report
report(result, path, *, title=None, ground_truth_tool=None, registry=None, funky_heatmap=True, metric_groups=None)
Write an HTML report for result to path.
Parameters
| Name | Type | Description | Default |
|---|---|---|---|
result |
RunResult | The RunResult from beam.rank. |
required |
path |
str | Path | Output file path. | required |
title |
str | None | Optional report title. Defaults to a generated one. | None |
ground_truth_tool |
str | None | Optional name of the tool documented to rank first, drawn on the ranking figure for comparison. Used by vignettes that carry a known truth; left None for a plain benchmark CSV. |
None |
registry |
Registry | None | Optional Registry for the per-dataset critical-difference computation. Defaults to a fresh registry. |
None |
funky_heatmap |
bool | Include the funky-heatmap glyph table with its rank-robustness panels (leave-one-dataset-out span, SMAA acceptability, aggregation consensus). Default True. Set False for a leaner report or when the matplotlib glyph table is not wanted. | True |
metric_groups |
Sequence[str] | None | Optional group label per metric, in the order of result.metric_ids, used to colour the funky-heatmap columns. Ignored when funky_heatmap is False. |
None |