Data processing method and apparatus, DMA controller and computer readable storage medium

Abstract

A data processing method and apparatus, a DMA controller and a computer readable storage medium are provided. The method comprises: generating first DMA read configuration information and first DMA write configuration information according to characteristic information of an input characteristic map, and determining second DMA read configuration information and second DMA write configuration information (201); reading data from the input characteristic map according to the first DMA read configuration information; according to the first DMA write configuration information, outputting the readdata to an operation logic (202); reading an operation result of the operation logic according to the second DMA read configuration information; according to the second DMA write configuration information, storing the read operation result in an output characteristic map (203). By using the method, data transfer in a CNN can be realized by the DMA controller, and the data transfer in the CNN doesnot need to be realized by a CPU, so that the burden of the CPU is lightened, the data is moved more efficiently, the effect of the CNN operation is accelerated, and the flexibility is maintained at the same time.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a data processing method, equipment, DMA (Direct Memory Access, direct memory access) controller and a computer-readable storage medium. Background technique [0002] In machine learning, CNN (Convolutional Neural Network, Convolutional Neural Network) is a feedforward neural network whose artificial neurons can respond to surrounding units within a part of the coverage area, and it has excellent performance for large-scale image processing. CNN is a multi-layer neural network, each layer is composed of multiple two-dimensional planes, and each plane is composed of multiple independent neurons. Generally, CNN can be composed of a convolutional layer and a pooling layer. The function of the convolutional layer is to extract various features of the image, and the function of the pooling layer is to perform two feature extractions on the original feature signal to r...

AI Technical Summary

Problems solved by technology

[0003] A variety of data transfer tasks are involved in CNN. The data transfer tasks are implemented by the CPU (Central Processing Unit, central processing unit). The data transfer efficiency is low, which adds too much burden to the CPU.

Images