Getting Started
The best way to get started is to walk through a couple of use case that take us from raw data to model predict, then onto a capability pipeline. The two use case are both classification challenges, with the first using the titanic dataset to show each capability and the second, a more real world situation of customer churn.
Following each use case is the re-usable pipeline where the capability recipes have been placed on a remote repository. We create a new synthetic dataset and run it through the remote pipeline.
use case one: Using the Titanic dataset, predict if a passenger is likely to survive.
use case one remote: run the predict pipeline as a reusable service.
use case two: Using a customer banking dataset, predict customer churn.
use case two remote: run the churn pipeline as a reusable service.