Multiplier device, multiplication operation method, operation chip, electronic equipment and storage medium

Abstract

The invention provides a multiplier, a multiplication operation method, an operation chip, electronic equipment and a storage medium. The multiplier comprises a multiplier preprocessing module, an encoding module, an addition module and a partial product selection module, wherein the multiplier preprocessing module is used for generating different encoding input values from received multipliers according to different operation bit widths; the encoding module is used for generating different encoding values according to the different encoding input values and performing operation according to the different encoding values and a received multiplicand to obtain a first partial product; and the addition module is used for accumulating a first partial product for corresponding times according to different operation bit widths to generate different second partial products. Multiplication of various mixed bit widths is supported, a multiplier unit can be multiplexed for multiplication with different precisions, consumption of hardware resources is reduced, and time delay and energy consumption can be effectively reduced for a large number of convolution operations which need to be achieved and include operations of multiple complex multiplication and addition combinations.

Description

technical field [0001] The present disclosure relates to the technical field of multiplication, and in particular to a multiplier, a multiplication method, an operation chip, an electronic device, and a computer-readable storage medium. Background technique [0002] Deep learning is one of the important application technologies for artificial intelligence (AI), which is widely used in computer vision, speech recognition and other fields. Among them, Convolutional Neural Network (CNN) is a deep learning efficient recognition technology that has attracted attention in recent years. It directly inputs the original image or data, and performs several layers of convolution with multiple feature filters. product and vector operations to produce highly accurate results in image and speech recognition. The size of the filter can range from 1×1, 3×3 small blocks to 5×5, 7×7 or even 11×11 large-scale convolution operation blocks, so the convolution operation is also a very cost-effec...

AI Technical Summary

Problems solved by technology

It can be understood that the calculation amount of multiple network layers in the neural network is very large, and such calculations often need to use the calculation results performed earlier. Using the above-mentioned traditional adders and multipliers will occupy a large amount of resources in the neural network processor. Brings extremely high delay and energy consumption

Images