使用自定义开源 MLFLOW 对 LLM 驱动的应用程序进行高级实验跟踪.pdf

编号:167672 PDF 11页 837.09KB 下载积分:VIP专享
下载报告请您先登录!

使用自定义开源 MLFLOW 对 LLM 驱动的应用程序进行高级实验跟踪.pdf

1、2024 Databricks Inc.All rights reserved1Advanced Experiment Advanced Experiment Tracking for LLMTracking for LLM-Powered Applications Powered Applications with Customized Openwith Customized Open-Source MLflowSource MLflowManas Mukherjee,Sr.Staff Software Engineer,IntuitManas Mukherjee,Sr.Staff Soft

2、ware Engineer,IntuitDate:06/11/2024 Date:06/11/2024 2024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reserved LLM Applications Overview Why Open Source MLflow Evaluation Process Tracking&Selection of Run&Review Next Steps2AgendaAgendaLLM powered app evaluations tracking using ML

3、FlowLLM powered app evaluations tracking using MLFlow2024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reservedLLM ApplicationsLLM Applications at Intuitat IntuitIntuit Assist:Guides small business owners in QuickBooksCreates launch announcements for marketers in MailchimpHelps C

4、redit Karma members make smart money decisionsTransforms tax preparation in TurboTax3Introduction to LLM-Powered Applications3Importance of ExperimentationImportance of ExperimentationTailoring prompts and configurations for diverse use casesEnsuring high-quality,reliable AI-driven solutionsContinuo

5、us improvement through iterative testing2024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reservedFlexibility Across Cloud Platforms&ToolsFlexibility Across Cloud Platforms&ToolsSupports various cloud tools:notebooks,spark,pipelines etc.Enables users to stay within their preferre

6、d cloud solutions and workflowsOffers a multi-tenant solution with an intuitive internal UXSelfSelf-managed backend datamanaged backend dataSeamless integration with internal databases and integrations with others MDLC components.mlflow server-backend-store-uri postgresql:/$DB_USER:$DB_PWD$DB_HOST:$

友情提示

1、下载报告失败解决办法
2、PDF文件下载后,可能会被浏览器默认打开,此种情况可以点击浏览器菜单,保存网页到桌面,就可以正常下载了。
3、本站不支持迅雷下载,请使用电脑自带的IE浏览器,或者360浏览器、谷歌浏览器下载即可。
4、本站报告下载后的文档和图纸-无水印,预览文档经过压缩,下载后原文更清晰。

本文(使用自定义开源 MLFLOW 对 LLM 驱动的应用程序进行高级实验跟踪.pdf)为本站 (张5G) 主动上传,三个皮匠报告文库仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对上载内容本身不做任何修改或编辑。 若此文所含内容侵犯了您的版权或隐私,请立即通知三个皮匠报告文库(点击联系客服),我们立即给予删除!

温馨提示:如果因为网速或其他原因下载失败请重新下载,重复下载不扣分。
客服
商务合作
小程序
服务号
折叠