Network model compression method, device and computer equipment of deep neural network

Abstract

Embodiments of the present invention provide a network model compression method, device, and computer equipment for a deep neural network, wherein the network model compression method for a deep neural network includes: obtaining the original deep neural network; The importance of each computing unit is analyzed, and the computing unit whose importance is lower than the preset importance in the network layer is determined as the computing unit to be deleted; the computing unit to be deleted in each network layer in the original deep neural network is deleted, and the network model compression is obtained. later deep neural network. Through this solution, the efficiency of target recognition and target detection can be improved.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to a network model compression method, device and computer equipment of a deep neural network. Background technique [0002] As an emerging field in machine learning research, DNN (Deep Neural Network) analyzes data by imitating the mechanism of the human brain. It is an intelligent model that analyzes and learns by establishing and simulating the human brain. Currently, it is more popular DNN includes: CNN (Convolutional Neural Network, convolutional neural network), RNN (Recurrent Neural Network, recurrent neural network), LSTM (Long Short Term Memory, long short-term memory network), etc. Since DNN can quickly and accurately identify and detect targets through the operation of multiple network layers in the network model, it has been widely used in target detection and segmentation, behavior detection and recognition, and speech recognition. [0003] With the development...

AI Technical Summary

Problems solved by technology

[0003] With the development of target recognition and target detection technology, the target features are becoming more and more complex, and more and more target features need to be extracted. In this way, in the design of the DNN network model, the network layer and the computing units in each network layer The number is increasing significantly, resulting in an increase in the computational complexity of target recognition and target detection, and a large number of network layers and computing units will consume too much memory and bandwidth resources, affecting the efficiency of target recognition and target detection

Images