Hi! Is there a way to cache Node outputs and avoid running it if the cached result exists?

https://kedro.readthedocs.io/en/stable/kedro.io.CachedDataSet.html Maybe this is useful? In general, you should think about what needs to be saved, keeping all nodes’ output will also increase memory consumption.

Thanks for the reply. I'm already using the CachedDataSet, yet when calling

SequentialRunner().run(pipeline)

it executes all nodes. The solution I found was to use

SequentialRunner().run_only_missing(pipeline)