Weak Label Model Predictor

Applies a weak label model learned with the Weak Label Model Learner to data. For each row, it appends the probability distribution over the class values as predicted by the weak label model. This node is a key component for the realization of weak supervision approaches as popularized by Snorkel . For more information, please see the Weak Label Model Learner.

Options

Class column name

The name of the label (or class) column that the labels provided by the label sources belong to. This value is used to create the column names for the probability columns in the output.

Append individual class probabilities

Select this option if you want to append the class probabilities as individual columns in addition to the class probability distribution column.

Remove source columns

Select if you want to remove the label source columns from the output table.

Views

This node has no views

Workflows

• 01_Guided_Labeling_for_Document_ClassificationKNIME Hub
• 01_Guided_Labeling_for_Document_Classification-FIXED-HEADERKNIME Forum
• 03_Weak_Supervision_for_Document_ClassificationKNIME Hub
• 05_Weak_Supervision_on_the_Adult_datasetKNIME Hub

Links

• No links available

Developers