So, let's start by understanding just what it is that we mean when we say pipeline.
Creating a Spark pipeline in Watson Studio
What is a pipeline?
An ML pipeline is characteristically used to automate ML workflows, essentially enabling sets of data to be transformed and correlated in a model that can then be tested and evaluated to achieve or estimate an outcome.
Such a workflow consists of four basic areas:
- Data preparation
- Training set generation
- Algorithm training/evaluation/selection
- Deployment/monitoring
Pipeline objectives
A pipeline consists of a sequence...
