Data processing method and related equipment

Abstract

The invention relates to the field of artificial intelligence, and discloses a data processing method, which comprises the steps of obtaining a transformer model comprising a target network layer and a target module, and processing data to be processed through the transformer model to obtain a data processing result; wherein the target module is used for performing target operation on the feature map output of the target network layer to obtain an operation result, and fusing the operation result with the feature map output to obtain updated feature map output. The target module is inserted into the transformer model, the operation result generated by the target module is fused with the input, the information carried in the feature map output by the target network layer in the transformer model is increased, and as parameter quantity of the target module and the computing power cost needed during operation are very small, namely, data processing precision of the model is improved on the premise of reducing the model parameter quantity and the computing power overhead.

Description

Technical field [0001] This application relates to the field of artificial intelligence, and in particular, to a data processing method and related equipment. Background technique [0002] Artificial intelligence (AI) is a theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a branch of computer science that attempts to understand the nature of intelligence and produce a new class of intelligent machines that can respond in a manner similar to human intelligence. Artificial intelligence is the study of the design principles and implementation methods of various intelligent machines, so that the machines have the functions of perception, reasoning and decision-making. [0003] With the continuous development of artificial inte...

AI Technical Summary

Problems solved by technology

[0005] The transformer model has many parameters and has high requirements for calculation and power consumption. Therefore, the transformer model can usually be compressed such as pruning to obtain a lighter transformer model. However, the compression process will greatly reduce the data processing accuracy of the transformer model.

Images