Azure Day Rome Reloaded 2019 - ML.NET Model Lifecycle with Azure DevOps
0 likes68 views
The document discusses the lifecycle of ML.NET models using Azure DevOps, outlining the process from data preparation to model deployment. It emphasizes the importance of unit testing and managing dataset quality, while also mentioning the integration capabilities with Azure ML. Additionally, it provides resources and tutorials for getting started with ML.NET.
1 of 20
Download to read offline
More Related Content
What's hot (20)
Ad
Similar to Azure Day Rome Reloaded 2019 - ML.NET Model Lifecycle with Azure DevOps (20)
Ad
More from azuredayit (14)
Recently uploaded (20)
Azure Day Rome Reloaded 2019 - ML.NET Model Lifecycle with Azure DevOps
- 1. ML.NET Model Lifecycle with Azure DevOps Marco Zamana Senior Developer 4ward Microsoft MVP AI
- 2. Thanks to
- 4. What is ML.NET? 4 Machine learning for .NET applications ML.NET runs on: Windows, Linux, and macOS using .NET Core
- 5. Machine learning for .NET applications 5
- 6. ML.NET CLI 6
- 7. Normal application lifecycle – building 7
- 8. Prepare Your Data Build & Train Run Model consumption End-user app using the ML model ML model ML model Model creation App/tools for training the ML model Datasets Normal ML Workflow
- 10. This is our sample ML model and trainer console app for the CI pipeline The code 10
- 11. If your build pipeline passes all the tests you defined, than you are good to go and deploy the just trained/built ML.NET model. Unit tests from Azure DevOps dashboard 11
- 12. «Providing new data» is the trigger 12
- 13. Unit tests help us “bad data” == bad quality 13
- 14. Azure files can help us for large dataset 14
- 15. Azure DevOps self-hosted agent is better 15
- 16. • Versioning datasets • Databases as training data • DevOps workflow Scenarios • ML Model Versioning • Integration with Azure ML and MLFlow Improovement in the future 16
- 17. Get started with ML.NET 1.0! Get started with ML.NET here. Tutorials and resources at the Microsoft Docs ML.NET Guide Sample apps using ML.NET at the machinelearning-samples GitHub repo Improovement 17
- 19. Thank You!!!
- 20. Thanks to