https://github.com/dem108/amlworkshop-iotedge-devops

This repo has some proposed agenda for Azure Machine Learning related hands-on workshops.
https://github.com/dem108/amlworkshop-iotedge-devops

automl azure azuremachinelearning azureml devops handson iot iotedge mlops workshop

Last synced: 11 days ago
JSON representation

This repo has some proposed agenda for Azure Machine Learning related hands-on workshops.

• Host: GitHub
• URL: https://github.com/dem108/amlworkshop-iotedge-devops
• Owner: dem108
• License: mit
• Created: 2019-05-09T14:02:16.000Z (almost 6 years ago)
• Default Branch: master
• Last Pushed: 2021-02-02T08:49:54.000Z (about 4 years ago)
• Last Synced: 2025-03-28T02:11:19.573Z (29 days ago)
• Topics: automl, azure, azuremachinelearning, azureml, devops, handson, iot, iotedge, mlops, workshop
• Language: Jupyter Notebook
• Homepage:
• Size: 17.6 MB
• Stars: 11
• Watchers: 1
• Forks: 4
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# AMLWorkshop-IotEdge-DevOps

> This repo caters different scenarios regarding Azure ML workshops

- [Agenda: Many Models with Azure ML](#agenda-many-models-with-azure-ml)

- [Agenda: ML/IoT/DevOps Hands on Workshop](#agenda-mliotdevops-hands-on-workshop)

---

## Agenda: Many Models with Azure ML

### Day 1: AutoML and Pipelines Basic - full instructions [here](https://github.com/dem108/AMLWorkshop-IotEdge-DevOps/blob/master/agendas/many-models/Day1-automl-pipeline-basic.md)

- Pre-requisites

  - Python skills

  - Understanding of [key concepts of Azure ML](https://docs.microsoft.com/en-us/azure/machine-learning/concept-azure-machine-learning-architecture)

  - Get familiar with Azure ML by running other experiments, trying own datasets, extending from previous workshops

  - Bring any questions on overall Azure ML, share your feedbacks so far before Day 1

- Morning

  - Intro to agenda, getting to know each other

  - Fundamentals

  - Discussions

- Afternoon

  - Automated ML

  - ML Pipeline

  - Discussions

### Day 2: Dive into Many Models - full instructions [here](https://github.com/dem108/AMLWorkshop-IotEdge-DevOps/blob/master/agendas/many-models/Day2-dive-into-many-models.md)

- Pre-requisites

  - Full understanding of key concepts on [ML Pipelines](https://docs.microsoft.com/en-us/azure/machine-learning/concept-ml-pipelines)

  - Watch videos ([1](https://channel9.msdn.com/Shows/Docs-AI/Building-Large-Scale-Machine-Learning-Forecasting-Models-using-Azure-Machine-Learnings-Automated-ML), [2](https://channel9.msdn.com/Shows/Docs-AI/Building-Large-Scale-Machine-Learning-Models-using-Azure-Machine-Learning), less than 30 min in total) on the solution accelerator for many models

  - Bring any questions on Pipeline before Day 2

  - (Optional) Create SSH key pairs (OpenSSH rsa format) and install [VSCode](https://docs.microsoft.com/en-us/azure/machine-learning/how-to-set-up-vs-code-remote) for remote connection

- Morning

  - Intro to [SA for many models](https://github.com/microsoft/solution-accelerator-many-models)

  - Set up dev env (new compute instance, conda env, optional: vscode remote connection)

  - Recap [ParallelRunStep](https://docs.microsoft.com/en-us/python/api/azureml-pipeline-steps/azureml.pipeline.steps?preserve-view=true&view=azure-ml-py)

  - Discussions

- Afternoon

  - Follow [path 1: AutoML for Many Models](https://github.com/microsoft/solution-accelerator-many-models#using-automated-ml-to-train-the-models) (for both training and inferencing)

  - (Optional) Follow [path 2: Custom ML models for Many Models](https://github.com/microsoft/solution-accelerator-many-models#using-a-custom-training-script-to-train-the-models)

  - Discussions

---

## Agenda: ML/IoT/DevOps Hands on Workshop

### Day 1: Azure ML Basic - full instructions [here](https://github.com/dem108/AMLWorkshop-IotEdge-DevOps/blob/master/agendas/mlops-edge/Day1-AzureML.md)

- Common

  - 09:30-10:00 Workshop overview, scope, expectations

- ML Track

  - 10:00-10:50 Dev environment setup: Azure ML service Workspace and Azure Notebooks. Authenticate, prepare compute (Azure ML Compute)

  - 11:00-11:50 Train first DL model on Azure Notebooks using Azure ML Compute

  - 13:00-14:50 Distributed training with Horovod on AML Compute, explore AML Workspace

  - 15:00-16:50 Create container images, deploy to Azure Container Instance (and/or Azure Kubernetes Service)

  - 17:00-17:50 Questions and answers

### Day 1 (halfday version): Azure ML Basic - full instructions [here](https://github.com/dem108/AMLWorkshop-IotEdge-DevOps/blob/master/agendas/mlops-edge/Day1-AzureML-halfday.md)

- Prepare (before workshop)

  - Check Azure subscriptoin

  - Install

  

- Afternoon

  - 14:00-14:50 Workshop overview, scope, expectations and getting started

  - 15:00-15:50 15:00-15:50 Visit AML studio, create computes and try Notebooks

  - 16:00-16:50 16:00-16:50 Try Automated ML

  - 17:00-17:50 17:00-17:50 Check out Designer and MLOps

### Day 2: IoT and Edge Basic

- IoT Track

  - 09:30-10:00 Dev environment setup, Azure Resource creation (IoT Hub, DPS, Cosmos DB, ASA, Storage, etc)

  - 10:00-10:30 Set-up Raspberry Pi

  - 10:40-11:00 Run D2C message application on Pi

  - 11:00-11:50 Provision a device using Azure IoT DPS (X.509 Individual Enrollment)

  - 13:00-13:50 D2C message, Azure Stream Analytics, Data to Storage/DB

  - 14:00-17:50 Custom Vision Edge module deployment

### Day 3: ML + IoT Edge + DevOps - full instructions [here](https://github.com/dem108/AMLWorkshop-IotEdge-DevOps/blob/master/agendas/mlops-edge/Day3-DevOps-ML-IotEdge.md)

- ML+IoT Edge+DevOps Track

  - 09:30-10:00 Day 1, 2 reflection, Day 3 expectations

  - 10:00-11:50 Dev environment setup: Use GitHub Desktop, Azure DevOps(create DevOps account, Organization), create from Azure ML template, customize Build Pipeline

  - 13:00-14:50 Customize Release Pipeline, Git clone using personal token, test CI build

  - 15:00-16:50 Integrate with IoT Edge deployment

  - 17:00-17:50 Questions and answers