Version 2.0

Overview:

ModelOp is extremely proud to release the 2nd generation of our flagship product: ModelOp Center! ModelOp Center v2.0 includes a substantial number of new features--running on top of an updated architecture-- to provide our Customers more flexibility in how they Deploy, Monitor, and Govern all of their models across the enterprise. We encourage you to visit our ModelOps Guide for an overview of all the essentials to running an enterprise ModelOps program, as well as our updated ModelOp Center architecture for an overview of the new services incorporated into the v2.0 architecture.

What’s New:

General:

• Refreshed Web UI (“Command Center”): incorporated a new overall layout and design to streamline the user experience and provide additional details about all models, their model lifecycles, and the model runtimes. This includes an Operations-focused landing page to surface critical issues and notifications for proactive attention to ensure your models are providing the highest value possible.

• CLI: revamped command line interface (CLI) to allow those that prefer interacting via a terminal to have easy access to interact with ModelOp Center, including getting model and job status, registering new models, adding new assets, and many more features.

• Model Factory Plugins: added new Jupyter and DataIku plugins to provide Data Scientists streamlined access to register and update models within ModelOp Center directly from their favorite model factory.

• Updated architecture: updated the core architecture that includes a number of new micro services, an enhanced service registry, a best-in-class gateway, and other enhancement to further enable scale and extensibility for the enterprise.

Deploy

• Model Registration:

  • Jupyter: added a new GUI-based Jupyter plugin that provides an intuitive wizard-based approach to register a new model, as well as mechanisms to update existing models, add attachments such as trained model artifacts, and to submit a model for productionization.

  • CLI: streamlined the approach to register new models via the CLI, with flexible options for adding in schemas, as appropriate.

  • DataIku DSS: added support to integrate with DataIku’s API’s to onboard code-based Python models created in DataIku DSS.

  • API: added new API to support registration and persistence of new models, allowing for systematic onboarding of new models, as required.

• Model Lifecycle (MLC) Orchestration

• Champion / Challenger

Score & Monitor

• Batch Jobs

• Runtimes

• Endpoints

• Notifications & Alerting

Govern

• Model Metadata

• Externalized Attachments

• Model Test Results

• Model Reproducibility Automation

• Model Traceability

Deprecations:

With the release of v2.0, there a few v1.x items that will be deprecated:

...

v1.x CLI: both the Go and Python v1.x CLI will be deprecated with the v2.0 software. All users should download the latest v2.0 CLI for usage with our latest v2.0 release.

...

View the release notes for ModelOp Center: