GET FIT SLOPES

Segments an input curve into piecewise linear regions and extracts the best-fit slope for each segment using a sliding-window linear regression. Use this worker to identify dominant slope regimes in stress-strain, force-displacement, or similar engineering curves, with optional filtering to a target slope value.

When to use

Classification: process.

Tagged: curve, digitization, fit_slopes, linear_fit, piecewise, r2, regression, segmentation.

Inputs

Label ID Type Default Required Description
Input Curve input_curve vector   The XY curve to analyse (e.g. force-displacement or stress-strain); supplied as a two-column vector — leave empty only when chaining from an upstream curve output.
Num Dig num_dig integer 1000   Number of uniformly spaced points used to re-digitize the input curve before segmentation; default 1000 — increase for high-frequency curves, decrease for speed.
Segment Length Ratio segment_length_ratio integer 0.01   Fraction of the total re-digitized point count that defines the sliding-window width for local linear fitting (e.g. 0.01 = 1 %); smaller values yield finer slope resolution.
Min R2 r2_limit integer 0.95   Minimum R² goodness-of-fit threshold (0–1) a segment must meet to be retained; default 0.95 — lower to accept noisier linear regions.
Merge Ratio merge_ratio integer 0.025   Relative tolerance used to merge adjacent segments whose slopes differ by less than this fraction (e.g. 0.025 = 2.5 %); reduces duplicate near-identical slope entries.
Target Slope target_slope integer 0   Reference slope value around which results are filtered; set to 0 (default) to disable target filtering and return all detected slopes.
Target Percentage target_percentage integer 0.05   Fractional tolerance around target_slope — only slopes within ±target_percentage of target_slope are returned (e.g. 0.05 = ±5 %); ignored when target_slope is 0.
Maximize Params maximize_params scalar slope,r2    

Outputs

Label ID Type Description
curve_get_fit_slopes_output_1 curve_get_fit_slopes_output_1 dataset Dataset table of detected piecewise linear segments, each row containing the x-start, x-end, fitted slope, and R² value for that segment, sorted by descending slope·R² score.

Disciplines

  • ai_ml.supervised.regression
  • data.curve.transform
  • data.statistics
  • engineering.material.characterization

Runnable example

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


Auto-generated from transformation schema. Worker id: curve_get_fit_slopes. Schema hash: f1e365ee47b4. Hand-curated docs in workerexamples/ override this page when present.