Peak Workflows enable you to automate routine tasks.
Contents
What are Workflows?
Workflows are a core function within Peak. They are used to automate routine data processing tasks. These can include:
- Transforming ingested data
- Retraining models
- Providing outputs
Workflow steps
Workflows are created by defining a series of steps which are then shown as a flowchart.
All workflows start with a trigger event and then comprise a series of one or more of the following step types:
- Input Data
Prepare your data ready to use in an application block, for example Amazon Personalize. - SQL Query
Run SQL queries that you have created and saved using SQL Explorer. - Standard Step
Trigger the scripts that make up your models. - Application Block
Configure the model training parameters for specific application blocks. - API Deployment
Deploy an API endpoint as part of workflow so that the results from specific models can be exposed. - Export Data
Export the results of a model either to a table in the tenant’s data lake or as a file that can be downloaded.
Getting started
Further information
- What are Workflows?
- Managing Workflows
- Configuring a Workflow for Amazon Personalize
- Trigger Workflows from GitHub Actions