Model Governance: Production Model Inventory

Owned by Brendan Kelly (Deactivated)
Last updated: Dec 16, 2020 by Dave Trier
3 min read

This article describes how to use the ModelOp Command Center as the central repository for governing models, including how a user can view all models across the enterprise in the ModelOp Center Production Model Inventory.

Table of Contents

 

Production Model Inventory

ModelOp Center provides a Production Model Inventory where you can view all models across the enterprise. Each user can see only the models to which they have access. The Production Model Inventory serves as a central entry point into your models where you can view details, modify, and work further on operationalizing your models.

To access the Production Model Inventory:

  1. Click Models on side bar.

     

  2. The list of available models appears, including the Model Name, Type of Model (e.g. Python3, R), Model Creator, and Last Modifier.

     

To search for a Model by Name:

  1. Click the Search icon in the top right.

    1.  

  2. In the “Filter by Name” field, begin typing in the name of your model.

    1.  

    2. The list of models automatically filters based on your input.

       

  3. To filter by model “tag”, in the “Filter by tag” field, begin typing in the name of the tag.

    1. A drop-down with available tags will be presented.

    2. Select the tag for your model

    3. The list of models automatically filters based on your input:

 

To compare 2 or more models (Champion/Challenger):

  1. Within the Production Model Inventory, select the checkbox in the “Champion/Challenger” column for each model that you would like to include in the comparison:

  2. Hit the “Compare Models” button

  3. The Champion/Challenger comparison screen will open to compare the latest Test Results for those selected models

  4. Note: in the left-hand menu bar, you may choose to include or exclude specific test results for each model.

 

See Model Governance: Standard Model Definition for more information about the information that is contained for each model within the Production Model Inventory.

 

Next Article: Model Governance: Bias & Interpretability >