plot_precision_icc

Note

This page is a reference documentation. It only explains the function signature, and not how to use it. Please refer to the Habitat Guide and Python API guide (v2.0) for usage.

plot_precision_icc(evidence: Any, *, lcl_threshold: float = 0.5, title: str = 'Precision screen: ICC and 95% CI', orientation: Literal['column', 'row'] = 'column') → Figure[source]

Draw per-feature ICC points with 95% CI whiskers.

Each row of evidence is one feature (optionally one experiment). The point is the ICC; the whisker is [lcl, ucl]. Colour marks precise vs unstable. A dashed line is the LCL threshold.

orientation="column" (default) puts features on x and ICC on y (vertical whiskers). orientation="row" flips that for long feature lists (horizontal whiskers).

Parameters:

  • evidence – Long-format table with columns feature, value, lcl, ucl. Optional precise (bool) and experiment. Typically PreciseFeatureSet.to_frame().

  • lcl_threshold – Cutoff drawn on the ICC axis.

  • title – Figure title (English / ASCII).

  • orientation – "column" or "row".

Returns:

The matplotlib Figure; the caller decides where it is saved.

Raises:

Examples using habit.viz.plot_precision_icc

Precise voxel features

Precise voxel features