regulatory_parity¶

regulatory_parity ¶

EPA AERMOD test-suite parity helpers.

Used to score a pyaermod-produced POSTFILE against an EPA reference POSTFILE from the bundled AERMOD test-case distribution (test_cases/aermet_24142_aermod_24142/postfiles/*.PST).

The scoring metric is the best-fit slope of paired (reference, candidate) concentrations through the origin, matching EPA's Compare_AERMOD_test_cases.R script. The script's published acceptance margin is ±0.001 around 1.0; we expose that as :data:DEFAULT_SLOPE_TOLERANCE and the convenience predicate :func:passes_parity.

Pairing is done on (x, y, date, ave, grp) join keys.

ParityScore `dataclass` ¶

Outcome of comparing a candidate POSTFILE to an EPA reference.

passes ¶

passes(tolerance: float = DEFAULT_SLOPE_TOLERANCE) -> bool

True if best-fit slope is within tolerance of 1.0.

score_postfile_pair ¶

score_postfile_pair(reference: Union[str, Path], candidate: Union[str, Path], *, case: str = '') -> ParityScore

Score a candidate POSTFILE against an EPA reference POSTFILE.

Both files must be in the same format (PLOT or UNFORM); the auto-detect in :func:pyaermod.postfile.read_postfile is applied to each.

Pairing is on (x, y, date, ave, grp). Receptors / timesteps that appear in only one file are dropped before scoring (their count is not subtracted from n_paired).

passes_parity ¶

passes_parity(reference: Union[str, Path], candidate: Union[str, Path], *, tolerance: float = DEFAULT_SLOPE_TOLERANCE) -> bool

Convenience: True if the candidate's best-fit slope is within tolerance.

regulatory_parity¶