You’ve had enough of ‘clicking’ your way through creating resources in the Azure portal?
Hearing that it doesn’t make sense to automate?
You can even look at ARM templates?
Someone who created the whole infrastructure by hand has just left the company, and you need to step in?
Do you agree that writing a ton of deployment documentation doesn’t solve the problem?
If you’ve answered ‘yes’ at least once, then that workshop is for you!
You’ll learn the tools and approaches that can help you develop and deploy ML projects on Azure more efficiently, and not only those straightforward ones you can find in the tutorials. You’ll see a practical approach to implementing Infrastructure as Code (IaC) in ML projects using AzureML, Bicep, and Azure Pipelines.
You will broaden your knowledge on:
- what is bicep language, and how to efficiently use it in MLOps
- how to define and develop ML-related infrastructure in Bicep
- how to create an efficient CI/CD pipeline in Azure Pipelines for both the infrastructure and ML pipelines
I’ll also present a few success stories that could equip you with argumentation to convince your organisation that it is worth investing in MLOps automation.
Prerequisites:
- A laptop
- An Azure subscription allowing for Azure Machine Learning workspace deployment:
- Regions: North EU, West EU, UK South
- Registered subscription providers: Microsoft.MachineLearning and Microsoft.MachineLearningServices
- Azure DevOps Services account (free)
- General knowledge about Azure, Azure Pipelines, python
- Visual Studio Code
- Conda installed (https://conda.io/projects/conda/en/latest/user-guide/install/download.html)
Course Features
- Lectures 0
- Quizzes 0
- Duration 3 hours
- Skill level All levels
- Language English
- Students 12
- Assessments Yes
