Skip to content

Latest commit

 

History

History
48 lines (37 loc) · 1.77 KB

pm_service_api_spark.md

File metadata and controls

48 lines (37 loc) · 1.77 KB
Raw
copyright lastupdated
years
2016, 2017
2017-08-09

{:new_window: target="_blank"} {:shortdesc: .shortdesc} {:screen: .screen} {:codeblock: .codeblock} {:pre: .pre}

Service API

The Machine Learning service is a set of REST APIs called from any programming language, permitting integration of Data Science Experience developed analytics in your applications. Bind your Bluemix applications to the Machine Learning service instance and generate the predictive analytics that your applications need to deliver higher value to your users. Manage models in the administration dashboard, and use the dashboard to create online, batch, or streaming deployments integrated with your applications.

Develop applications against the Data Science Experience files (Spark and Python models) that are deployed on a service instance through the powerful REST APIs:

  • Retrieve the metadata for a given predictive model
  • Deploy models and manage deployed models
    • Online deployment (scoring)
    • Batch deployment (supporting IBM Object Storage and DashDB)
    • Streaming deployment (supporting IBM MessageHub)
  • Retrieve the metadata for a given deployment
  • Generate predictive analytics by making score requests against deployed models

You can easily explore the Machine Learning service capability via the available Swagger representation of the REST API.

See the following sections for REST API examples of: