Data management method, neural network processor and terminal equipment

Abstract

The invention is applicable to the technical field of neural networks, and provides a data management method, a neural network processor and terminal equipment. The neural network processor to which the method is applied comprises a task manager, a memory unit, a controller, a cache space, a computing unit, a first DMA for writing data into the cache space, and a second DMA for reading data from the cache space. The method comprises the steps that the task manager configures working parameters of each DMA according to a data caching strategy, wherein the working parameters comprise a data storage mode, the size of an access space, an access starting address and an access ending address; and each DMA and the controller jointly control the transmission and storage of the operation data amongthe memory unit, the cache space and the computing unit according to the working parameters. According to the method provided by the embodiment of the invention, the duration of caching the operationdata by the neural network processor can be reduced, and the calculation rate of the neural network is improved.

Description

technical field [0001] The present application belongs to the technical field of neural network, and in particular relates to a data management method, a neural network processor and terminal equipment. Background technique [0002] In the neural network processor, a cache space, such as Static Random Access Memory (SRAM), is provided between the memory unit for storing operation data and the computing unit. [0003] Used to cache computing data. [0004] The buffer space is configured with an input storage area, a weight storage area and a result storage area to simultaneously store different operation data, including input data, weight data and result data. At present, the size and storage mode of each storage area are usually fixed, but as the neural network calculation proceeds, the structure of the calculation data is constantly changing, for example, the input data will gradually decrease, while the weight data will gradually increase, making the weight The storage a...

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present application provides a data management method, a neural network processor and a terminal device, which can improve the problems of long time-consuming data caching and slow calculation speed in the neural network calculation process

Images