计算机科学 ›› 2025, Vol. 52 ›› Issue (6A): 240900022-4.doi: 10.11896/jsjkx.240900022
梁秉豪, 张传刚, 袁明明
LIANG Binghao, ZHANG Chuangang, YUAN Mingming
摘要： 随着企业数智化转型的持续推进,人工智能技术已经开始应用到企业内部管理、经营分析和生产效率提升等各个方面。然而,传统的AI应用研发流程涉及数据采集、数据清洗、特征提取、算法建模和应用研发等多个环节。整体技术门槛高,团队成员协作难,硬件资源利用率低,难以支撑数智化业务需求的敏捷落地。针对上述问题,提出了一套基于预训练大模型的AI应用服务平台。该平台主要面向AI应用研发和运营全过程管理进行设计,大幅降低了团队协作和资产管理难度。针对预备态、设计态和运行态中的核心流程,引入了预训练大模型和低代码技术,通过构建标注大模型、测试大模型和运营大模型,提升了AI应用的研发效率,同时实现了对运营数据的实时分析,保障了用户的使用体验,并大幅提升了硬件资源的利用率。
中图分类号:
| [1]BERNSTEIN D.Containers and Cloud:From LXC to Docker to Kubernetes[J].Cloud Computing,IEEE,2014,1(3):81-84. [2]NVIDIA.NVIDIA Virtual GPU Software Documentation v18.0[EB/OL].https://docs.nvidia.com/vgpu/18.0/index.html. [3]GU J,SONG S,LI Y,et al.GaiaGPU:Sharing GPUs in Container Clouds[C]//2018 IEEE Intl Conf on Parallel & Distributed Processing with Applications,Ubiquitous Computing & Communications,Big Data & Cloud Computing,Social Computing & Networking,Sustainable Computing & Communications.2018,469-476. [4]TAN Z,LI D W,WANG S,et al.Large Language Models for Data Annotation:A Survey [J].arXiv:2402.13446,2024. [5]JING Z,SU Y Y,HAN Y K,et al.When Large Language Mo-dels Meet Vector Databases:A Survey[J].arXiv:2402.01763,2024. [6]PASZKE A,GROSS S,MASSAF,et al.PyTorch:An Imperative Style,High-Performance Deep Learning Library[J].arXiv:2209.15428,2019. [7]ABADI M,AGARWAL A,BARHAM P,et al.TensorFlow:Large-Scale Machine Learning on Heterogeneous Distributed Systems [J].arXiv:1605.08695,2016. [8]ZHOU Y,YANG K.Exploring TensorRT to Improve Real-Time Inference for Deep Learning[C]//2022,IEEE 24th Int Conf on High Performance Computing & Communications;8th Int Conf on Data Science & Systems;20th Int Conf on Smart City;8th Int Conf on Dependability in Sensor,Cloud & Big Data Systems & Application(HPCC/DSS/SmartCity/DependSys).2022. [9]KWON W,LI Z H,ZHUANG S Y,et al.Efficient MemoryManagement for Large Language Model Serving with PagedAttention[J].arXiv:2309.06180,2023. [10]BEN W.MLflow:A Tool for Managing the Machine Learning Lifecycle[EB/OL].https://mlflow.org/docs/latest/index.html. [11]YANG A,YANG B S,HUI B Y,et al.Qwen2 Technical Report[J].arXiv:2407.10671,2024. [12]CHENG T H,SONG L,GE Y X,et al.YOLO-World:Real-Time Open-Vocabulary Object Detection[J].arXiv:2401.17270v2,2024. [13]GAO L,MADAAN A,ZHOU S,et al.PAL:Program-aidedLanguage Models[J].arXiv:2211.10435,2022. [14]YANG L,YANG C,GAO S T,et al.An Empirical Study of Unit Test Generation with Large Language Models[J].arXiv:2406.18181v1,2024. [15]NI C,WANG X Y,CHEN L S,et al.CasModaTest:A Cascaded and Model-agnostic Self-directed Framework for Unit Test Gene-ration[J].arXiv:2406.15743,2024. [16]XIA Y H,CHEN Y Y,SHI T Y,et al.AICoderEval:Improving AI Domain Code Generation of Large Language Models[J].ar-Xiv:2406.04712,2024. [17]MENG W,LIU Y,ZHU Y,et al.LogAnomaly:UnsupervisedDetection of Sequential and Quantitative Anomalies in Unstructured Logs[C]//Twenty-Eighth International Joint Conference on Artificial Intelligence(IJCAI-19).2019. [18]TAO S,MENG W,CHENG Y,et al.LogStamp:Automatic Online Log Parsing Based on Sequence Labelling[J].Performance Evaluation Review,2022(4):49. [19]ZHOU X H,LI G L,LIU Z Y.LLM As DBA [J].arXiv:2308.05481,2024. |
|
||
首页