Use the advanced setup option to run a Python notebook that uses sample data to demonstrate how to configure model evaluations.
Use a notebook to build, train, and deploy a model to monitor the model deployment. The notebook also provides historical data that generates insights that you can view. The notebook uses the German Credit risk model that is also used for the auto setup option to provide sample data. For more information, see the Overview of the sample data.
Before you begin
To use the advanced setup option, you must select the Default Spark 3.3.x & Python 3.9 or the Runtime 22.2 on Python 3.10 runtime environment when you create a Jupyter Notebook in the notebook editor. The runtime environments require service credentials for the following services:
- Watson OpenScale
- IBM watsonx.ai Runtime
- Db2 Warehouse
watsonx.ai Studio, watsonx.ai Runtime, Watson OpenScale, and other supplemental services are not available by default. An administrator must install these services on the IBM Cloud Pak for Data platform. To determine whether a service is installed, open the Services catalog and check whether the service is enabled.
Running the advanced setup
- Download the Watson OpenScale and Watson ML Engine.ipynb file from Github. You can sign in and clone the watson-openscale-samples repository to download the file or click Copy raw contents to paste the file content into a new IPYNB file.
- Open watsonx.ai Studio and select one of your projects.
- From the Assets tab, click New asset > Work with data and models in Python or R notebooks.
- Select the From file tab and specify a notebook name.
- Select the Default Spark 3.3.x & Python 3.9 or the Runtime 22.2 on Python 3.10 runtime environment from the Select runtime menu.
- Upload the IPYNB file and click Create. The Working with watsonx.ai Runtime notebook loads and opens in your project.
Now you can follow the steps in the notebook to run the advanced setup and deploy the German Credit risk model to your service instance for model evaluations.
Learn more
Installing a Python module to set up Watson OpenScale
Parent topic: Evaluating AI models