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

Memory allocation method, device, device, readable storage medium and program product

A memory allocation and memory technology, which is applied in the computer field to achieve good results, reduce memory usage, and solve memory bottlenecks.

Active Publication Date: 2022-02-25
TENCENT TECH (SHENZHEN) CO LTD
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In view of the shortcomings of the existing methods, the present application proposes a memory allocation method, device, equipment, computer-readable storage medium and computer program products to solve the problem of how to reduce the memory usage in the reasoning process of the model

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
  • Memory allocation method, device, device, readable storage medium and program product
  • Memory allocation method, device, device, readable storage medium and program product
  • Memory allocation method, device, device, readable storage medium and program product

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0046] Embodiments of the present application are described below with reference to the drawings in the present application. It should be understood that the implementation manner described below in conjunction with the accompanying drawings is an exemplary description for explaining the technical solutions of the embodiments of the present application, and does not limit the technical solutions of the embodiments of the present application.

[0047] Those skilled in the art will understand that unless otherwise stated, the singular forms "a", "an", "said" and "the" used herein may also include plural forms. It should be further understood that the terms "comprising" and "comprising" used in the embodiments of the present application mean that the corresponding features can be implemented as the presented features, information, data, steps, operations, elements and / or components, but do not exclude The realization is other features, information, data, steps, operations, elemen...

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

Embodiments of the present application provide a memory allocation method, device, device, computer-readable storage medium, and computer program product, and relate to the field of artificial intelligence. The method includes: obtaining a weight file of at least one target model; According to the graph class, determine the memory allocation information of the session class corresponding to any target model in at least one target model; according to the memory allocation information of the session class corresponding to any target model, in any target model During the inference process, the memory of the session class corresponding to any target model is reused. The embodiment of the present application realizes the multiplexing of the memory of the session class corresponding to any target model, thereby reducing the memory occupation in the reasoning process of any target model.

Description

technical field [0001] The present application relates to the field of computer technology, specifically, the present application relates to a memory allocation method, device, equipment, computer-readable storage medium and computer program product. Background technique [0002] With people's continuous exploration in the field of deep learning, deep learning models are developing in two directions. One aspect is that the model becomes larger and larger, and the other aspect is that the proportion of device-side deployment increases. For server deployment, larger deep learning models require more memory during inference; for mobile terminal deployment, mobile terminals have limited memory during inference; the above two phenomena often cause memory to become a bottleneck, resulting in server-side concurrency Problems such as data reduction, idle computing power, and the inability to deploy larger models on mobile terminals. Quantization schemes are used in the prior art, a...

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 Patents(China)
IPC IPC(8): G06F9/50G06N3/04G06N3/08G06N5/04
CPCG06F9/5016G06F9/5022G06N3/0418G06N3/08G06N5/04G06F2209/5011
Inventor 杨伟光
Owner TENCENT TECH (SHENZHEN) CO LTD