Hi, I'm wondering about the kedro-boot plugin. In the example, you show really well how to run pipelines, but I'm wondering what's the best practice to access a dataset? I managed to do it using:

some_number = (
    get_kedro_boot_session("project")
    ._context.catalog._get_dataset("features_store")
    .load()
    .iloc[0][0]
)

(In the example project), is it how it should be done?

Hi ! if i understand well you question, you want to manipulate the dataset directly within your app ? you could use the Kedro Catalog in a Standalone way :

config_loader = OmegaConfigLoader()
conf_catalog = config_loader["catalog"]
catalog = DataCatalog.from_config(conf_catalog)
feature_store = catalog.load("feature_store")

Or you can create a pipeline namespace that return your desired results, then you can call it with

some_number = session.run("your_namespace")

Your code is still valid, but it touch some private attributes, it may break in futur releases.

Great, thanks!