Data processing method and data processing device for deep learning reasoning framework

• Main Authors: TIAN YONGZHEN, LU XIN, SUN ZENGZENG, WANG ZHE
• Format: Patent
• Language: Chinese; English
• Published: 17.08.2021
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention discloses a data processing method and device for a deep learning reasoning framework, and the method comprises the steps: responding to a data arrangement mode of a reasoning model which is not supported by the reasoning framework, determining a data arrangement mode conversion strategy of the input data and the output data of the reasoning operator according to the dimensionality of the input data received by the reasoning operator, the dimensionality of the output data correspondingly output and the correlation between the reasoning operator and the data arrangement mode; and according to the determined conversion strategy, converting the data arrangement mode of the input data of the reasoning operator and/or converting the data arrangement mode of the output data of the reasoning operator. According to the technical scheme, the reasoning performance of the deep learning reasoning framework on different Layout deep learning models can be improved. 公开了一种用于深度学习推理框架的数据处理方法和数据处理装置，其中，所述数据处理方法包括：