Splitting data into parts

Hi,
I would like to split data into 10 parts. How do I do this in Knime?
Thanks

Do you mean splitting rows or splitting columns? And how do you want to split them?

Probably best if you append some sample data.

1 Like

I built this workflow that takes some data, uses random numbers to sort them and then split hem into a given number of parts and store them in KNIME tables. Still wondering if there is an easier way.

You could also combine an iteration of the Partitioning node.

image

kn_example_split_data_10_parts.knar (892.7 KB)

Thank you for your responses. I have to perform 10 fold cross-validation on 10% of the dataset and test on 90% of the dataset. Also, this has to be repeated 20 times. Looking to perform the cross-validation manually by dividing the dataset.

OK if you want cross-validation there are already nodes for that. You might put a loop around so you could multiply that.

https://www.knime.com/nodeguide/analytics/optimization/cross-validation-with-svm

Or you might want to try out the powerful H2O.ai nodes

https://nodepit.com/workflow/public-server.knime.com%3A80%2F04_Analytics%2F15_H2O_Machine_Learning%2F04_H2O_Crossvalidation

2 Likes

Thank you, will try these out.