I´m currently trying to sum up the outputs of a auto-binner node. Using the mini-workflow shown below.
I´m using the Value Counter to sum up the binned data columns one by one.
Each column has the same number of bins (6) with "equal width".
So now when the data is binned and counted, the empty bins are lost.
If I binned a column and "bin 4" is empty, I´m trying to pass the information that Count(bin 4) = 0 but it is lost by the Value Counter, because he can only count what he can see...
How can I solve this problem?
Thank you a lot for your help!
If you just want to count number of rows for each bin you can use a GroupBy node and then join its output with a manually created table, consisting of all possible bins, so whenever count equals to 0, you will have a missing value for the corresponding bin.
Please, find attached a sample workflow.
thank you a lot for your input.
I already constructed something similar, yet your idea seems to be more elegant, so thank you!
Unfortunately there still remains the problem, that when I change the number of bins in the auto-binner, I will always have to manually change the table in the table in the table-creator-node.
I hoped to somehow extract all the bin-names from the node and paste them into a table using a flow variable(or something like that^^). But I haven´t found a way yet...
Okay, I solved it!
For anyone interested:
I used a code snippet (R in my case) to generate a column with all the possible bins listed)
Before that, I generated a local flow variable with the number of bins, passed it to the auto binner and then on to the R Snippet to the for loop.
This way I can generate a table containing all possible Bin- names (Bin1,Bin 2, ...) and concatenate the table to my real-data-table. That means, that every Bin-Count will be biased by +1, which I can simply substract later (still having the empty bins available).