深入探讨 SageMaker Unified Studio 中的 AI_ML 开发 [重复].pdf

《深入探讨 SageMaker Unified Studio 中的 AI_ML 开发 [重复].pdf》由会员分享，可在线阅读，更多相关《深入探讨 SageMaker Unified Studio 中的 AI_ML 开发 [重复].pdf（11页珍藏版）》请在三个皮匠报告上搜索。

1、 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.A N T 3 2 5-RA deep dive into AI/ML development in SageMaker Unified StudioDr.Shivagami Gugan(she/her)Chief TechnologistAWS,Middle East,Africa and TurkeySashi Varana

2、si(she/her)Head of Data&AI,NAMERAWS WWSO Geo Specialists 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.What to expect from this talkThis is a chalk talk-there will be whiteboarding and demosInteraction interact with us as we ask questions and ask us questions This is a 300-level

3、 session we will dive deep into the concepts of AIMLOps using Amazon SageMaker Unified Studio 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.Question to AudienceWhat are the challenges when building data-driven AI/ML applications?2025,Amazon Web Services,Inc.or its affiliates.All

4、 rights reserved.AgendaWhat are the challenges when building data-driven AI/ML applications?How do you modernize data pipelines,accelerate AI development,and enforce regulatory compliance?Taking your Project from development to production using well-architected best practicesEffective AIOps:How can

5、your Team Collaborate seamlesslyDemos:The flexibility of Amazon SageMaker Unified Studio 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.The ML-Lifecycle:Detailed ViewData AggregationData WranglingData BalancingCleanNormalizeRegularizeUnbiasDetect BiasMitigate BiasData LabellingFe

6、ature EngineeringFeature ImportanceTransformingEncoding JoiningFeature Store(Offline)Pre-ProcessingPost-ProcessingData ArtifactAlgorithm SelectionExperimentsTrialsTrial ComponentsCode ArtifactModel TrainingLocalModel ArtifactArtifact LineageHostedModel RegistryMetrics ValidationH