Conformal Scorer (Regression)

Compares predictions made by Conformal Regressor with actual values. There are two types of scoring:

• Error rate – the proportion of test instances where the true target is outside the prediction interval. Calculated as 1 - Number of validate predictions/Total
  See Additional prediction information option section for parameters description.
• Mean interval size - the value of mean prediction interval. Calculated as Sum of all interval sizes/Total
  See Additional prediction information option section for parameters description.

Options

Target column

A column that contains the real regression values of the data.

Upper bound

A column that contains prediction upper bounds produced by Conformal Regressor. Should be double column type.

Lower bound

A column that contains prediction lower bounds produced by Conformal Regressor. Should be double column type.

Additional prediction information

Includes additional columns with some prediction metrics to the output table.

• Median interval size – the value of median prediction interval.
• Max interval size – the value of maximum prediction interval.
• Min interval size – the value of minimum prediction interval.

Views

This node has no views

Workflows

• Conformal prediction regression (advanced)KNIME Hub
• Conformal prediction regression (simple)KNIME Hub
• Conformal predictive systems (advanced)KNIME Hub
• Conformal predictive systems (simple)KNIME Hub
• Conformal_regression_by_RedfieldKNIME Hub
• Show all 8 workflows

Links

• No links available

Developers