Wrap Up: API Deployment#
Congratulations!#
You’ve just completed the API Deployment course, where you learned to work with real-time API services in Dataiku.
Here are a few of the main takeaways from this course:
In a real-time scoring use case, we create an API endpoint in the API Designer of a Dataiku project, package it in an API service, and deploy it to an API node via the API Deployer.
One common type of API endpoint is a prediction model. Exposing the model as an API endpoint enables us to generate a prediction for an incoming request.
In the API Designer, you can test queries to make sure that the endpoint responds as expected. You can also configure query enrichments.
Enrichments can enhance features by using a lookup on another table. An enrichment can be useful when an endpoint uses certain features to perform its function, but some of these features aren’t available to the client making the API request.
The API Deployer (a component of the Dataiku Deployer) is the tool that’s used for deploying an API service to an API node. This Deployer can be local or remote (standalone), and an administrator of a Dataiku instance can configure the Deployer as needed.
Learn more#
Now that you’ve completed this course, you may wish to consult the reference documentation for more information on API Node & API Deployer: Real-time APIs.
When ready, continue progressing on the MLOps Practitioner learning path.