Data processing method, related equipment and computer storage medium

• Main Authors: ZHANG CHANGZHENG, TU DANDAN, BAI XIAOLONG
• Format: Patent
• Language: Chinese; English
• Published: 24.01.2020
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The embodiment of the invention discloses a data processing method, which comprises the following steps that: computing equipment receives data to be processed, and inputs the data to be processed into a neural network model; the to-be-processed data is processed according to at least one two-dimensional convolution kernel in the neural network model to obtain result data; wherein the to-be-processed data comprises D channels, and each channel in the D channels comprises two-dimensional data with the size of H * W, and D, H and W are all positive integers. By adopting the embodiment of the invention, the problems of relatively large calculation amount, relatively high calculation complexity and the like during convolution operation in an existing neural network can be solved. 本发明实施例公开了数据处理方法，包括：计算设备接收待处理数据，将所述待处理数据输入神经网络模型；根据所述神经网络模型中的至少一个二维卷积核对所述待处理数据进行处理，以获得结果数据；其中，所述待处理数据包括D个通道，所述D个通道中的每个通道包括大小为H×W的二维数据，D、H以及W均为正整数。采用本发明实施例，能够解决现有神经网络中进行卷积运算时存在的计算量较大、计算复杂度较高等问题。