Convolutional neural network compression method based on channel number search

Abstract

The invention provides a convolutional neural network compression method based on channel number search, and the method comprises the steps: firstly selecting a target image data set for image recognition, and dividing the target image data set into a training set and a test set; secondly, inputting the training set into a convolutional neural network for training, and outputting an importance index of each channel corresponding to the convolutional neural network; comparing the importance index value with a set threshold value, and abandoning a channel corresponding to the importance index lower than the threshold value to obtain an improved convolutional neural network; and finally, replacing the convolution layer of the improved convolutional neural network with the deep convolution layer to obtain a lightweight network model, and inputting the test set into the lightweight network model to verify the identification performance of the lightweight network model. According to the method, the lightweight convolutional neural network model is constructed by combining the search of the number of channels with the improvement of the network convolution mode, so that the parameters ofthe network model are greatly reduced, and the operation speed of the model is increased.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a convolutional neural network compression method based on channel number search. Background technique [0002] Convolutional neural networks have shown great advantages in the field of computer vision. However, due to the disadvantages of high storage and high consumption of deep convolutional neural networks, it is difficult to deploy them on edge computing devices, such as automated robots and drones. , smartphones, etc. To solve this problem, a lot of research on convolutional neural network compression has been carried out. The more common techniques mainly include parameter quantization, knowledge distillation, low-rank decomposition and pruning. Among the above methods, pruning is considered to be one of the most effective tools for compressing network models, which reduces computational complexity by discarding unimportant weights or entire filters. If running...

AI Technical Summary

Problems solved by technology

[0005] In view of the deficiencies in the above-mentioned background technology, the present invention proposes a convolutional neural network compression method based on channel number search, which solves the technical problems of high complexity and low efficiency in the existing convolutional network technology

Images