A fixed-point acceleration method and device for convolution calculation

Abstract

The invention provides a fixed-point acceleration method and device for convolution calculation, and relates to the technical field of convolution calculation. The fixed point acceleration method forconvolution calculation comprises the following steps: carrying out first conversion operation on a floating point matrix in a preset convolution algorithm to obtain a converted fixed point matrix; Replacing the floating point matrix in the preset convolution algorithm with a fixed point matrix to obtain a replaced convolution algorithm; performing convolution calculation according to the replacedconvolution algorithm and the input data to obtain a convolution calculation result; and performing a second conversion operation on the convolution calculation result to obtain a convolution calculation result of the input data. In the embodiment of the invention, the method comprises the following steps: obtaining a target object; replacing a floating point matrix in a preset convolution algorithm by a fixed point matrix to obtain a replaced convolution algorithm; the method does not comprise a floating point matrix any more, can be directly applied to fixed-point calculation, reduces multiplication in the convolution process, can effectively reduce access and calculation amount of data volume in the fixed-point convolution algorithm, and can improve the performance of convolution calculation.

Description

technical field [0001] The invention relates to the technical field of convolution calculations, in particular to a fixed-point acceleration method and device for convolution calculations. Background technique [0002] Convolutional Neural Networks (CNN) is a type of feed-forward neural network that includes convolution calculations and has a deep structure. It has a wide range of applications in machine vision, natural language processing, speech recognition, and AI technology applications in-depth application. [0003] The convolutional neural network contains a large number of convolution calculations. The convolution calculation is a time-consuming process for the CPU (Central Processing Unit, CPU). The whole process includes the process of data access and calculation, especially for comparison. Good precision is calculated using floating point data. [0004] However, when floating-point data is used for calculation, the amount of calculation and data access of the CPU...

AI Technical Summary

Problems solved by technology

[0004] However, when floating-point data is used for calculation, the amount of calculation and data access of the CPU is relatively large, and the performance of the CPU for convolution calculation is difficult to guarantee.

Images