Can MLFlow be used without the `with mlflow.start_run()` block?

Question

I want to track an entire notebook and log the parameters of cleaning steps that occur before training a model. I'd like to use mlflow to do it, but on all the docs it looks like you have to track models using this format:

with mlflow.start_run():
    ...

Is there a way to track an entire notebook using mlflow without the with block?

Answer 1

Yes, it's possible to start and end an mlflow run without the context manager syntax.

To do this you can simply write:

mflow.start_run()
...
your_run_code_here
...
mlflow.end_run()

You can read more about it in the mlflow documentation.