Kedro is an open-sourced Python framework for creating maintainable and modular data science code.

Kedro

Hi all,

*What are the requirements for Kedro-viz to allow layer functionality?*

I am currently trying to visualise a single pipeline and have defined a layer attribute for each dataset (including memory datasets) that is used in that pipeline. The layer feature on Kedro Viz however is greyed out and it doesnt let me split the datasets and functions into the layers I defined. Is this because there are datasets not used in the pipeline that dont have layer attributes? (The whole Kedro catalog has >300 datasets so I am hoping that I do not have to define a layer for each one). FYI I am running kedro-viz == 5.2.0

you can use the recently introduced <https://docs.kedro.org/en/stable/data/kedro_dataset_factories.html|Dataset factories > to write way less YAML and apply the same layer to groups of nodes

if you’re on an older version of kedro you can maybe do this with some clever hooks but it’s a bit funky

So I need to have a layer attribute for every dataset in my Kedro repo (including memory datasets)? (I am on Kedro 17.6 btw)

Thanks for the heads up about dataset factories!

dataset factories won’t work for 0.17.x

you may be able to achieve this with the old jinja syntax or hooks