Network model deployment method, device and equipment and readable storage medium

Abstract

The invention discloses a network model deployment method, apparatus and device, and a computer readable storage medium. The method comprises: obtaining a network model file obtained by adopting a target framework; determining a target analysis mode according to the target framework, and analyzing the network model file according to the target analysis mode to obtain an intermediate file; obtaining a structure file by utilizing the intermediate file, and performing format conversion processing on the structure file to obtain a standard structure file and a control file matched with the FPGA device; acquiring a plurality of operator codes, and generating a bit stream file by utilizing the plurality of operator codes, the standard structure file and the control file; and storing the bit stream file in FPGA equipment, and completing the deployment of the network model on the FPGA equipment. According to the method, through two times of conversion, rapid deployment of the network model ofany architecture on the FPGA device can be realized, and the deployment flexibility and deployment speed are improved.

Description

technical field [0001] The present invention relates to the technical field of deep learning, and in particular to a network model deployment method, a network model deployment device, a network model deployment device, and a computer-readable storage medium. Background technique [0002] In order to improve the data processing speed of the neural network model, many manufacturers deploy the neural network model on the FPGA device and use the FPGA device for high-speed calculation. Related technologies After deploying the neural network model under a certain framework to the FPGA device, if you want to deploy the neural network model under other frameworks, you need a lot of other work to make the FPGA device support the neural network model under the new framework, for example, you may need to Make adjustments to the FPGA device. This makes related technologies less flexible and less efficient when deploying neural network models. [0003] Therefore, how to solve the prob...

AI Technical Summary

Problems solved by technology

[0004] In view of this, the purpose of the present invention is to provide a network model deployment method, a network model deployment device, a network model deployment device, and a computer-readable storage medium, which solve the problem of poor deployment flexibility and slow deployment speed in related technologies. question

Images