This how-to walkthrough is designed to take you through
superduperdb usage step-by-step.
To build more understanding of
superduperdb read the "Fundmentals" section.
The documents in this section build on one-another, and all assume that you have configured and connected to
You'll probably also want to ready about the query API which is relevant to your database:
Time to Explore
📄️ Connecting to the Database
In this document we instantiate the variable db based on configuration and overrides.
📄️ Setting up tables and encodings
superduperdb has flexible support for data-types. In both MongoDB and SQL databases,
📄️ Inserting data
After configuring and connecting, you're ready to insert some data.
📄️ Working with external data sources
This functionality is currently supported for MongoDB only
📄️ Working with and inserting large pieces of data
This functionality is currently only supported by the MongDB API
📄️ Selecting data
After inserting data to superduperdb, it may be queried with a Select query.
📄️ Adding Models to the Database
SuperDuperDB integrates with both AI models and AI APIs
📄️ Training models directly on your datastore
Similarly to applying models to create predictions, training models is possible both procedurally and declaratively in superduperdb.
📄️ Applying the models
Model and Predictor instances may be applied directly to data in the database without first fetching the data client-side.
📄️ Configuring models to ingest features from other models
Sometimes the outputs of one model should be "chained together" to become inputs of another model.
📄️ Setting up and accessing vector-search
Vector-search refers to the task of searching through vectors
📄️ Serializing components with SuperDuperDB
When adding a component to SuperDuperDB,
📄️ Creating complex stacks of functionality
With the declarative API, it's possible to create multiple