GET R2 FIT TIME-HISTORY¶
Computes a rolling polynomial-fit R² time-history for one or more input curves by sliding a fixed-width window along each curve. Use this worker to quantify how linearly (or how well a chosen polynomial order) a signal behaves locally over time.
When to use¶
Classification: process.
Tagged: curve_quality, goodness_of_fit, linear_fit, polynomial_fit, r2, rolling_window, time_history.
Inputs¶
| Label | ID | Type | Default | Required | Description |
|---|---|---|---|---|---|
| Base Curves | base_curves | vector | — | One or more input time-history curves (x–y vector pairs) on which the rolling R² fit will be computed; accepts multiple curves simultaneously. | |
| Window Size | window_size | string | 0.01 | Width of the sliding window in the same x-axis units as the input curves (default 0.01); controls the local interval over which each polynomial fit is evaluated — smaller values capture faster local changes. | |
| Number Of Digitize Points | num_dig | string | 1000 | Number of evenly-spaced digitisation points used to re-sample each curve before fitting (default 1000); the effective value is max(original point count, this value) to ensure sufficient resolution. | |
| Order | order | string | 1 | Polynomial order used for the local least-squares fit when computing R² (default 1 = linear); increase to 2 or higher to measure goodness-of-fit for nonlinear trends. |
Outputs¶
| Label | ID | Type | Description |
|---|---|---|---|
| R2 histories | curves_get_linear_fit_output_1 | vector | One R² time-history curve per input curve, where each y-value (0–1) represents the coefficient of determination of the polynomial fit within the sliding window centred at that x-position. |
Disciplines¶
- data.curve.transform
- data.signal_processing
- 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_get_linear_fit_r2
Auto-generated from transformation schema. Worker id: curves_get_linear_fit_r2. Schema hash: a29625572b16. Hand-curated docs in workerexamples/ override this page when present.