April 24, 2025

Unified version 25.03.05

Published

April 24, 2025

We've released the ability to manage model stakeholder types within the ValidMind Platform, support for custom configuration of test-driven blocks via the ValidMind Library, an updated Python API reference within our documentation, and more.

Release highlights — 25.03.05

Our documentation now follows the new unified versioning scheme for our software as of our 25.01 release on January 31, 2025.

ValidMind Library (v2.8.13)

New introduction for model development notebook series

We've revamped our old Introduction for model developers notebook into a series of four introductory notebooks — ValidMind for model development:

  1. Set up the ValidMind Library
  2. Start the model development process
  3. Integrate custom tests
  4. Finalize testing and documentation

Access Notebooks on GitHub

These new notebooks break down using ValidMind for your end-to-end model documentation process based on common model development scenarios:

  • Learn the basics of the ValidMind Library with these interactive notebooks designed to introduce you to basic ValidMind concepts and get you familiar with tasks such as how to work with documentation templates, running and logging tests with ValidMind, and more.
  • After you've completed your learning journey with these notebooks, you'll have a fully documented sample model ready for review.

ValidMind Platform (v1.31.10)

Ability to manage model stakeholder types

We've introduced the ability to define custom model stakeholder types, allowing you to control granular permissions on each model in your inventory:

Manage model stakeholder types

  • Model stakeholders determine specific responsibilities and access levels for model review and approval processes for each model in your model inventory, such as read or edit access to model inventory fields.
  • Each model stakeholder also belongs to user groups1 which determine which models they can see, and have user roles2 with attached role permissions3 which define the level of access they have to overarching ValidMind Platform features.

Model stakeholders are now also available as a field defined in conditional arguments within model workflow steps.4

Documentation

Improved Python API reference

We've completely redesigned our ValidMind Library Python API reference using the same tools we use to produce the rest of our product documentation, allowing us to more easily keep this information up to date and ensure its accuracy.

ValidMind Library Python API reference

  • Now featuring more intuitive navigation wrapped in a familiar sidebar, enhanced code signature styling, and integration with our main docs site search, these improvements aim to empower users to maximize the potential of the ValidMind Library.
  • The updated reference structure mirrors the Python package layout, ensuring backward compatibility with our old Python API reference while providing a more reader-friendly experience.

Enhancements

ValidMind Library (v2.8.13)

Support for display configuration of test-driven blocks5

You can now customize the display of test results when inserted as test-driven blocks in the ValidMind Platform when logging results via the ValidMind Library with result.log().

result.log()

You can now toggle the display of the following attributes:

  • hideTitle: Title
  • hideText: Description
  • hideParams: Parameters
  • hideTables: Tables
  • hideFigures: Figures

For example, to show figures but hide tables:

test = vm.tests.run_test(
    "validmind.data_validation.TabularDescriptionTables:raw_dataset",
    input_grid={
        "dataset": [vm_raw_dataset],
    },
)
test.log(config={"hideFigures": False, "hideTables": True})

Bug fixes

ValidMind Platform (v1.31.10)

Improved table of contents navigation for model documentation

Previously, users experienced non-optimal behavior with the table of contents sidebar for their model's Documentation pages,6 such as being unable to scroll to reveal additional headings when the sections of the Document Outline exceeded the length of the visible page, and navigation that auto-collapsed counterintuitively.

Now, users can scroll to reveal previously hidden table of contents headings via a fixed navigation sidebar.

Fixed data table navigation in large test-driven blocks

Previously, when test results logged to the ValidMind Platform contained more than 20 pages of data in a table, the previous and next buttons did not render correctly for pagination when added as test-driven blocks7 into documentation.

Now, test results logged and inserted as test-driven blocks into the documentation display pagination navigation buttons correctly regardless of the size of the data table.

How to upgrade

ValidMind Platform

To access the latest version of the ValidMind Platform,8 hard refresh your browser tab:

  • Windows: Ctrl + Shift + R OR Ctrl + F5
  • MacOS: ⌘ Cmd + Shift + R OR hold down ⌘ Cmd and click the Reload button

ValidMind Library

To upgrade the ValidMind Library:9

  1. In your Jupyter Notebook:

    • Using JupyterHub: Hard refresh your browser tab.
    • In your own developer environment: Restart your notebook.

  2. Then within a code cell or your terminal, run:

    %pip install --upgrade validmind

You may need to restart your kernel after running the upgrade package for changes to be applied.