Watson Machine Learning Service instance

The Watson Machine Learning service supplies the credentials you need to train machine learning models and deploy assets.

Attention: Starting on September 1, 2020, existing “v1” Watson Machine Learning service instances will be deprecated in favor of “v2” plans that support new features, such as deployment spaces. Before you associate an new service instance, use the Watson Studio migration tool or use migration APIs to move existing assets to a project, then follow these instructions to associate a v2 Watson Machine Learning service instance so you can access new features.

Your machine learning instance is one of the associated services that displays on your Service instances page. The service instance governs what resources you can use to create and deploy machine learning models. You can view an existing service instance, delete an instance, or create a new one.

Tip: If your machine learning service instance is deprecated, it will be tagged with a reminder to upgrade. Upgrade service

Creating a Watson Machine Learning service instance

For details on the services available from Watson Studio, see IBM Cloud services. To create a service instance:

  1. Click Service instances from the Watson Studio services section in the navigation pane to view your service instances.
  2. If you don’t have a machine learning service, click Add new service to open the IBM Cloud service catalog.
  3. Select Machine Learning in the AI/Machine Learning category.
  4. Select a region/location to deploy in. For example, choose Dallas.
  5. Select a pricing plan. For example, choose the Lite free version and click Create.

The service instance is available for you to associate with projects, experiments, and deployment spaces.

Associating a service instance with a space

When you create a space, you must associate a machine learning service instance with the space if an association does not exist.

  1. From the Settings page of a space, click Associate instance in the Watson Machine Learning Associated service section.
  2. If you have a valid machine learning service instance, you can associate it with the space.
  3. If you do not have a valid machine learning instance, click Learn more to open the services catalog page, where you can create a new service instance.

Removing a service instance

Before you remove a service instance from a deployment space, delete any deployments associated with the instance.

  1. Click Service instances from the Watson Studio navigation pane to view your service instances.
  2. Can click Remove from the action menu for the service name.