Did you know that you can upload, store, use, and run tests on your own custom model in DataRobot? It’s easy! Look no further than the Model Registry.
- Create your model package:
- Pickle (or otherwise package) your model
- Create your dependencies Requirements
- Create custom.py file (if needed, you can get help to create that file)
- From the Model Registry tab, upload your model package to DataRobot.
Once deployed to a prediction server managed by DataRobot, you can make predictions via the API and monitor your deployment with a suite of capabilities.
I hope you find this tip helpful. I’m working on other tips that I think will help everyone. Please share your tip ideas with me, @Megan, and keep checking back for more tips.