COMPUTE ERROR BETWEEN TWO SETS OF CURVES BASED ON R2, MSD, NMAD, RE

Computes a scalar error metric between two sets of curves using one of eight distance/difference measures: NMAD, MSD, RE, R², Fréchet, Euclidean, DTW, or Partial DTW. Use this worker to quantify the agreement between reference and comparison curve sets in a simulation-vs-test or model-vs-model context.

When to use

Classification: process.

Tagged: curve_comparison, curves, dtw, error_metric, euclidean, frechet, goodness_of_fit, msd.

Inputs

Label ID Type Default Required Description
Reference Curves ref_curves vector   One or more reference (baseline) curves — e.g. physical-test or target simulation time-histories — against which the comparison curves are measured; accepts multiple vector curves.
Compare Curves compare_curves vector   One or more curves to evaluate against the reference set — e.g. simulation results or candidate model outputs; must be on a compatible domain to the reference curves.
Error Type type_of_error scalar nmad   Error metric to apply: nmad (Normalized Mean Absolute Difference, default), msd (Mean Square Difference), re (Relative Error), r2 (Coefficient of Determination), frechet (Fréchet Distance), euclidean (Euclidean Distance), dtw (Dynamic Time Warping), or pdtw (Partial Dynamic Time Warping); leave as ‘nmad’ when a normalised, unit-independent scalar summary is needed.

Outputs

Label ID Type Description
Error curves_error scalar Scalar error value computed between the reference and comparison curve sets using the selected metric; dimensionless for NMAD/RE/R²/DTW variants, or in curve-units² for MSD.

Disciplines

  • data.correlation
  • data.curve.pair
  • data.statistics

Runnable example

A runnable example is registered for this worker. Open the example workflow on the d3VIEW canvas: /api/workflow/example?id=curves_error


Auto-generated from transformation schema. Worker id: curves_error. Schema hash: 447bebe6175c. Hand-curated docs in workerexamples/ override this page when present.