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

Classification method based on pruning convolutional neural network and related equipment

A technology of convolutional neural network and classification method, applied in the field of classification method and related equipment based on pruning convolutional neural network, can solve the problem of time-consuming pruning process, and achieve the effect of saving time-consuming pruning

Pending Publication Date: 2022-08-09
际络科技(上海)有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In order to better analyze the sensitivity of each layer after pruning, it is necessary to perform fine-tuning training (fine-tuning training) on ​​the model after pruning. branch configuration, it is necessary to analyze the pruning sensitivity of all layers to determine unimportant tensors, and the pruning process takes a long time

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
  • Classification method based on pruning convolutional neural network and related equipment
  • Classification method based on pruning convolutional neural network and related equipment
  • Classification method based on pruning convolutional neural network and related equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048] In order to make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention. , not all examples. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0049] figure 1is a schematic flowchart of a classification method based on a pruned convolutional neural network provided by an embodiment of the present invention; such as figure 1 As shown, a classification method based on pruned convolutional neural network includes the following steps:

[0050] S101, acquiring a picture to be classified.

[0051] In this step, the picture to be classified is a picture of a ...

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 provides a classification method based on a pruning convolutional neural network and related equipment. The classification method comprises the following steps: acquiring a to-be-classified picture; inputting a to-be-classified picture into the pruned classification model to obtain a corresponding classification result; wherein the classification model weight in the pruned classification model is obtained by searching a pre-trained pruning sensitivity analysis model through a predefined neural network structure search method; the pre-trained pruning sensitivity analysis model is obtained by training based on an initial classification model, a training picture set, a label set corresponding to the training picture set and a weight mask training set. The pruning time can be greatly saved.

Description

technical field [0001] The invention relates to the technical field of model pruning, in particular to a classification method and related equipment based on a pruned convolutional neural network. Background technique [0002] Currently, deep learning models require a lot of computing power, memory, and electricity. When it is necessary to perform real-time inference, run models on the device side, and run deep learning models with limited computing resources, deep learning models with small size and high accuracy are required, so model compression can achieve this goal, model pruning It is a type of model compression. [0003] Model pruning is mainly used to reduce the amount of computation in the convolutional neural network. Usually, the purpose of reducing the computation amount of the entire neural network is to cut off the unimportant tensors in the weight of the neural network. Before pruning unimportant tensors, it is necessary to determine the sparsity rate of eac...

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): G06K9/62G06N3/04G06N3/08
CPCG06N3/082G06N3/084G06N3/045G06F18/214G06F18/2415
Inventor 陆强
Owner 际络科技(上海)有限公司