3.6 ML4AML Use Case

This section describes the post-installation steps for ML4AML use case.

To upgrade the model:
  1. Execute the following scripts.
    • importWorkspaceSQL.sh
    • importNotebooksSM.sh
    • importNotebooksASC.sh
    • importNotebooksAIF.sh
    • importNotebooksAMLES.sh

      Note:

  2. For new model development, users need to start with the Admin notebook CS 8.1.2.9.0 and then proceed with the model training and scoring process.

    To create a model, train, and score, see the OFS Compliance Studio Use Case Guide.

  3. To execute older notebooks, the user needs to select the previous version of the Conda environment (CS 8.1.2.8.0) and run the batches.

    For Model Scoring/Annual Model Validation, see the How to Execute Model Scoring/Annual Model Validation with the Batch Framework section in the OFS Compliance Studio Administration and Configuration Guide.

    For Monthly Model Validation, see the How to Execute Monthly Model Validation with the Batch Framework section in the OFS Compliance Studio Administration and Configuration Guide.

Scenario Conversion Utility (SCU)

To execute SCU:
  1. Execute the following scripts.
    • importWorkspaceSQLSCU.sh
    • importNotebooksSCU.sh
  2. Refresh the Materialized view.
  3. Generate Scenario and Threshold notebook.

    For more information, see the Using Scenario Conversion Utility for ASC section in the OFS Compliance Studio Administration and Configuration Guide.

Automated Scenario Calibration (ASC)

To upgrade the ASC model:
  1. Run the Scenario Conversion Utility.
  2. Create a threshold sets using bulk upload. For more information, see How to Upload Thresholds in Bulk for ASC section in the OFS Compliance Studio Use Case Guide.
  3. Create new ASC Definition. For more information, see the Automated Scenario Calibration (ASC) section in the OFS Compliance Studio Use Case Guide.

Note:

This patch will create backups of the affected jars, files, and directories, storing them in a backup folder.