Download model packages¶
You can download model packages (.mlpkg files) for DataRobot models running on remote prediction environments directly from their deployments, or via the model Leaderboard. You can then run prediction jobs with a portable prediction server outside of DataRobot.
Before proceeding, ensure that your deployment supports model package downloads. The deployment must have a DataRobot build environment and an external prediction environment, which you can verifying using the Governance Lens in the Deployment inventory:
After selecting your deployment, go to the Predictions > Portable Predictions tab. This page offers a snippet for using the downloaded model package with the Portable Prediction Server.
Click Download model package. The download appears in the downloads bar when complete.
Once downloaded, you can use the provided code snippet to launch the Portable Prediction Server with the downloaded model package. The Portable Prediction Server will monitor your model for performance and track prediction statistics. Copy the snippet and specify your API token, local file names, and file paths.
The ability to download a model package from the Leaderboard depends on the pricing plan for your organization. Reference the pricing differences documentation for more information.
If you have built a model with AutoML and want to download its model package for use with the Portable Prediction Server, navigate to the model in the Leaderboard and select the Predict > Portable Predictions tab.
From this tab, you can download the model as a model package (.mlpkg) without engaging in the deployment flow. To do so, select Download .mlpkg. The download appears in the downloads bar when complete.
Once downloaded, the example command below can be utilized to launch the Portable Prediction Server with monitoring enabled. You will need to specify your API token, local file names, and paths before launching.