Data processing method and related device

Abstract

The embodiment of the invention discloses a data processing method, which is applied to the field of artificial intelligence, and comprises the steps of obtaining a to-be-compressed network and a plurality of pieces of data, with the to-be-compressed network being a classification network; inputting the plurality of data into the to-be-compressed network to obtain a plurality of first output results, with the plurality of first output results being in one-to-one correspondence with the plurality of data; determining a one-hot one hot label corresponding to each first output result in the plurality of first output results; respectively determining a first similarity between each first output result in the plurality of first output results and the one-hot label; and according to the first similarity corresponding to each first output result in the plurality of first output results, determining at least one piece of target data in the plurality of pieces of data, with the at least one piece of target data being used for compressing the to-be-compressed network. Through the method, a large amount of data similar to the original training data of the to-be-compressed network can be obtained, and therefore it is guaranteed that network compression can be effectively achieved.

Description

technical field [0001] The present application relates to the field of computer technology, in particular to a data processing method and related devices. Background technique [0002] In recent years, deep neural networks have achieved great success in various applications in the field of computer vision, such as image classification, object detection, image segmentation, etc. However, the deep neural network model often contains a large number of model parameters and a large amount of calculation, which makes it difficult to apply to some devices with low computing power (such as terminal devices, embedded devices, integrated devices, etc.). [0003] In related technologies, a deep neural network compression algorithm is proposed, which can compress the teacher network model with large storage space requirements and high computational complexity into a student network model with small storage space requirements and low computational complexity, so that the student network ...

AI Technical Summary

Problems solved by technology

[0004] However, in some cases, the training data of the network to be compressed cannot be obtained, which makes it difficult to effectively compress the neural network

Images