Reconfigurable computation structure meeting requirement for arbitrary-dimension convolution and computation scheduling method and device

Abstract

The invention relates to a reconfigurable computation structure meeting the requirement for arbitrary-dimension convolution and a computation scheduling method and device. The reconfigurable computation structure comprises an interface controller and a reconfigurable computation module, the reconfigurable computation module comprises at least one multiply-accumulate computation processing array, each multiply-accumulate computation processing array comprises multiple multiply-accumulate operation processing units, each multiply-accumulate operation processing unit is configured with a corresponding internal bus, and the multiply-accumulate operation processing units are connected in pairs through the internal buses and then connected with a control bus; the interface controller conducts scheduling management on the connection mode between the multiply-accumulate operation processing units and the time-sharing multiplexing frequency of the multiply-accumulate operation processing units through the control bus. According to the reconfigurable computation structure, for arbitrary-dimension convolution, convolution computation is achieved by rapidly reconfiguring the processing units with different computation functions, the flexibility of variable-dimension convolution computation is improved, the parallelism and streamline of the computation process are fully mined, and the convolution computation efficiency is greatly improved.

Description

technical field [0001] The invention belongs to the technical field of convolutional reconfigurable computing, and in particular relates to a reconfigurable computing structure and a computing scheduling method and device suitable for arbitrary dimensional convolution requirements. Background technique [0002] Convolution is a method of mathematical integral transformation and an important operation in analytical mathematics. It is widely used in electronic communication, physics, statistics, earthquake prediction, image processing, and oil field exploration. For one-dimensional discrete convolution, suppose the input is N 1 point sequence f(n 1 ) and N 2 point sequence h(n 2 ), the output sequence N 1 +N 2 -1 point sequence g(n), then: [0003] [0004] For two-dimensional discrete convolution, suppose the input is M 1 ×N 1 dimension matrix f(m 1 ,n 1 ) and M 2 ×N 2 dimension matrix h(m 2 ,n 2 ), output (M 1 +M 2 -1)×(N 1 +N 2 -1) dimensional matrix g(...

AI Technical Summary

Problems solved by technology

The convolution calculation in CNN is somewhat different from the one-dimensional or two-dimensional convolution calculation process. It cannot be realized based on the frequency domain calculation method, and can only be calculated in the time domain by multiplying and accumulating.

The above two convolution calculation processes have a large number of inherent characteristics that can be designed in parallel. Therefore, it is necessary to design a reasonable calculation structure and multi-dimensional parallelization to realize the calculation process to improve the calculation efficiency of convolution.

Images