Flyte enables production-grade orchestration for machine learning workflows and data processing created to accelerate local workflows to production.

Flyte

<@U06RKAMK0UC> how can I use argparse with pyflyte script

To use `argparse` with a `pyflyte` script, you need to modify your script to handle command-line arguments. However, it's recommended to use `pyflyte run` for running workflows as it simplifies the process. Here's how you can do it:

1. *Modify your script to use `argparse`*:
   Add an `argparse` block to handle command-line arguments.

2. *Run the script using `pyflyte run`*:
   Use `pyflyte run` to execute the script with the specified arguments.

Example:

```
import argparse
from flytekit import task, workflow

@task
def my_task(name: str) -> str:
    return f"Hello, {name}!"

@workflow
def my_workflow(name: str) -> str:
    return my_task(name=name)

if __name__ == "__main__":
    parser = argparse.ArgumentParser(description="Run Flyte workflow")
    parser.add_argument("--name", type=str, required=True, help="Name to greet")
    args = parser.parse_args()
    print(my_workflow(name=args.name))
```

Run the script with `pyflyte run`:

```
pyflyte run my_script.py my_workflow --name "Flyte"
```

Would you like more detail?

Sources:
- <https://github.com/flyteorg/flyte/blob/master/docs/getting_started_with_workflow_development/running_a_workflow_locally.md|Running a workflow locally> :+1: 0 button :-1: 0 button 💡 *Hint:* Mention @runllm in the thread for followups.