Option "Max unique values per group" in Moving Aggregator node

[skozer]

Hi, I am fresh in Knime Analytics platform.
When I am trying to understand this option “Max unique values per group” in Moving Aggregator node, I get confused in:

1. Limit value to “1”, doesn’t work on concatenate method, but work on List method - Can I know which ones it can be applied?
2. Even List can be applied, any list with second element will be replaced with missing, evening it is not unique value, why it is happened?

THX.

[Iris]

Hi @skozer

this parameter is mostly for speeding up the node. It controls a internal size.

Limiting to something small would not be helpful. It should be put to a big value. Typical leaving the default is just fine. But if you have columns with a lot of different values, you might want to increase it.

[skozer]

@Iris
Many thanks for your answer. Now I am clear on the applicatipn you mentioned.
I still want to dig a little bit more on what kind of aggregation methods it can be applied?

[Iris]

It cannot be really applied or used. It is more limiting your calculations, to improve the overall speed.

Let’s assume you want to calculate the number of customers per product group. You will use the Unique Count Option. And this only works if you have less than 10.000 unique customers. If you have more, the node will return a missing value and give you a warning message.