Administrative Information
| Title | Data architecture |
| Duration | 60 |
| Module | B |
| Lesson Type | Tutorial |
| Focus | Practical - Organisational AI |
| Topic | Data architecture |
Keywords
Data Architecture,Machine Learning pipeline,MLOps,
Learning Goals
- To know the fundamentals of Machine Learning System architectures
- To know how to design and automate a Machine Learning pipeline
- To know the basic aspects of some ML production pipeline
- To know how to configure a ML production pipeline on Cloud
Expected Preparation
Learning Events to be Completed Before
None.
Obligatory for Students
- Data Analysis Process
- Machine Learning Models
Optional for Students
- DevOps
- CI/CD
References and background for students
None.
Recommended for Teachers
- Google cloud architcture
- Bakshi, K. (2012, March). Considerations for big data: Architecture and approach. In 2012 IEEE aerospace conference (pp. 1-7). IEEE.
Lesson materials
Instructions for Teachers
Topics to be covered
- Introduction to what a data architecture is
- MLOps
- Design and Automation of a Machine Learning pipeline
- Architecture of a Machine Learning system in production
- Orchestration of the ML pipeline.
- Configuration of a Continuous Integration
- Continuous Delivery CI/CD system for the ML pipeline using the Cloud.
Outline
| Duration (min) | Description | Concepts | Activity | Material |
|---|---|---|---|---|
| 5 | MLOps | ML System, ML Pipeline, DevOps, Continuous Integration, Continuous Delivery, Continuous Testing |
Lecture | External Slides |
| 10 | Design and Automation of a Machine Learning pipeline | CI/CD Pipeline, CT Pipeline | Lecture | External Slides |
| 20 | Architecture of a Machine Learning system in production | Data validation, Preprocessing, Model Development, Model Analysis, Model Deployment, TFX |
Lecture | External Slides |
| 10 | Orchestration of the ML pipeline | Orchestrator, Kubeflow | Lecture | External Slides |
| 15 | Configuration of a Continuous Integration/ Continuous Delivery CI/CD system for the ML pipeline using the Cloud | ML pipeline on Cloud | Running Example | Online Tutorials |
Acknowledgements
The Human-Centered AI Masters programme was Co-Financed by the Connecting Europe Facility of the European Union Under Grant №CEF-TC-2020-1 Digital Skills 2020-EU-IA-0068.