A workflow is a visual representation of a series of tasks organized in a logical order to achieve a specific goal. Each task is shown as a node, and workflows are designed to automate processes, streamline activities, and respond effectively to different events.
Key characteristics of workflows include:
- Inputs: Datasets utilized in your workflow, datasets inside your project or external APIs.
- Nodes: Individual components representing specific tasks or actions within the workflow.
- Output: The result generated by your workflow (optional). It can be a dataset stored in a project or Workspace datasets.
How to build a Workflow?
Here's a comprehensive guide to building a workflow using Atlas:
1. Create a new workflow
Begin by entering a project and navigate to the Workflow mode in the left side navigation.
Then click on New workflow
You can start from scratch or select a suitable template closest to the idea you are building.
2. Add workflow input
Datasets that you want to use in your workflow.
Drag and drop a input components from the right panel or just drag the dataset from the left layers panel directly.
3. Add logic nodes
Logic nodes perform tasks and actions within your workflow.
Logic nodes let you apply filters, run calculations, transform data, and actions.
Drag them in from the right panel and connect them to your inputs. You can chain multiple nodes to build up your logic.
Think of each node as a step in your process. Want to clean data? Filter it? Calculate something? This is the place.
4. Add workflow output
Optionally, include an output node at the end of your workflow.
You can return the output data in a dataset that is stored in the project or in Workspace datasets.
5. Run workflow
Then click on the Run button to run the workflow.
Congratulations! You've successfully created a dynamic workflow tailored your needs.
