Skip to content

Details

Hands-on hackathon / workshop on how to operationalise a model using Azure Machine Learning Service.

In this workshop we will show you how to build an end-to-end machine learning pipeline from experimentation to operationalisation.

Agenda:

10:00 - 11:00 End-to-end Data Science Process using Azure Machine Learning Service (Presentation - Fatos Ismali/Muffajul Ali)

11:00 - 13:00 Machine Learning Experimentation

  • Perform data exploratory analysis using Python / Jupyter Notebooks and
    Azure Data prep SDK.
  • Train a model and log parameters with Azure Machine Learning Service
  • Register the model in Azure Machine Learning Service

13:00 - 14:00 Lunch / Networking (kindly sponsored by PA Consulting)

14:00 - 15:00 - Azure ML Ops with CI/CD (Demo + Presentation - Lorea Arrizabalaga)

15:00 - 17:00 Model Operationalization

  • Deploy the model to Docker container
  • Deploy the docker container to an Azure Container Instance or
    Kubernetes Cluster
  • Expose the model as an HTTP endpoint
  • Use the model for prediction

17:00 - Close

More detail regarding development environments and logistics to follow.

UPDATE:

To follow the exercises, an Azure subscription is necessary. If you already have an Azure subscription feel free to use that one. Otherwise, we have 60 Azure subscriptions that we have allocated and we will be handing those over to you on the day so you can run the exercises on Azure Cloud.

Instructions for installing pre-requisites:

Windows/Mac/Linux:

  1. Please make sure you have Anaconda installed https://www.anaconda.com/distribution/#download-section
  2. Make sure your Anaconda is added to your Path environment variables
  3. pip install --upgrade azureml-sdk[explain,automl]
  4. pip install azureml-widgets

For those using Windows (additional steps below):

Events in London, GB

Members are also interested in