0 / 0
Working with unstructured text models

Working with unstructured text models

Watson OpenScale supports monitoring unstructured image or text data for explainability, but it requires a custom configuration. Fairness and drift metrics are not supported for unstructured image or text data types.

You must complete the following steps to set up your environment. For information about configuring support for non-space-delimited languages, such as Japanese, Chinese, and Korean, see Enabling non-space-delimited language support.

  1. Set up your environment.

    1. Install the Watson OpenScale and IBM Watson Machine Learning packages.
    2. Configure your credentials.
    3. Install the libraries that are needed for creating the models and doing analysis. These include the following libraries:
      • pandas
      • pyspark (if not using Watson Studio)
  2. Create and deploy your text-based model.

    1. Load training data into a pandas frame.
    2. Train the model on the data.
    3. Publish the model.
    4. Deploy and score the model.
  3. Configure Watson OpenScale by assigning the APIClient, subscribing the asset, and scoring the model.

  4. Configure explainability.

    1. Enable the explainability.
    2. Get explanations for the transactions.

Next steps

Creating custom evaluations and metrics

Parent topic: Configuring model monitors

Generative AI search and answer
These answers are generated by a large language model in watsonx.ai based on content from the product documentation. Learn more