Unlock instant, AI-driven research and patent intelligence for your innovation.

Cache optimization method and device applied to deep learning network

A deep learning network and cache optimization technology, applied in the computer field, can solve problems such as low cache optimization efficiency, and achieve the effect of improving cache optimization efficiency

Active Publication Date: 2018-10-02
HANGZHOU HIKVISION DIGITAL TECH
View PDF9 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problem of low cache optimization efficiency in each layer of deep learning network training by proposing a cache optimization method and device applied to deep learning networks

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Cache optimization method and device applied to deep learning network
  • Cache optimization method and device applied to deep learning network
  • Cache optimization method and device applied to deep learning network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] In order to make the purpose, technical solution and advantages of the present application clearer, the following will clearly and completely describe the technical solution of the present application in combination with specific implementation methods of the present application and corresponding drawings. Apparently, the described implementations are only some of the implementations of this application, not all of them. Based on the implementation manners in this application, all other implementation manners obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.

[0036] The technical solutions provided by various embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0037] Such as Figure 1a Shown is a schematic diagram of the traditional cache usage method, where the horizontal axis is the time axis; from Figure 1a In the...

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

No PUM Login to View More

Abstract

The invention discloses a cache optimization method and device applied to a deep learning network. The cache optimization method includes the following steps that: analog operation is performed on theN-th layer of the deep learning network; and whether a first predetermined cache block is occupied is detected after the simulation operation is performed, if the first predetermined cache block is occupied, a second predetermined cache block is allocated for the output data of the N-th layer, and when a preset condition is met, the occupied first predetermined cache block is released. With the cache optimization method and device applied to the deep learning network of the invention adopted, the problem of the complexity of cache optimization during the training of each layer of the deep learning network can be solved, and cache optimization and allocation can be simpler and more efficient. The method and device can be applied to different networks.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a cache optimization method applied to a deep learning network and a device corresponding to the method. Background technique [0002] Deep learning is a new field in machine learning research, and it is a relatively effective artificial intelligence method. By simulating the learning behavior of the human brain, it learns relevant knowledge from the data for subsequent predictions. Deep learning uses so-called "networks" for learning, where the network is composed of multiple "layers" (e.g., convolutions), each layer is trained on the output of the previous layer (or previous few convolutions) as input, and then Use the training result as the input of the next layer. The training process is the process of orderly computing each layer. [0003] In the training of deep learning, a large amount of intermediate data will be generated. For training needs, these data usually need...

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
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06N99/00G06N3/08
CPCG06N3/08G06N99/00
Inventor 郑星王鹏叶挺群彭剑峰周智强
Owner HANGZHOU HIKVISION DIGITAL TECH