Getting started
In this section, we give a quick introduction to writing and running Union workflows on your local machine.
To get started, you will need to do the following:
- Install Python 3.8 or higher
- Install Flytekit with
pip install -U flytekit
.
Create a "Hello, world!" workflow
To create an example workflow file, copy the following into a file called example.py
:
from flytekit import task, workflow
@task
def say_hello(name: str) -> str:
return f"Hello, {name}!"
@workflow
def hello_world_wf(name: str = 'world') -> str:
res = say_hello(name=name)
return res
Tasks and workflows
In this example, the file example.py
contains a task and a workflow. These are simply Python functions decorated with the @task
and @workflow
decorators, respectively. The workflow is the top-level construct which you run. The workflow, in turn, invokes the task.
2. Run the example workflow in a local Python environment
Run the workflow with pyflyte run
. The syntax is:
$ pyflyte run <script_path> <task_or_workflow_name>
In this case:
$ pyflyte run example.py hello_world_wf
You should see the following output:
Running Execution on local.
Hello, world!
Since the @workflow
function takes an argument called name
, you can also pass that in:
$ pyflyte run example.py hello_world_wf --name Ada
Then, you should see the following output:
Running Execution on local.
Hello, Ada!
Next steps
In the following sections, we will walk through setting up a simple but production-level Union project and deploying it to your Union instance in the cloud.