Neural network model processing method and device, reasoning method and device and electronic equipment

Abstract

The embodiment of the invention provides a neural network model processing method and device, a reasoning method and device, and electronic equipment. The processing performance of a neural network model can be improved. The neural network model processing method comprises the steps: acquiring a graph model of the neural network model, wherein the graph model comprises a plurality of nodes, and each node in the plurality of nodes comprises an operator; dividing the plurality of nodes into at least two types according to the characteristics of the plurality of nodes; combining the plurality of nodes according to a division result to form at least two types of sub-models; and compiling the at least two types of sub-models to obtain at least two types of executable sub-programs, the at least two types of executable sub-programs being used for running on at least two types of processors. By adopting the method provided by the embodiment of the invention, neural network processing is performed through various types of processors, the characteristics of the various processors can be fully utilized, and the overall processing performance of the neural network is improved.

Description

technical field [0001] The present application relates to the field of computer technology, and more specifically, to a neural network model processing method, reasoning method, device and electronic equipment. Background technique [0002] In recent years, Deep Learning has become one of the hottest research directions in the field of Artificial Intelligence (AI), developing rapidly in vision, speech, natural language and other application fields, and empowering various industries. [0003] At present, the vast majority of researchers or enterprises engaged in deep learning research and application use Graphics Processing Unit (GPU) for training and reasoning of neural network models. However, with the continuous development of neural network models, GPU In terms of performance and power consumption, it is increasingly unable to meet the actual needs. Therefore, academia and industry have begun to vigorously develop neural network-specific processors, such as neural network...

AI Technical Summary

Problems solved by technology

[0003] At present, the vast majority of researchers or enterprises engaged in deep learning research and application use Graphics Processing Unit (GPU) for training and reasoning of neural network models. However, with the continuous development of neural network models, GPU In terms of performance and power consumption, it is increasingly unable to meet the actual needs. Therefore, academia and industry have begun to vigorously develop neural network-specific processors, such as neural network processors (Neural network Processing Unit, NPU) or tensor processing units (Tensor Processing Unit). , TPU) in order to improve the processing performance of the processor on the neural network model. However, not all operations and data in the neural network model are suitable for processing by a neural network dedicated processor. The network-specific processor runs the neural network model, which cannot optimize the processing performance of the processor for the neural network

Images