Hello, I was going through the tutorial on visualizing pipelines in notebooks () and found it quite insightful, especially for data teams that primarily work within notebook environments. However, I’m a bit unsure about the recommended approach for running a pipeline when it's defined directly within a notebook. I understand that normally, executing a pipeline requires initializing a

KedroSession

. But in cases where the

Pipeline

and

DataCatalog

is already defined in the notebook itself, what would be the best practice for running it?

Hi @Felipe Monroy, you can add them like this

catalog[param_name] = param_value

as long as you’re able to create

params_dict,

which should look something like this:

params_dict = {
  "params:a": {"b": 1},
  "params:a.b": 1
}

Thank you! I watched the video and thought about the possibility of using Kedro entirely within a notebook. I believe the only missing part is the feed dictionary, everything else works perfectly