summary

Overview ¶

Summary reads results.tsv and prints, per case, models, effort and judge, how many runs there were, the pass rate (the share of runs earning every point), the mean score and its range, and what a run used: one run says little, since a model answers differently from one run to the next, and a best run hides the others (ADR-0014). Fewer than five runs are marked `few runs`; scores earned by a model that no longer answers on this machine (the models-seen file, ADR-0004) are marked `replaced`.

go run ./tests/evaluation/summary [results.tsv]