Computer data model training method

Abstract

The invention relates to the technical field of computer data processing, and discloses a computer data model training method. The computer data model training method comprises the following steps: S1, data acquisition; S2, data testing; S3, data training; S4, parameters acquisition; and S5: model processing. According to the method, the training data set is divided into the support set training data and the query set training data for combined type diversified training, so that multiple groups and multiple types of training can be completed after a small amount of data is combined when data model training is performed on a computer, and the learning ability of the whole model is improved; therefore, possibility is provided for application of the model in various scenes, the training efficiency and the prediction effect of the model are improved through various statistics, model parameter errors caused by parameter errors can be effectively avoided through the parameter preprocessing matrix, and the stability and accuracy of operation of the whole model are improved.

Description

technical field [0001] The invention belongs to the technical field of computer data processing, in particular to a computer data model training method. Background technique [0002] The data model is the formal framework used to provide information representation and operation means in the database system, and it is also the abstraction of data characteristics. The data model describes the static characteristics, dynamic behavior and constraints of the system from the abstract level. The content of the description can be divided into There are three parts: data structure, data operation and data constraint. Data structure: the data structure in the data model mainly describes the type, content, nature and relationship between data. The data structure is the basis of the data model, and data operations and constraints are based on the data structure. Different data structures have different operations and constraints. Data operations: Data operations in the data model mainl...

AI Technical Summary

Problems solved by technology

[0004] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides a computer data model training method, which effectively solves the need for data model training in the process of building the data model. The existing data model training The method generally relies on a large amount of data to train and obtain model parameters. This learning method limits the feasibility of the model in many application scenarios: in specific real situations, it is also difficult to obtain a large amount of data. During the training process of the model Only exposed to data samples related to a specific task, when faced with new tasks, its adaptability and generalization ability are weak, unable to adapt to a variety of scenarios, and the existing data model training cannot correct wrong parameters , which affects the stability and accuracy of the data model

Images