A high-precision neural network engineering method based on look-up table calculation

Abstract

The invention discloses a high-precision neural network engineering method based on table look-up calculation, which belongs to the technical field of artificial intelligence neural network deep learning. Sparse the neural network; adopt non-linear quantization to the weight parameters of the neural network, and perform index quantization on the data; use the look-up table method to quickly realize the forward calculation of the neural network, and the present invention achieves the design through the sparse rate automatic distribution strategy. In the case of a fixed sparse rate, the maximum possible reserved network precision is used. The weight parameter nonlinear quantization and data index quantization strategy is used to compress the quantization width to 8bit or below while preserving high calculation accuracy. Finally, by constructing a 256x256 The 32bit high-precision fast calculation table realizes the acceleration of the forward process and reduces the precision in the calculation process without loss, with high calculation accuracy and fast calculation speed.

Description

technical field [0001] The invention relates to a neural network engineering method, and more specifically, relates to a high-precision neural network engineering method based on table lookup calculation, which belongs to the technical field of artificial intelligence neural network deep learning. Background technique [0002] Artificial Neural Network (ANN) is a computing model. It is a research hotspot in the field of artificial intelligence since the 1980s. It is mainly composed of a large number of nodes (called neurons connected to each other, each node represents A specific output function, called the activation function, each connection between two nodes represents a weighted value for the signal passing through the connection, called weight, which is equivalent to the memory of the artificial neural network. Network The output of the network depends on the connection method of the network, the weight value and the activation function are different. [0003] A typica...

AI Technical Summary

Problems solved by technology

[0009] The purpose of the present invention is to solve the problem of network precision loss in the traditional engineering process, and to provide a query-based system with technical characteristics such as high calculation precision and fast calculation speed, and can save high calculation precision while reducing the quantization width. High-precision Neural Network Engineering Method for Table Computing

Images