Hi, I’ve got a dataset which looks like this [image] and so on. I would like to create a table which looks more like this: [image] and so on. So sort of like transpose it but at the same time using something like groupby, because I want to transpose it but because the cells are several times repeating, if I transpose it I get as many columns as I have rows but I only want as many columns as cells I have, then list the values of each cell below its own column. Also, it would be great if I could merge headers as shown but if it’s impossible it’s also good if it just says cella_1 over the time and the voltage in each column. I thought about loops or something but i really am a beginner KNIME user so I just can’t figure this out. I hope I was clear and I would be grateful if you could help me with this. Thank you

Hello @BendegzHegedus Your challenge output results to be a ‘nested hierarchy in column header’ table. And as you pointed out, it’s somehow based in grouping: then the best choice is a Group Loop node and some column name rearrangement. What is conceptually missed in your description is an output ‘row sense’ column, further than a row ID. In this case could be something like measurement_round … Your workflow could be something like this; considering the data is sorted by time: [image] Data Wrangling _ Nesting Hierarchy in Column Header – gonhaddock Knime Forum: https://forum.knime.com/t/create-columns-from-rows/47972 [image] [image] BR PS.- Be aware that once your dada is arranged in this way, it couldn’t be usable for some analysis or visualizations, and return it back to original shape can require complex back transformations.

create columns from rows

KNIME Analytics Platform

mlauber71 October 9, 2022, 9:46pm 6

@BendegzHegedus you could try and adapt this example of how to create an Excel file with a complex header