence experiments in universities,this paper designs and implements a cloud computing-based artificial intelligence teaching platform.This article studies the virtualization,resource scheduling technology and related principles in cloud computing,combines single sign-on to achieve user authentication,and develops users’ online experiment interfaces based on Jupyter.Design and implement a cloud computing-based artificial in telligence teaching platform.The work of this paper is mainly divided into three parts.The first part is the design and implementation of an online experiment module.Based on the Docker container virtualization technology,a unified,out-of-the-box artificial intelligence online experiment programming environment is realized.Using Docker Swarm container orchestration technology achieves the user’s application and scheduling of computing resources anytime,anywhere,and achieves the purpose of recycling users’ computing resources to the cluster after a fixed time.The second part is the design and implementation of the user data storage module.It investigates and compares multiple distributed file storage technologies,and finally selects NFS as the storage method for user training data,experimental code,and training results.The third part is the design and implementation of the user authentication network management module.Using the OAuth2.0 protocol in the single sign-on scheme and the Python Django web development framework,the OAuth2.0 protocol authentication server is established as the user authentication gateway for the user experiment interface Jupyter Lab.The practical use of the talented classes of the School of Computer of Guangxi Normal University in artificial intelligence teaching proves that the cloud computing-based artificial intelligence teaching platform designed and implemented in this paper has good practical value and application prospects. |
