Convolutional neural network acceleration method and system, terminal and storage medium

Abstract

The invention provides a convolutional neural network acceleration method and system, a terminal and a storage medium. The method comprises the following steps: generating an RISC-V processor soft core by using a source code generator; constructing an RISC-V single core by setting an extended DMA of the RISC-V processor soft core, a memory controller and a distributed memory module; constructing amany-core acceleration array with a preset specification by utilizing the RISC-V single core; and accessing the many-core acceleration array to a convolutional neural network system, wherein the convolutional neural network system comprises a main processor and convolutional neural network hardware. According to the invention, the memory access bandwidth in the calculation process can be greatlyimproved, the memory access delay is reduced, the calculation performance of the convolutional neural network is improved, and the calculation acceleration of the convolutional neural network is realized.

Description

technical field [0001] The present invention relates to the technical field of convolutional neural networks, in particular to a convolutional neural network acceleration method, system, terminal and storage medium. Background technique [0002] With the advent of the era of big data, massive data has shown exponential explosive growth with the improvement of computer performance, and various deep learning algorithms represented by convolutional neural networks have been widely used. However, based on the neural network hierarchy and convolutional computing structure, the huge amount of calculations and parameters brought about have increasingly become the performance bottleneck of convolutional neural networks, especially the large number of parameter storage and memory access delays have become computing bottlenecks. Contents of the invention [0003] Aiming at the above shortcomings of the prior art, the present invention provides a convolutional neural network accelera...

AI Technical Summary

Problems solved by technology

However, based on the hierarchical and convolutional computing structure of the neural network, the huge amount of calculations and parameters brought about have increasingly become the performance bottleneck of the convolutional neural network, especially the large number of parameter storage and memory access delays have become computing bottlenecks.

Images