Classification model training system and implementation method thereof

Abstract

The invention provides a classification model training system and an implementation method thereof. The implementation method comprises the steps of S1, establishing a front end management display interface of a SPARK algorithm classification model training system, and defining a front end interactive request interface of the front end management display interface; S2, establishing a rear end service data source system of the SPARK algorithm classification model training system; S3, based on the front end interactive request interface of the front end management display interface, establishinga rear end service control interface, and establishing a corresponding relation of the rear end service control interface and the front end interactive request interface; S4, establishing internal business logic based on a SPARK algorithm training and optimizing classification model in the rear end service control interface. According to the classification model training system and the implementation method thereof, the operation procedure of classification model training can be effectively simplified, accordingly the labour intensity of developers is effectively reduced, and the developing efficiency is improved.

Description

technical field [0001] The present invention relates to the technical field of information processing, and more specifically, to a classification model training system and an implementation method thereof. Background technique [0002] At present, machine learning using the machine learning library SPARK.MLlib of the SPARK algorithm has become a common machine learning method. In order to use SPARK.MLlib for classification algorithm model training conveniently and quickly, and because the classification algorithm belongs to supervised learning, a large number of labeled samples need to be prepared in advance, divided into training samples and test samples, and then SPARK.MLlib uses these labeled samples To train the classification algorithm model, it is necessary to continuously adjust the samples and model parameters to optimize the classification algorithm model. [0003] The commonly used method of optimizing the classification model requires manually adding training sam...

AI Technical Summary

Problems solved by technology

Manually adding training samples and optimizing model parameters requires developers to spend a lot of time and energy on data preparation and program operation, resulting in low development efficiency

Images