File compare_results.py

eval/compare_results.py:None–None · view source on GitHub ↗

Source from the content-addressed store, hash-verified

1	#!/usr/bin/env python3
2	"""Compare two SWE-bench harness summary reports side-by-side.
3
4	Reads two ``<model>.<run_id>.json`` files produced by

nothing calls this directly

mainFunction · 0.70

exitMethod · 0.45

no test coverage detected