Neural network model pruning method and device, computer equipment and storage medium

Abstract

The invention relates to the field of artificial intelligence, and provides a neural network model pruning method and device, computer equipment and a storage medium. The method comprises the steps ofobtaining to-be-processed data, inputting the to-be-processed data into an initial neural network model, and respectively obtaining a feature map output by each layer of the initial neural network model; decomposing the feature map output by each layer to obtain redundancy data and channel index data corresponding to the feature map; determining pruning channel data of each layer based on the redundancy data and the channel index data; and carrying out pruning processing on the initial neural network model based on the pruning channel data. By processing the feature map, redundancy data and channel index data are obtained at the same time, a basis is provided for determining pruned channel data, other constraints do not need to be added, pruning operation can be effectively carried out, operation is easy, and control is convenient.

Description

technical field [0001] The present application relates to the technical field of artificial intelligence, in particular to a neural network model pruning method, device, computer equipment and storage medium. Background technique [0002] With the development and progress of computer technology, deep learning has developed rapidly in various fields. Among them, neural network models have been widely used in many tasks such as image classification, object detection, semantic segmentation, and instance segmentation. With the continuous development of applications, the amount of parameters and calculations of neural network models are also increasing, which limits the application of neural network models to devices with limited computing resources such as mobile terminals. [0003] Therefore, how to obtain a smaller neural network model is of great significance for application scenarios with limited computing resources. Pruning refers to deleting some connections and weight pa...

AI Technical Summary

Problems solved by technology

Pruning refers to deleting some connections and weight parameters in the neural network, thereby reducing the size of the neural network model, while shortening the inference time to a certain extent, but without significantly reducing the performance of the neural network

However, the traditional neural network model pruning method is cumbersome and difficult to control effectively.

Images