COMPUTE THE DIFFERENCE BETWEEN CURVES AND A BASELINE¶
Computes the pairwise difference (or distance) between every curve in a set and a designated baseline curve using a configurable match metric (raw, absolute, Euclidean, Fréchet, DTW, etc.). Use this worker when you need a quantitative similarity or deviation score for a batch of curves against a single reference.
When to use¶
Classification: process.
Tagged: baseline, batch, curve_match, curves, difference, digitize, distance, dtw.
Inputs¶
| Label | ID | Type | Default | Required | Description |
|---|---|---|---|---|---|
| Curves | curves | vector | — | Collection of input curves (vector of curve objects) to be compared against the baseline; supply two or more curves — at least one non-baseline curve is required for a meaningful result. | |
| Baseline Curve Index | baseline_curve_index | scalar | 0 | Zero-based integer index into the ‘curves’ collection that identifies the reference (baseline) curve; defaults to 0 (first curve). | |
| Match Type | match_type | list | raw | Distance/difference metric to apply: ‘raw’ (signed point-wise diff), ‘abs’ (absolute diff), ‘max’ (peak difference), ‘max-max-diff’ (max-to-max delta), ‘euclidean-distance’, ‘frechet’, ‘dynamic-time-warping’, or ‘pdtw’; defaults to ‘raw’. | |
| Digitize | digitize | list | 10 | Number of evenly-spaced points to re-sample each curve to before matching (0 = no resampling); choose higher values (200–1000) for smoother metrics on dense signals; defaults to 10. | |
| Normalize | normalize | list | no | Whether to normalize each curve to [0, 1] before computing the match metric (‘yes’ / ‘no’); set to ‘yes’ when curves have different amplitude scales; defaults to ‘no’. | |
| Xmin | xmin | scalar | 0 | Skip values before this point | |
| Xmax | xmax | scalar | 100000000000000000000 | Skip values after this point | |
| Sync | sync | list | yes | ||
| Enforce Monotonicity | skip_drop | list | no | Options to choose show to clip curves before comparing | |
| Diff Ratio Threshold | diff_ratio_threshold | scalar | 0.1 | Only applies if the diff type is ratio based | |
| Remove Outlier Points | remove_outliers | list | no | Remote outliers | |
| Distance Measurement Type | distance_measurement_type | list | euclidean | Only applied for Frechet, DTW, pDTW | |
| Scale By Area Diff | scale_by_area_diff | select | no | The match error is multiplied by integral difference between the two curves | |
| Clip Method | clip_method | list | old | Only DTW, pDTW |
Outputs¶
| Label | ID | Type | Description |
|---|---|---|---|
| curves_match_all_output_1 | curves_match_all_output_1 | dataset | Dataset containing one row per non-baseline curve with the computed match/distance score(s) relative to the baseline curve, along with curve identifiers and the chosen metric name. |
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_match_all
Auto-generated from transformation schema. Worker id: curves_match_all. Schema hash: 5d6434f338ff. Hand-curated docs in workerexamples/ override this page when present.