DevOps and AI

Transcript

1. DevOps and Machine Learning Henk Boelman Cloud Advocate @ Microsoft @hboelman github.com/hnky henkboelman.com
2. Machine Learning Ability to learn without being explicitly programmed.
3. Programming Algorithm Data Answers
4. Machine Learning Algorithm Data Answers
5. Machine Learning Model Data Answers
6. Machine Learning Model Data Answers
7. Machine Learning Predictions Data Model Data Answers
8. Sophisticated pretrained models To simplify solution development Azure Databricks Machine Learning VMs Popular frameworks To build advanced deep learning solutions TensorFlow Keras Pytorch Onnx Azure Machine Learning Language Speech … Azure Search Vision On-premises Cloud Edge Productive services To empower data science and development teams Powerful infrastructure To accelerate deep learning Flexible deployment To deploy and manage models on intelligent cloud and edge Machine Learning on Azure Cognitive Services
10. DevOps is the union of people, process, and products to enable continuous delivery of value to your end users. “ ”
11. Developers Data Scientists Operations
12. Ask a sharp question Collect the data Prepare the data Select the algorithm Train the model Use the answer The data science process
13. Azure Machine Learning A fully-managed cloud service that enables you to easily build, deploy, and share predictive analytics solutions.
14. What is Azure Machine Learning? Set of Azure Cloud Services Python SDK Prepare Data Build Models Train Models Manage Models Track Experiments Deploy Models That enables you to:
15. Prepare your environment Experiment with your model & data Deploy Your model into production
16. Step 1: Prepare your environment
18. Datasets – registered, known data sets Experiments – Training runs Pipelines – Training workflows Models – Registered, versioned models Endpoints: Real-time Endpoints – Deployed model endpoints Pipeline Endpoints – Training workflow endpoints Compute – Managed compute Environments – defined training and inference environments Datastores – Connections to data Azure Machine Learning
19. Demo: Azure Machine Learning
20. Step 2: Create a ML pipeline to deliver a model
21. Pipelines Azure ML Service Pipelines Azure Pipelines
22. Azure Machine Learning Pipelines Workflows of steps that can use Data Sources, Datasets and Compute targets Unattended runs Reusability Tracking and versioning
23. Azure Pipelines Orchestration for Continuous Integration and Continuous Delivery Gates, tasks and processes for quality Integration with other services Trigger on code and non-code events
24. Create a pipeline step Input Output Runs a script on a Compute Target in a Docker container. Parameters
25. Create a pipeline Dataset of Simpsons Images Prepare data Train the Model with PyTorch Processed dataset model Register the model Blob Storage Account Model Management
26. Submit the pipeline to the cluster
27. Demo: Create an Azure ML Pipeline
28. Jupyter Notebook Compute Target Docker Image Data store 1. Snapshot folder and send to experiment 2. create docker image 3. Deploy docker and snapshot to compute 4. Mount datastore to compute 6. Stream stdout, logs, metrics 5. Launch the script 7. Copy over outputs Experiment
29. Azure Machine Learning Service Pipelines A repeatable process to deliver a ML model
31. Continuous Integration
32. Code and comments only (not Jupyter output) Plus every part of the pipeline And Infrastructure and dependencies And maybe a subset of data Source Control
33. Everything should be in source control! Except your training data which should be a known, shared data source
34. Triggered on code change Refresh and execute AML Pipeline Code quality, linting, and unit testing Pull request process Continuous Integration
35. Demo: Setup Azure Pipeline for AMLS Pipeline
36. Step 3: Deploy your model
37. Trigger on model registration Deploy to test and staging environments Run integration and load tests Control: rollout, feature flags, A/B testing Continuous Delivery
38. Control model rollout! The same way you do with other software
39. AMLS to deploy The Model Score.py Environment file Docker Image
40. Demo: Deploy with Azure DevOps
41. Complete Pipeline
42. @hboelman github.com/hnky henkboelman.com Thank you! Read more on: henkboelman.com