Model compression method and device, electronic equipment and computer storage medium

Abstract

The invention provides a model compression method and device, electronic equipment and a computer storage medium. The method comprises the steps that training sample data are acquired, and the training sample data comprise label sample data; Respectively training the teacher network model and the student network model by using the training sample data to obtain an adaptive distillation loss function and a focus loss function; and performing back propagation on the student network model according to the adaptive distillation loss function and the focus loss function to obtain a trained studentnetwork.

Description

technical field [0001] The present application relates to the field of artificial intelligence, in particular to a model compression method, device, electronic equipment and computer storage medium. Background technique [0002] In recent years, deep learning networks have achieved great success in object detection applications in the field of computer vision. However, since the deep learning network model often contains a large number of model parameters, the amount of calculation is large, the processing speed is slow, and its application is mostly in the cloud, so it still faces huge challenges in the implementation of the terminal. [0003] In order to reduce the redundancy of the network model, researchers at home and abroad have proposed a distillation learning algorithm. In distillation learning, the training of the student network model is guided by refining or distilling the knowledge of the teacher network with a complex structure to the student network model with ...

AI Technical Summary

Problems solved by technology

However, the performance of the distilled student network is not ideal, and there is still a certain gap between the performance of the teacher network and the detection performance of the teacher network.

Moreover, the current distillation learning is based on a two-stage (Two-stage) target detection network, and the application of single-stage (One-stage) target detection has not yet been explored.

Images