Deep learning model establishement method and device, and image processing method and device

Abstract

The invention provides a deep learning model establishment method and device and an image processing method and device, and the method comprises the steps: setting the maximum number of input channelsand the maximum number of output channels of a convolutional layer as the volume parameters of a weight pool of the convolutional layer in a first deep learning model; multiplying the plurality of proportional parameters by the volume parameters of the weight pool in sequence to obtain a plurality of sub-weight pools; and sequentially performing convolution calculation on the sub-weight pools, establishing a plurality of second deep learning models according to the calculated output channel number of each sub-weight pool, importing the second deep learning models into a heuristic algorithm model, and outputting a third deep learning model meeting a preset test index. According to the method, the second deep learning models with different width combinations are divided, the heuristic algorithm model is utilized, and the third deep learning model with the optimized width combination is screened out from the second deep learning models, so that the purpose of automation of model width setting is achieved, and resources needed by model training are reduced.

Description

technical field [0001] The invention belongs to the technical field of computers, and in particular relates to a method for establishing a deep learning model, an image processing method and a device. Background technique [0002] The deep learning model originates from the research on artificial neural networks. The setting process of the model width (that is, the number of channels of each convolutional layer) parameters of the deep learning model is related to the performance of the deep learning model, as well as the relationship between the deep learning model and the business or operation. match between them. [0003] At present, the model width design of the deep learning model is ever-changing. The width refers to the number of output channels of the convolutional layer of the model. Different convolutional layers often need to design different numbers of output channels, that is, the number of different feature maps for feature representation. Now A commonly used d...

AI Technical Summary

Problems solved by technology

[0005] The present invention provides a method for establishing a deep learning model, an image processing method and a device, so as to solve the problem in the prior art that it takes too much time to adjust the model width and reduces the efficiency of model establishment

Images