Quantization method and device of image recognition model, computer equipment and storage medium

Abstract

The embodiment of the invention discloses an image recognition model quantification method and device, computer equipment and a storage medium, and belongs to the technical field of digital information transmission. The method comprises the steps: determining a first image set used for model quantization, and determining a second image set used for verifying model precision; determining a first quantization parameter of a to-be-quantized first image recognition model according to the first image set; for a target network layer in the first image recognition model, performing a precision test on the first image recognition model according to the second image set and the first quantization parameter to obtain a second quantization parameter matched with the target network layer; and according to the second quantization parameter, quantizing the target network layer in the first image recognition model to obtain a second image recognition model. Because the second quantization parameter is matched with the target network layer, the target network layer is quantized through the second quantization parameter, and the precision value of the quantized second image recognition model can beimproved.

Description

technical field [0001] The embodiments of the present application relate to the technical field of neural networks, and in particular to a quantification method, device, computer equipment, and storage medium of an image recognition model. Background technique [0002] In the field of neural network technology, image recognition needs to be performed through image recognition models in many scenarios; in order to speed up the running speed of image recognition models and reduce the power consumption of computer equipment, a common solution is to quantify the image recognition model and convert the image The parameters in the recognition model are quantized from floating-point data to fixed-point data. [0003] In related technologies, the computer equipment quantifies the image recognition model as follows: the computer equipment determines the first image recognition model to be quantified, and the input data, weight data and output data of each network layer of the first i...

AI Technical Summary

Problems solved by technology

However, this quantization parameter does not necessarily match each network layer; therefore, when quantizing the floating-point data of each network layer through this quantization parameter, the precision value of the fixed-point data obtained is low, resulting in the quantization of the first The accuracy value of the second image recognition model is low

Images