0 / 0
Setup options for model evaluations
Last updated: Nov 07, 2024
Setup options for model evaluations

Choose a path for configuring model evaluations according to your preference and level of expertise.

You can choose to set up watsonx.governance to configure evaluations for machine learning or generative AI models.

To set up evaluations for machine learning models, you can use one of the following options:

Automatic setup

The automatic setup option is a guided, no-code experience that sets up and configures a machine learning environment, a database, and a sample model. Follow the steps in the tour to learn to evaluate the sample model in Watson OpenScale. After the setup is complete, you can configure the service with your own database and add your own models to the dashboard. See Automatic setup.

Manual setup

Prepare to evaluate models by connecting to a database, setting up machine learning providers, specifying the environment type (pre-production or production), and optionally adding integrated services. See Manual setup.

Advanced setup

For data scientists who prefer working in Notebooks, learn how to use the OpenScale REST APIs or Python SDK to provision and configure the Watson OpenScale service. See Advanced setup.

Note: This module requires that Python 3 is installed, which includes the pip package management system. For instructions, see [Installing a Python module to set up model evaluation](cloud-alt-setup.html).

Parent topic: Evaluating AI models with Watson OpenScale