File eval_size_routing.py

tools/experiments/eval_size_routing.py:None–None · view source on GitHub ↗

Source from the content-addressed store, hash-verified

1	"""Evaluate cheap-vs-capable model agreement across size buckets.
2
3	Sample manpages stratified by gz size, extract with two models,
4	compute per-file option-set agreement, aggregate by bucket.

nothing calls this directly

mainFunction · 0.70

no test coverage detected