Data processing method and training method of data processing model

Abstract

The invention provides a data processing method and a training method of a data processing model. The data processing method comprises the following steps that: target data is acquired, wherein the target data comprises a target image; an ith processing layer group is made to process the target data to output an ith processing result, wherein the ith processing layer group comprises first Ni common feature processing layers and an ith branch layer group which are connected in sequence, wherein the branch layer group comprises a plurality of branch feature processing layers, i is larger than or equal to 1, and the sum of Ni + 1 is larger than or equal to Ni and larger than or equal to 1; and a final processing result of the target data is determined according to the at least one processing result. According to the data processing method and training method of the data processing model of the embodiment of the invention, the data processing model with a common layer group and a plurality of branch layer groups can be used for completing a plurality of data processing tasks end to end, and the processing capability of the data processing tasks is effectively improved by combining the improvement of each branch layer group on the common layer group and the reverse influence of the common layer group on each branch layer group, and the data processing efficiency and accuracy are improved.

Description

technical field [0001] The present disclosure relates to the technical field of machine learning, and in particular, to a data processing method and a training method for a data processing model. Background technique [0002] With the increasing popularity of smartphones, many companies have launched mobile phone broken screen insurance. In the process of claim settlement, the customer or repairer is first required to take a picture of the mobile phone screen with the IMEI code (International Mobile Equipment Identity, International Mobile Equipment Identity), and perform the following image recognition and verification work: (1) Screen out the pictures with the mobile phone screen (2) Determine whether the mobile phone screen is really broken; (3) Identify the IMEI code in the image and judge whether the mobile phone is an insured mobile phone according to the IMEI code; (4) Determine the mobile phone model corresponding to the IMEI code The model is the same. [0003] In...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present disclosure is to provide a data processing method and a data processing model training method, which are used to overcome the limitations and defects of related technologies at least to a certain extent in the multi-task processing process caused by the matching of multiple models to training samples Defects such as high requirements and low data processing efficiency in the actual application process

Images