Least Confident Uncertainty Scorer

Calculates the least confident uncertainty score of a class probability distribution. Input are rows containing class probabilities that must sum up to 1. The score is calculated as one minus the highest class probability of the data row. Notice: This method is very sensitive to noise!

Options

Column Selection

Include the columns containing the class probabilities. The values must sum up to 1 for each data row.

Output column name

Set the name of the appended output column.

Invalid Input Handling

Specify the action if a data row of the input is invalid. Invalid could mean a missing value in the input or an invalid distribution (the probabilities must sum up to 1). If Fail is selected, the node will fail. Otherwise, the node just gives a warning and puts missing values in the output for the corresponding rows.

Views

This node has no views

Workflows

• No workflows found

Links

• No links available

Developers