Process optimization method and device based on machine learning, terminal and storage medium

Abstract

The invention provides a process optimization method based on machine learning. The process optimization method comprises the following steps: establishing a first association relationship between a preset model and model variable information and a second association relationship between the preset model and model scene information; obtaining a target model corresponding to the simulation trial calculation request instruction; obtaining an input parameter of the target model by calling a Groovy engine; calling a model engine to encrypt the target model; calling a model engine to start a process process in Java, and calling a Python environment and a model file in Anaconda through the process; calling a model engine to decrypt the input parameters of the target model; and simulating trial calculation request data through target model operation based on a Python environment in Anaconda and the model file. The invention further provides a device, a terminal and a storage medium. Anacondacovering complete Python third-party class libraries is directly called through Java to operate the model, and the accuracy and speed of model operation can be improved.

Description

technical field [0001] The present invention relates to the field of model deployment, in particular to a process optimization method, device, terminal and storage medium based on machine learning. Background technique [0002] Artificial intelligence is one of the most cutting-edge technologies at present, and it is also the most breakthrough invention in the technological world that human beings can imagine. As an important subcategory of artificial intelligence, machine learning has naturally become a key research direction. At the same time, Python, as the first voice tool in the field of machine learning, has a wide range of application scenarios. However, the current mainstream systems are basically written in java. To achieve intelligent attributes in the java system, it is an inevitable choice for java to call the Python machine learning model. [0003] At present, there are two common ways for Java to call Python scripts. One is to realize it through the class libr...

AI Technical Summary

Problems solved by technology

[0003] At present, there are two common ways for Java to call Python scripts. One is to realize it through the class library provided by jython. It is implemented by converting the Python machine learning model to PMML, and then using java to call PMLL. However, after converting the Python machine learning model to PMLL, Java needs to parse the content of the PMLL file, which is cumbersome to operate. The model and algorithm library model obtained by PMLL conversion Compared with the predicted results, there are deviations, and the PMLL file is large, occupying too many system resources

Images