Graph rewriting processing method and device, computing equipment and readable medium
A technology of operators and original graphs, applied in the direction of neural architecture, biological neural network models, etc., can solve the problems of high development cost, low development efficiency, time-consuming and labor-intensive, etc., and achieve the effect of reducing research and development costs, improving efficiency, and strong flexibility
Pending Publication Date: 2020-06-26
XILINX INC
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] However, due to the current hard-coded implementation, developers need to manually code to match all the provided optimization modes when implementing graph rewriting, which is not only time-consuming and laborious, but also has a high error rate, resulting in high development costs and low development efficiency.
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0068] Preferred embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although preferred embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
[0069] figure 2 It is a flow chart of Embodiment 1 of the processing method for graph rewriting of the present invention. Such as figure 2 As shown, the processing method for graph rewriting in this embodiment may specifically include the following steps;
[0070] S100. Analyzing the neural network model of the front-end of the network deployment framework, and obtaining an original graph corresponding to the neural network model, the o...
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
Login to View More Abstract
The invention provides a graph rewriting processing method and device, computing equipment and a readable medium. The method comprises the steps: analyzing a neural network model at the front end of anetwork deployment framework, obtaining an original graph corresponding to the neural network model, wherein the original graph comprises a plurality of single-operation operator operators; and rewriting the original graph by using a pre-generated graph rewriting interface to obtain a rewritten graph supported and processed by the rear end of the network deployment framework, wherein the optimization operator is a combination of at least two adjacent single-operation operators in the original graph. According to the technical scheme, graph rewriting can be automatically realized, so that thecalculation / compiling capability of the rear end can be utilized to the greatest extent, the optimization operation of the neural network model is realized, research personnel do not need to compile codes in the realization process, time and labor are saved, the research cost is effectively reduced, and the research efficiency is improved.
Description
technical field [0001] The invention relates to the field of computer application technology, in particular to a graph rewriting processing method and device, computing equipment and a readable medium. Background technique [0002] In recent years, with the development of computer technology, artificial intelligence (AI for short) has entered a development boom, and the research on neural network models is the most popular. [0003] For example, a convolutional neural network (CNN for short) is the most common network model among neural network models. CNN is a feed-forward neural network consisting of one or more convolutional layers and a fully connected layer at the top (corresponding to a classic neural network), as well as associated weights and pooling layers. Convolutional neural networks can give better results in image and speech recognition than other deep learning structures. Compared with other deep, feed-forward neural networks, convolutional neural networks n...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06N3/04
CPCG06N3/045
Inventor 李天平孙晓明
Owner XILINX INC