A matrix multiplication method, graphics processor and electronic device

Abstract

The invention provides a matrix multiplying method, a graphic processor and electronic equipment. The process for executing the matrix multiplying method between a first matrix and a second matrix comprises: performing a primary operation, which is to multiplying N matrix units in the first matrix and N matrix units in the second matrix for N times in turn; performing a secondary operation, in which a saving result matrix unit is obtained by accumulating a multiplication result matrix unit Cj; when the saving result matrix unit accumulates the multiplication result matrix unit Cj for a preset number, using the saving result matrix unit as an intermediate result matrix unit; and accumulating the intermediate result matrix unit to obtain the final result matrix unit of the primary operation. When the technique provided by the invention is used, because the temporary accumulation result is cleared after the secondary operation is ended for a plurality of times, the difference between the temporary accumulation result and the number of the digits of the decimal fraction of the matrix elements is very small, and higher accuracy is retained.

Description

technical field [0001] The invention relates to graphics technology, in particular to a matrix multiplication operation method, a graphics processor and electronic equipment. Background technique [0002] A graphics processing unit (GPU, Graphic Processing Unit) has far more computing units than a CPU, so it has been more and more used in the field of high-performance computing. Matrix multiplication is one of the most important operations in numerical computing. In a sense, it is a common basis for numerical computing operations. Therefore, studying how to achieve efficient and accurate GPU matrix multiplication is an important issue in the study of GPU general computing. . [0003] Since a large number of operations inside the GPU are single-precision floating-point operations, the cumulative error increases rapidly when the size of the matrix increases. After testing, when the matrix size exceeds 5000, the error between the GPU calculation results and the CPU calculatio...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is that in the prior art, the errors generated by the intermediate results of matrix multiplication are not processed, resulting in excessive errors in the final GPU calculation results, which affects the accuracy of the calculation.

Images