基于消息队列的LightGBM超参数优化

计算机工程与科学

基于消息队列的LightGBM超参数优化

南东亮1,2，王维庆1，王海云1

（1.新疆大学电气工程学院,新疆乌鲁木齐 830047；

2.国网新疆电力有限公司电力科学研究院，新疆乌鲁木齐，830011）

收稿日期:2018-07-06 修回日期:2018-12-29 出版日期:2019-08-25 发布日期:2019-08-25
基金资助:
自治区教育厅高校重大专项(XJEDU2017I002)；自治区重点研发任务(2016B02019)

Optimization of LightGBM hyper-parameters

based on message queuing

NAN Dong-liang1,2，WANG Wei-qing1,WANG Hai-yun1

（1.School of Electrical Engineering,Xinjiang University,Urumqi 830047;

2.Electric Power Research Institute,State Grid Xinjiang Electric Power Corporation,Urumqi 830011,China）

Received:2018-07-06 Revised:2018-12-29 Online:2019-08-25 Published:2019-08-25

摘要/Abstract

摘要：

为了提高LightGBM超参数优化效率，同时得到全局最优模型，提出了以消息队列方式并行优化LightGBM超参数方法。根据超参数的预选范围，将每一组超参数发送到队列中，各节点从队列获取到消息后以并行方式进行模型训练并验证准确率，最后选出准确率最高的模型计算待预测的数据集。实验结果表明，与传统的网格搜索、贝叶斯法、随机搜索方法以及消息队列串行优化相比，消息队列并行优化超参数方法时间最短，AUC值最大。

关键词: 消息队列, LightGBM, 超参数

Abstract:

In order to improve the optimization efficiency of light gradient boosting machine (LightGBM) hyper-parameters, and obtain the global optimal model, we propose a parallel optimization method for LightGBM hyper-parameters based on message queuing mode. According to the pre-selection range of hyper-parameters, each set of hyper-parameters is sent to the queue in message mode. Every node trains the model with the message obtained from the queue in parallel mode and verifies the accuracy rate of the model. Finally, the model with the highest accuracy is selected to calculate the dataset to be predicted. Experimental results show that compared with the traditional grid search, Bayesian optimization, random search method and message queuing serial optimization, the proposed method has the highest speed and largest area under curve (AUC ) value.

Key words: message queuing, LightGBM, hyper-parameter

南东亮1,2，王维庆1，王海云1. 基于消息队列的LightGBM超参数优化[J]. 计算机工程与科学.

NAN Dong-liang1,2，WANG Wei-qing1,WANG Hai-yun1.

Optimization of LightGBM hyper-parameters

based on message queuing

[J]. Computer Engineering & Science.

[1]	顾楚梅, 曹建军, 王保卫, 徐雨芯, . 基于蚁群参数优化的LightGBM辐射源个体识别[J]. 计算机工程与科学, 2023, 45(01): 85-94.
[2]	孙庞博, 符琦, 陈安华, 蒋云霞. 基于组合预测模型的小样本轴承故障分类诊断 [J]. 计算机工程与科学, 2021, 43(09): 1684-1691.
[3]	李静, 莫思敏 . 基于改进遗传算法的深度神经网络优化研究[J]. 计算机工程与科学, 2021, 43(08): 1503-1511.
[4]	陈孝全刘波田晓朋. 可编程交换机架构的呼叫中心研究[J]. J4, 2008, 30(9): 25-28.