Configuring global machine learning providers

Global machine learning providers can be used in all the decision automations that are created in the same instance, and can be used by all users.

Before you begin

Note: You must have the Administer platform for decision services permission to access the administration page. This permission is associated with the Decision Designer Platform Administrator predefined role. For more information, see Manage user permissions.

About this task

You can have global or local machine learning providers for your decision automations:

Global providers: Available from all the decision automations created in the same instance. You can configure global machine learning providers in the Global providers on the administration page.
Local providers: Accessible only from a specific decision automation. You can configure local machine learning providers on the Settings page of a decision automation.

The following machine learning providers are supported:

Three remote machine learning providers: IBM Watson® Machine Learning, Amazon SageMaker, and IBM® Open Prediction Service.
An embedded machine learning provider. This provider allows you to import any type of Predictive Model Markup Language (PMML) file and run it directly in your automation.

For more information about these providers, see Managing local machine learning providers.

Before you import out-of-box samples that use predictive models, you must set up and configure a sample machine learning provider called Sample Machine Learning Provider in the Global providers tab. For more information about importing samples, see Creating decision services.

Procedure

To create a global machine learning provider, retrieve the URL of the provider of your choice.

Table 1. Machine learning provider options
Provider	Credentials
Watson Machine Learning	Enter the following service credentials to authenticate with your Watson Machine Learning service instance: API key Space ID Authentication URL URL This information can be found on Watson Studio. The Authentication URL is populated with a default value automatically.
Amazon SageMaker	Enter the following service credentials to authenticate with your SageMaker instance: Region Access key ID Access key value
Open Prediction Service	Enter the URL of your Open Prediction Service instance. You can get the URL while you install your instance. For more information about installing Open Prediction Service instances, see the instructions on GitHub.
Embedded Machine Learning	No credentials needed.

Go to the administration page by adding /admin-platform at the end of the URL of Decision Designer in your browser:
```
https://<decision_designer_url>/admin-platform
```
Open the Global providers tab, and then click New.
Select the provider type in the drop-down list.
Enter the name and description for the global machine learning provider.

Important: If you are configuring a provider to use out-of-box samples, enter Sample Machine Learning Provider as the name.
Complete the required fields depending on the provider type. You must use HTTPS in the URL field, not HTTP.
Click Test connection to verify the connection.
When it is successfully connected, click Save.

To update or delete a provider, click the overflow menu next to the provider name, and then click Edit or Delete.

What to do next

Business users can now connect to the machine learning provider. For more information, see Managing local machine learning providers.