Tutorial 1A: Intro to ML Model Deployment and Management

In this tutorial, you will launch MLOps test drive from aquarium and access the url, which takes you to H2O.ai Platform Studio where you will launch Driverless AI to create a project that is shared with MLOps and then launch MLOps to access that shared project. In MLOps, you will explore the features and then deploy one of the Machine Learning model’s in the project to a development environment. After the model is deployed, you will make scoring requests and get back a scored result using curl. For our particular use case, we will deploy a hydraulic model that predicts hydraulic cooling condition. The result will be percentage values for 3 meaning the hydraulic operates at close to total failure, 20 meaning it operates at reduced efficiency and 100 meaning it operates at full efficiency.

Key:

Complete
Failed
Available
Locked
Objective
Select the "Read" button to begin.
Select the "Read" button to begin. By the end of this tutorial, you will deploy a Driverless AI machine learning model into a development environment using MLOps.
Prerequisites
Select the "Read" button to begin.
Select the "Read" button to begin. For this tutorial, you will need an MLOps Test Drive environment from Aquarium.
Task 1: Explore H2O.ai Platform Studio
Select the "Read" button to begin.
Select the "Read" button to begin. From H2O.ai Platform Studio web application, you will have two launch options: launch Driverless AI or MLOps. In this tutorial, we will start by launching Driverless AI and then later launch MLOps.
Task 2: Create Driverless AI Project and Share it with MLOps
Select the "Read" button to begin.
Select the "Read" button to begin. You will learn to create a Driverless AI project and then share it with MLOps. You will add a Driverless AI experiment to a Driverless AI project. Then you will launch MLOps and see it has access to your Driverless AI project.
Task 3: Machine Learning Operations Concepts
Select the "Read" button to begin.
Select the "Read" button to begin. In the concepts section, we cover AI Lifecycle with H2O.ai products, barriers for AI adoption at scale, what success looks like today for deploying Machine Learning models, what MLOps is, the key components of MLOps, model deployment backend process of MLOps with Kubernetes, monitoring levels for Machine Learning models, typical model lifecycle flow for the production lifecycle management, production model governance, MLOps impact, and MLOps Architecture.
Task 4: Tour of MLOps UI
Select the "Read" button to begin.
Select the "Read" button to begin. You will go through a tour of MLOps UI to see a dashboard of projects, a dashboard of a specific project, a dashboard of a specific project’s ML models, actions that can be performed on a model, more details of a model (adding comments, seeing parameters and metadata), and events of a specific project. You will also learn how to share a project with multiple users.
Task 5: Interactive Scoring via MLOps Model Deployment
Select the "Read" button to begin.
Select the "Read" button to begin. You will learn to deploy a ML model from a specific project using MLOps to a development environment. Alternatively, you could also deploy a model to a production environment. Once the ML model is deployed, you will copy the scoring sample curl request, make a scoring sample request and get back the scored result using curl.
Task 6: Challenge
Select the "Read" button to begin.
Select the "Read" button to begin. You will be provided with some project ideas on how you can go further with your MLOps Machine Learning model deployment knowledge. You can share a new project from Driverless AI with MLOps, deploy that new ML model. You can perform predictions by performing programmatic scoring requests with the programming language of your choice that has an HTTP client.
Next Steps
Select the "Read" button to begin.
Select the "Read" button to begin. After you complete the tutorial, check out an H2O.ai webinar on “Getting the Most Out of Your Machine Learning with ModelOps” or try out one of the Driverless AI model deployment tutorials.
Appendix A: AI Glossary
Select the "Read" button to begin.
Select the "Read" button to begin. If there are terms in the tutorial that you want more clarification on like how it is relevant to MLOps, H2O-3, or Driverless AI or a general definition, we provide you the link to the AI Glossary in H2O.ai Community.
You must be logged in to post to the discussion