Network division method and system, electronic equipment and storage medium
A network division and point division technology, applied in the field of deep learning, can solve problems such as long distance, large transmission delay, and benefit concealment
Active Publication Date: 2019-06-18
INSPUR SUZHOU INTELLIGENT TECH CO LTD
View PDF5 Cites 2 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, traditionally, due to the long distance between the mobile terminal and the cloud, the transmission delay is large, which obscures the benefits of utilizing cloud computing
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0059] The following will clearly and completely describe the technical solutions in the embodiments of the application with reference to the drawings in the embodiments of the application. Apparently, the described embodiments are only some of the embodiments of the application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of this application.
[0060] In the prior art, when the network scale is large, it takes a long time to perform reasoning only on the mobile terminal. An optional method is to perform deep learning inference on the cloud, that is, save the network model in the cloud, and when reasoning is required, the mobile terminal transmits the data to the cloud, and the cloud server performs deep learning inference and returns the inference result to the mobile terminal. However, traditionally, due to the long distance bet...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a network division method, system and device, and a medium, and the method comprises the steps: taking an exit point of a deep learning network as a preset division point, anddetermining the calculation time after the deep learning network is divided according to the preset division point; obtaining network bandwidths and delays of the edge end and the mobile end, and determining transmission time by using the bandwidths and the delays; obtaining the total reasoning time by using the calculation time and the transmission time; selecting the outlet point of the reasoning total time with the minimum time from the reasoning total time of all the outlet points, and taking the outlet point as a target division point; and dividing the deep learning network by using the target division point, enabling the edge end to calculate by using the front part of network, transmitting a calculation result to the mobile end, and calculating by using the rear part of network by using the mobile end to obtain an inference result. According to the method, the target division point is selected for the deep learning network by using the total reasoning time, the deep learning network is divided according to the target division point, calculation reasoning is performed by using the divided deep learning network, the reasoning time is shortened, and the efficiency is improved.
Description
technical field [0001] The present application relates to the technical field of deep learning, and more specifically, relates to a network division method and system, an electronic device, and a computer-readable storage medium. Background technique [0002] Currently, deep learning is mainly used in the cloud and mobile. Specifically, the cloud includes tens of thousands of server nodes with powerful computing capabilities that can be used for deep learning training and reasoning. The mobile terminal includes personal assistants, mobile phones and other mobile devices, which usually have weak computing power, and are generally only used for deep learning reasoning, not for deep learning training. [0003] When the network scale is large, it takes a long time to infer purely on the mobile side. An optional method is to perform deep learning inference on the cloud, that is, save the network model in the cloud, and when reasoning is required, the mobile terminal transmits t...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): H04W16/02H04W16/18
Inventor 李龙赵雅倩李仁刚
Owner INSPUR SUZHOU INTELLIGENT TECH CO LTD