Calculation graph execution method, computer equipment and storage medium

Abstract

The embodiment of the invention discloses a calculation graph execution method, computer equipment and a storage medium, and the calculation graph execution method comprises obtaining a binary instruction executed by an artificial intelligence processor corresponding to a calculation graph according to an operation instruction of a fusion operator when a universal processor compiles the calculation graph with the fusion operator.

Description

technical field [0001] The present application relates to the technical field of information processing, and in particular to a calculation graph execution method, computer equipment and storage media. Background technique [0002] The deep learning framework is the first layer facing applications in the entire deep learning ecosystem. In the early days of Caffe, Layer was regarded as the basic element of building a convolutional neural network. In later deep learning frameworks, such as TensorFlow and MXNet, although different names were used, such as Operator, the core idea of ​​Caffe's Layer is The above is still similar, and the neural network calculation is further divided into various common tensor data-oriented operators, and the user builds the neural network model by combining operators and data. [0003] The above is the interface design idea of ​​the deep learning framework for higher-level applications, and it is also the idea of ​​the deep learning framework's ...

AI Technical Summary

Problems solved by technology

[0006] The embodiment of the present application provides a calculation graph execution method, computer equipment and storage medium, which can solve the IO bottleneck when the artificial intelligence processor executes the learning task

Images