Neural network model quantification method and device and electronic equipment

A neural network model and technology of electronic equipment, applied in the field of machine learning, can solve problems such as high computational complexity and inability to run target equipment

Pending Publication Date: 2020-07-10
MEGVII BEIJINGTECH CO LTD
View PDF0 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the existing quantized neural network model still has floating-point parameters in the normalization layer, so it still needs a floating-point unit and cannot run on target devices that only support fixed-point or low-bit

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
  • Neural network model quantification method and device and electronic equipment
  • Neural network model quantification method and device and electronic equipment
  • Neural network model quantification method and device and electronic equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] First, refer to figure 1 An example electronic device 100 for implementing a neural network model quantization method, apparatus and electronic device according to an embodiment of the present invention will be described.

[0036] like figure 1 Shown is a schematic structural diagram of an electronic device. The electronic device 100 includes one or more processors 102, one or more storage devices 104, an input device 106, an output device 108, and an image acquisition device 110. These components pass through a bus system 112 and / or other forms of connection mechanisms (not shown). It should be noted that figure 1 The components and structure of the electronic device 100 shown are only exemplary, not limiting, and the electronic device may also have other components and structures as required.

[0037] The processor 102 can be implemented in at least one hardware form of a digital signal processor (DSP), a field programmable gate array (FPGA), and a programmable log...

Embodiment 2

[0044] This embodiment provides a method for quantifying a neural network model, which can be executed by the above-mentioned electronic equipment such as a computer, see figure 2 The flow chart of the method for quantifying the neural network model is shown, the method mainly includes the following steps S202 to S208:

[0045] Step S202, obtaining a neural network model; wherein, the neural network model includes a convolutional layer, a normalization layer, and a quantized activation layer, and the output features of the quantized activation layer are integer features.

[0046] The above-mentioned neural network model can be a model that completes the training of a quantized neural network model, and the neural network model can be a convolutional neural network, wherein the convolutional layer, normalization layer, and quantization activation layer in the convolutional neural network network layer are sequentially connection, since there are still floating-point parameters...

Embodiment 3

[0085] On the basis of the foregoing embodiments, this embodiment provides an example of applying the foregoing neural network model quantification method to quantify the convolutional neural network, see e.g. image 3 The flow chart of convolutional neural network quantization shown in the figure can be executed with reference to the following steps S302 to S306:

[0086] Step S302: During the iterative training of the convolutional neural network, the parameters of the target network layer of the convolutional neural network are quantized until the iterative training ends, and a trained convolutional neural network is obtained.

[0087] The above-mentioned target network layer includes a convolutional layer, a normalization layer, and a quantized activation layer, and may also include other network layers. During the training phase of a convolutional neural network, see e.g. Figure 4 The flow chart of parameter dumping and deployment of the convolutional neural network is ...

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 neural network model quantification method and device, and electronic equipment, and relates to the technical field of machine learning, and the method comprises the steps: obtaining a neural network model, wherein the neural network model comprises a convolution layer, a normalization layer and a quantization activation layer, and the output feature of the quantization activation layer is an integer type feature; converting the parameters of the convolution layer into integer type parameters; combining the normalization layer and the quantization activation layer to obtain a combined layer; and converting the parameters of the merging layer into integer type parameters to obtain a quantized neural network model. According to the method, the operation complexity ofthe neural network model is reduced, so that the quantized neural network model can be accelerated to run on target equipment only supporting fixed-point or low-bit-width operation.

Description

technical field [0001] The invention relates to the technical field of machine learning, in particular to a neural network model quantification method, device and electronic equipment. Background technique [0002] At present, the neural network model has been widely and successfully applied in many fields such as speech recognition, text recognition, and image and video recognition. In some specified target tasks, the trained neural network model needs to be further deployed to the target device for acceleration. Since the general neural network model is a double-precision or single-precision floating-point number operation, in order to make as many target devices as possible satisfy the operation To meet the demand, the researchers quantified the neural network model to reduce the computational complexity and computational unit overhead of the neural network model. However, the existing quantized neural network model still has floating-point parameters in the normalizatio...

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): G06N3/08G06N3/04
CPCG06N3/04G06N3/08
Inventor 周舒畅林大超李翔张志华杨弋王田
Owner MEGVII BEIJINGTECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap