Voting Loop End

This node defines the loop end in a sequence of models that are applied to a test dataset. In each iteration the selected prediction columns are stored, and the most frequent value is determined from this collection of prediction columns. A new column containing the majority class of all predictions is appended to the output table. The selected prediction columns are appended (depending on the 'Remove individual predictions' option) to the resulting table; all other columns are not part of the output table.

Options

Winner column

Column that is used to determine the most frequent value.

Remove individual predictions

Removes all prediction columns generated during the loop iterations; only the final, most frequent prediction is appended to the input table.

Views

This node has no views

Workflows

• Tree_EnsemblesKNIME Hub

Links

• voting between multiple classifiersby qqilihq on 2018-06-01

  forum.knime.com/p/43590

Developers