Skip to content

pm_custom_models.md

Latest commit

 

History

History
85 lines (60 loc) · 4.17 KB

pm_custom_models.md

File metadata and controls

85 lines (60 loc) · 4.17 KB
copyright
years
2016, 2017
lastupdated 2017-11-16

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

Development and persistence of the custom model

Using the {{site.data.keyword.pm_full}} service, you can deploy a model and generate predictive analytics by making score requests against the deployed model. {: shortdesc}

Working with custom models

Scenario name: Product line prediction.

Scenario description:

Our client is running one of the most famous chain stores in Europe. They would like us to figure out their clients' interests in terms of their product line such as personal accessories, camping equipment, and outdoor protection. A data scientist develops a predictive model and shares it with you (the developer). Your task is to deploy the model and generate predictive analytics by making score requests against the deployed model.

Development and persistence of the custom model

Using Data Science Experience

Use Data Science Experience to create custom models. After you sign up, you must sign in to complete the following steps.

  1. Create an organization and a space. The first time you sign in, you'll be asked for it. Click Continue to accept the default values.

  2. After the organization is created, go to Projects and click New project.

  3. Specify a name and description for your project and click Create. The project name you specified will also be used as your Target Container's name.

  4. After the project is created, you can perform one of the following tasks:

Using local environment

You can also use environment of your choice to develop model and later publish, deploy and score using Watson Machine Learning common API client library available on pypi. For more information about the client library, see the sample notebook and documentation.

Note: The common API client library is in beta.

Deployment and scoring of the custom model

You can use the API to deploy and score online, batch, and streaming models.

Learn more

Ready to get started? To create an instance of a service or bind an application, see Using the service with Spark and Python models or Using the service with IBM® SPSS® models.

For more information about the API, see Service API for Spark and Python models or Service API for IBM® SPSS® models.

For more information about IBM® SPSS® Modeler and the modeling algorithms it provides, see IBM Knowledge Center.

For more information about IBM® Data Science Experience and the modeling algorithms it provides, see https://datascience.ibm.com.