Data mining model publishing method, model and model service management method

Abstract

The invention discloses a data mining model publishing method, a model and a model service management method. The data mining model publishing method comprises the steps of creating a data mining training task online and performing online training to generate a data mining model; collecting data mining model information and storing the data mining model information in a model information database,and storing a data mining model in a model server; and creating data mining model service information according to the data mining model information, storing the data mining model service informationin a model service information database, configuring an environment variable for service operation according to the model service information, storing an environment variable configuration file in amodel server, and creating a data mining model service according to the environment variable configuration file. According to the method, the data mining model can be published as Web service, the functions of model prediction, model information query and the like are supported, and reduction of Web application development cost and scientific management of the model and the model service are facilitated.

Description

technical field [0001] The invention relates to the field of online deployment of machine learning algorithm models, in particular to a data mining model release method and a model and model service management method. Background technique [0002] With the development of machine learning, data mining and other fields, deploying models online to provide online access to model services makes it easier for developers to use machine learning models to solve business problems. [0003] The traditional model online service method is mainly to import the trained model into a PMML file and store it on the web server. It is necessary to load the model inside the web application and build a separate model application for each model, so that the model can be deployed as online service. Although PMML files can be used across multiple platforms, there are disadvantages such as large model files and slow model loading speed; and web application developers need to build different model ap...

AI Technical Summary

Problems solved by technology

Although PMML files can be used across multiple platforms, there are disadvantages such as large model files and slow model loading speed; and web application developers need to build different model applications according to different models, which increases the development cost; each model service embedded Internal to the application, making model maintenance difficult

Images