Check patentability & draft patents in minutes with Patsnap Eureka AI!

Cross-platform deployment method for deep neural network model

A deep neural network, cross-platform technology, applied in the field of cross-platform deployment of deep neural network models, can solve problems such as inconvenient platform data extraction, unstable data reading, and insufficient deployment, to maintain accuracy and ensure normal effect of transmission, improved accuracy

Pending Publication Date: 2021-04-09
中科视拓(南京)科技有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The present invention provides a method for cross-platform deployment of a deep neural network model, which can effectively solve the problem that the cross-platform deployment of deep neural network models currently on the market is not rigorous enough in the above background technology, resulting in data loss and data deployment failure, which cannot facilitate the platform. Direct data extraction has caused the problem of data reading instability

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
  • Cross-platform deployment method for deep neural network model
  • Cross-platform deployment method for deep neural network model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] Such as Figure 1-2 As shown, the present invention provides a technical solution, a method for cross-platform deployment of a deep neural network model, including a deployment system and deployment steps;

[0032] The deployment system includes a connection module, a terminology specification module, a network security module, a deployment module and a power supply module;

[0033] The connection module, the term specification module, the network security module and the deployment module are all connected to the power supply module by signal;

[0034] The deployment steps include the following steps:

[0035] S1. First, perform data connection by connecting the deep neural network model related to the module, and determine the normal compatibility of its data;

[0036] S2, then define and determine the deep neural network model through the term module to prevent ambiguity;

[0037] S3, and then transfer the data of the deep neural network model through the network s...

Embodiment 2

[0051] Such as Figure 1-2 As shown, the present invention provides a technical solution, a method for cross-platform deployment of a deep neural network model, including a deployment system and deployment steps;

[0052] The deployment system includes a connection module, a terminology specification module, a network security module, a deployment module and a power supply module;

[0053] The connection module, the term specification module, the network security module and the deployment module are all connected to the power supply module by signal;

[0054] The deployment steps include the following steps:

[0055] S1. First, perform data connection by connecting the deep neural network model related to the module, and determine the normal compatibility of its data;

[0056] S2, then define and determine the deep neural network model through the term module to prevent ambiguity;

[0057] S3, and then transfer the data of the deep neural network model through the network s...

Embodiment 3

[0071] Such as Figure 1-2 As shown, the present invention provides a technical solution, a method for cross-platform deployment of a deep neural network model, including a deployment system and deployment steps;

[0072] The deployment system includes a connection module, a terminology specification module, a network security module, a deployment module and a power supply module;

[0073] The connection module, the term specification module, the network security module and the deployment module are all connected to the power supply module by signal;

[0074] The deployment steps include the following steps:

[0075] S1. First, perform data connection by connecting the deep neural network model related to the module, and determine the normal compatibility of its data;

[0076] S2, then define and determine the deep neural network model through the term module to prevent ambiguity;

[0077] S3, and then transfer the data of the deep neural network model through the network s...

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 deep neural network model cross-platform deployment method, and the method comprises a deployment system and a deployment step; the deployment system comprises a connection module, a term specification module, a network security module, a deployment module and a power supply module, wherein the connection module, the term specification module, the network security module and the deployment module are in signal connection with the power supply module. The method is scientific and reasonable in structure and safe and convenient to use, the term specification module can better determine terms, the accuracy of the terms is greatly kept, the network security module ensures normal transmission of data, the deployment module can better deploy and transfer the degree neural network model, and the security and the accuracy of normal reading of later data are greatly improved.

Description

technical field [0001] The invention relates to the technical field of deep neural network models, in particular to a cross-platform deployment method for deep neural network models. Background technique [0002] In supervised learning, the problem of the previous multi-layer neural network is that it is easy to fall into local extreme points. If the training samples are enough to fully cover the future samples, the learned multi-layer weights can be used to predict new test samples. But it is difficult to get enough labeled samples for many tasks. In this case, simple models such as linear regression or decision trees can often get better results than multi-layer neural networks (better generalization, worse Training error), in unsupervised learning, there was no effective way to construct a multi-layer network in the past. The top layer of a multi-layer neural network is a high-level representation of the underlying features, such as the bottom layer is pixels, and the nod...

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
IPC IPC(8): G06K9/62G06N3/08H04L29/06
CPCG06N3/08H04L63/1483G06F18/241
Inventor 王梓丞
Owner 中科视拓(南京)科技有限公司
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com