Machine learning parameter configuration method for multi-level users

• Main Authors: ZHANG XUFENG, LI BOCHANG, XUE YUAN, YU HONGFENG, CHENG GANG, DIAO WENHUI, CHU NING, WANG QIANG, LIU BIXIN
• Format: Patent
• Language: Chinese; English
• Published: 27.10.2020
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The method mainly solves the problem that in the application process of the machine learning application program, due to users with zero basis, a little basis and professional level users share the application, the machine learning application program needs to meet the requirements of multi-level users. The method comprises the following steps: (1) in a machine learning stage, aiming at a plurality of typical scenes, adapting data of each scene to machine learning, and determining optimal parameter configuration of each scene; (2) in a parameter sorting stage, sorting each group of optimal parameter configurations in an ascending order according to a frequency-based strategy to form an ordered optimal parameter set; and (3) in an application running stage, for an application scene, providing a default parameter mode, a single parameter configuration mode and a detailed parameter configuration mode for a user to support the running of the application program, and putting the optimal parameter obtained in the d