You can use APIs, SDKs, and notebook tutorials to evaluate your model in more ways with the Watson OpenScale service. Use the model evaluation and prompt setup APIs to prepare for model evaluations, the Python SDK to customize metrics for model monitoring, and the tutorials to demonstrate tasks with Jupyter Notebooks.
Model evaluation API
You can use the Watson OpenScale v2 API to prepare and manage data for traditional model evaluations with the Watson OpenScale service.
Prompt setup API
You can use the prompt setup API to configure artifacts for evaluating prompt templates in deployment spaces and projects with the watsonx.governance service. For more information, see Python SDK.
Python SDK
The Watson OpenScale Python SDK is a Python library that performs model evaluations. You can use the Python SDK to directly configure the data mart database, add your machine learning engine, or select and monitor deployments. For more information, see Metrics computation with the Python SDK.
Learn more
Parent topic: Evaluating AI models with Watson OpenScale