A convolutional neural network quantization circuit and quantization method

Abstract

The invention discloses a convolutional neural network quantization circuit and a quantization method, belonging to the technical field of artificial intelligence data processing, including an original parameter pool, a comparator array, a quantization parameter calculation unit, and an arithmetic operation unit. The original parameter pool is used for storing The original parameter data required for each layer of convolutional neural network calculations, including each channel data and bias data of all convolution kernels in each layer; the comparator array is used to perform statistical operations on the data in the original parameter pool, and iteratively compares to obtain The maximum value and the minimum value of the parameters of each layer of the convolutional neural network; the quantization parameter calculation unit is used to perform arithmetic operations on the maximum value and the minimum value to obtain the parameters used for model quantization; the arithmetic operation unit is used for The model is quantized, and the obtained results are expressed in integer format with the number of bits specified by the unsigned bit. The invention can reduce the power consumption of the system through quantization, so that the deployment and operation of the convolutional neural network on the terminal can be more reliably guaranteed.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence data processing, in particular to a convolutional neural network quantization circuit and a quantization method. Background technique [0002] As an important direction in the development trend of artificial intelligence, convolutional neural network has been in a state of intense development. Various new models and new algorithms emerge in an endless stream, continuously injecting new impetus into this field. Among them, the increase in the depth and scale of the network model is the main development direction. In the process of continuous improvement in accuracy, the deployment and implementation of the neural network is facing great challenges. [0003] Due to the limitations of the development of mobile terminal processing and storage levels, the artificial neural network whose depth and size have doubled can only be run on processing machines with large-scale computing resour...

AI Technical Summary

Problems solved by technology

[0003] Due to the limitation of mobile terminal processing and storage level development, the artificial neural network with doubled depth and size can only be run in processing machines with large-scale computing resources, and the power consumption and running speed of the device cannot meet the requirements.

End-to-end transplantation cannot be performed, and thus cluster deployment cannot be performed

Some new networks suitable for various scenarios cannot be applied and verified, which hinders the optimization of algorithms and the progress of networks to a certain extent

Images