Skip to content

prasanthpul/azureml-workshop-2019

 
 

Repository files navigation

Azure ML E2E Workshop

Target Audience

Anyone who wants a comprehensive E2E understanding of Azure ML.

Key Goals

  1. Understand the product E2E
  2. Ensure that the work we are doing in Mn will address key gaps in usability / E2E user experience
  3. Open bugs, fix docs & ensure commitments from product area leads on whether / how those bugs will be fixed

Agenda

(1 hour) Module 1, infra setup, ARM, workspace setup, computes, datastores, setup**

(2 hours) Module 2, Datasets, Model Training (AML, HyperDrive and AutoML), Model Interpretability**

Notebooks to run and research:

AML training, HyperDrive and Interpretability:

Automated ML:

Complementary tutorials and HowTo:

https://docs.microsoft.com/en-us/azure/machine-learning/service/how-to-version-track-datasets https://docs.microsoft.com/en-us/azure/machine-learning/service/how-to-train-with-datasets

Stretch Goal for Module 2: Get interpretability working on the scikit-learn training job.**

(2 hours) Module 3, MLOps (model management, deployment, inference, automation)

Tutorials for MLOps

Stretch Goal for Module 3 - Set up automated training and deployment on a schedule.

(1 hour) Module 4, Enterprise Readiness**

About

AzureML Workshop for the 2019 Euro Tour

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 93.7%
  • Python 5.0%
  • R 1.3%