Method, device and computer-readable storage medium for processing operation instructions

Abstract

The embodiment of the present invention discloses an operation command processing method, device and storage medium, including: the server obtains the operation command for the training environment built in its own learning platform; judges whether the obtained operation command is a parameter adjustment command; if The obtained operation instruction is a parameter adjustment instruction, and the training environment is not closed, and the training is performed according to the target parameter configuration information expected to be adjusted in the parameter adjustment instruction. It can be seen from the technical solution provided by the embodiment of the present invention that since the server has pre-judged whether the operation instruction is a parameter adjustment instruction, if it is determined to be a parameter adjustment instruction, it will control the training environment not to close, so when there is training parameter information caused by training requirements In the case of being modified, the restart of the training environment is avoided, thereby saving the time spent on actual training and improving user experience.

Description

technical field [0001] The embodiments of the present invention relate to the field of computer technology, and in particular, to a method, device, and computer-readable storage medium for processing operation instructions. Background technique [0002] The deep learning platform is a platform based on powerful graphics processing unit (GPU) computing resources and equipped with multiple mainstream deep learning frameworks to quickly build a deep learning development environment and comprehensively speed up application development. Existing deep learning platforms are often comprehensive platforms integrating development environment creation, training environment creation, and actual training. [0003] In related technologies, it is often necessary to pre-set corresponding training parameter information in the created training environment during actual training, and then perform actual training on these parameter information. When the training result does not meet the prese...

AI Technical Summary

Problems solved by technology

[0004] However, this method is a waste of time because the training environment needs to be restarted when modifying the training parameter information, and the user experience is poor.

Images