Importing a model from IBM Cloud Pak for Data

If you develop and train Spark, Scikit-learn, PMML, and XGBoost models on IBM Cloud Pak for Data, you can import those models into WML for z/OS for deployment and management.

Before you begin

  • Verify that the models you've trained on Cloud Pak for Data support the following machine learning algorithm libraries and release levels:
    • Spark 2.4.0 or 2.3.0
    • XGBoost 0.82 or 0.90
    • Scikit-learn 0.19.x or 0.22.x
  • Locate the following access and authorization information:
    • User name and password for the WML for z/OS web user interface.
    • Host name or IP address, port number, user name, and password of the Cloud Pak for Data system where your Spark, Scikit-learn, PMML, and XGBoost models are.

Procedure

  1. Sign into the WML for z/OS user interface with your user name and password.
  2. From the side bar, go to the Model Management page, select the Models tab, and click Import model.
  3. Click the From Cloud Pak for Data tab.
  4. Specify the host name or IP address, port number, user name, and password for the Cloud Pak for Data system.
  5. Select a deployment space where your Spark, Scikit-learn, PMML, or XGBoost models are saved.
  6. Select a model that you want to import from the specified deployment space.
  7. Specify a different name for the model if needed.
  8. Click Import model to import the model.
  9. Verify that the imported Spark, Scikit-learn, PMML, or XGBoost model shows up on the Models tab of the Model Management page.