> If I want to apply the same pipeline to each eml of the partition, I need to provide each time a different namespace as in the cook food example in the doc? Namespacing shouldn't be required with `PartitionedDataset`; that would be if you were to create multiple instantiations of the

Pipeline

object. In your case, you will have one

Pipeline

, one instance of each node in the pipeline, and each node will load/process/save an iterable. If this isn't clear, I can try and whip up a demo later. Maybe you can also just confirm if my understanding that you just want to see one instance of each node in Viz is correct.