Register models¶
In the Model Registry, models are listed as registered models containing deployment-ready model packages as versions. Each package functions the same way, regardless of the origin of its model. The Model Registry also contains the Custom Model Workshop, where you can create, deploy, and register custom models. Custom inference model packages and external model packages are exclusive to MLOps. After you register a model to the Model Registry, you can generate Compliance Documentation to provide evidence that the components of the model work as intended and the model is appropriate for its intended business purpose. You can also deploy the model to production from the Model Registry.
Availability information
Contact your DataRobot representative for information on enabling MLOps-exclusive model package options.
Topic | Describes |
---|---|
Model Registry | How DataRobot AutoML models, custom inference models, and external models are automatically or manually added to the Model Registry. |
Register DataRobot models | How to add a DataRobot model to the Model Registry from the Leaderboard. |
Register custom models (MLOps only) |
How to register custom inference models in the Model Registry. |
Register external models (MLOps only) |
How to register external models in the Model Registry. |
Deploy registered models | Deploy registered models at any time from the Registered Models tab in the Model Registry. |
Manage model packages | How to deploy, share, or archive models from the Model Registry. |
Generate model compliance documentation | How to generate model compliance documentation from model packages in the Model Registry. |
Customize compliance documentation with key values | Build custom compliance documentation templates with references to key values, adding the associated data to the template and limiting the manual editing needed to complete the compliance documentation. |
Custom jobs | Create custom jobs in the Model Registry to define tests for your models and deployments. |
Custom Model Workshop | How to bring pre-trained models into the Model Registry as custom inference models. |
Model logs for model packages (legacy) | View model logs for model packages from the Model Registry's legacy Model Packages tab to see successful operations (INFO status) and errors (ERROR status). |
Topic | Describes |
---|---|
Model Registry | How DataRobot AutoML models, custom inference models, and external models are automatically or manually added to the Model Registry. |
Register DataRobot models | How to add a DataRobot model to the Model Registry from the Leaderboard. |
Register custom models (MLOps only) |
How to register custom inference models in the Model Registry. |
Register external models (MLOps only) |
How to register external models in the Model Registry. |
Deploy registered models | Deploy registered models at any time from the Registered Models tab in the Model Registry. |
Manage model packages | How to deploy, share, or archive models from the Model Registry. |
Generate model compliance documentation | How to generate model compliance documentation from model packages in the Model Registry. |
Customize compliance documentation with key values | Build custom compliance documentation templates with references to key values, adding the associated data to the template and limiting the manual editing needed to complete the compliance documentation. |
Custom jobs | Create custom jobs in the Model Registry to define tests for your models and deployments. |
Custom Model Workshop | How to bring pre-trained models into the Model Registry as custom inference models. |
Import .mlpkg files exported from DataRobot AutoML | How to transfer .mlpkg files from DataRobot AutoML to DataRobot MLOps. |
Model logs for model packages (legacy) | View model logs for model packages from the Model Registry's legacy Model Packages tab to see successful operations (INFO status) and errors (ERROR status). |