Overview

Steam is an “instant on” platform that streamlines the entire process of building and deploying applications. It is the industry’s first data science hub that lets data scientists and developers collaboratively build, deploy, and refine predictive applications across large scale datasets. Data scientists can publish Python and R code as REST APIs and easily integrate with production applications.

This document describes how to start and use Steam and the Steam Scoring Service. Note that this document assumes that an admin has successfully installed and started Steam on a YARN edge node using the instructions provided in the Steam Installation and Setup steps.

Note: Before you begin using Steam, be sure that your minimum version of H2O is 3.10.0.7. Earlier versions are not supported. If necessary, follow the instructions on the H2O Download page for your platform to upgrade H2O.

• Installing and Starting Steam
  • Requirements
  • Installing and Starting Steam on YARN
  • Installing and Starting Steam on a Local Machine
  • Steam Start Flags
  • Next Steps

• User Management Overview
  • Terms
  • Privileges/Access Control
  • Authorization
  • User Management Workflow
  • User Management Example
  • Next Steps

• Logging in to Steam
  • The Steam UI
• Projects
  • Creating a Project
  • Models
  • Deployment
  • Configurations
  • Collaborators
• Clusters
  • Launch a New Cluster
• Users
  • Creating Roles
  • Deleting Roles
  • Adding Users
  • Editing Users
  • Deactivating/Reactivating Users
  • Changing Permissions
• Using Steam with H2O Flow
• Stopping Steam

• Appendix A: Prediction Service Builder
  • About the H2O Prediction Service Builder
  • Build and Run the Builder Service
  • Running an Example War File
  • Starting the H2O Prediction Service
  • Making Predictions
  • H2O Prediction Run-Time Stats
• Appendix B: CLI Command Reference
  • build model
  • create dataset
  • create datasource
  • create identity
  • create project
  • create role
  • create workgroup
  • deactivate identity
  • delete cluster
  • delete dataset
  • delete datasource
  • delete engine
  • delete model
  • delete project
  • delete role
  • delete service
  • delete workgroup
  • get all cluster-types
  • get all entity-types
  • get all permissions
  • get cluster
  • get clusters
  • get dataset
  • get datasets
  • get datasource
  • get datasources
  • get engine
  • get engines
  • get identities
  • get identity
  • get model
  • get models
  • get permissions
  • get project
  • get projects
  • get role
  • get roles
  • get service
  • get services
  • get workgroup
  • get workgroups
  • import model
  • link identity
  • link role
  • login
  • register cluster
  • reset
  • start cluster
  • stop cluster
  • stop service
  • unlink identity
  • unregister cluster
  • update role
  • update workgroup
  • upload engine
  • upload file