Refactor/adforce eval - #2

Merged

sdat2 merged 17 commits into

mainfrom

refactor/adforce-eval

Aug 18, 2026

Commits on Aug 16, 2026

move comp to eval
sdat2
committed
678f00f
appendix
sdat2
committed
54d9d1a
hydra-ize adforce.eval: config tree, 5 mains, lazy dirs, drift tests

sdat2
and
claude
committed
9386b05
continued refactor
sdat2
committed
7051fbf
eval pairs + matrix grammar: model-vs-model, published-artifact pins

sdat2
and
claude
committed
ff5cccf
eval launch orchestration: drive_storm seam, plan/dry-run, harvest

sdat2
and
claude
committed
88d2d15
finished eval move
sdat2
committed
6f0f49a
eval Phase 2a: tide-on obs machinery, tide validation, TWL/skew scoring

sdat2
and
claude
committed
e1e0eee
eval Phase 2b: constituent-level tidal diagnosis + verdict

sdat2
and
claude
committed
0ead349

Commits on Aug 17, 2026

eval: Manning's-n friction axis + the tidal-diagnosis sweep (GCP-only)

sdat2
and
claude
committed
d067e62
gcp: mannings_sweep.sh runner + PE*-strip dirfix in eval.launch

sdat2
and
claude
committed
16f750e
gcp: fix adcirc image build -- symlink conda's prefixed ar/ranlib/nm

sdat2
and
claude
committed
dbfd1d3
eval.launch: clear GlobalHydra before wrap compose (GCP launch crash)

sdat2
and
claude
committed
15b5921
eval: friction_cf axis (fort.15 CF) -- the mannings sweep was a NWP=0 no-op

sdat2
and
claude
committed
49f8232

Commits on Aug 18, 2026

tidal diagnosis: friction_cf sweep results -- semidiurnal knob, diurnal saturation

sdat2
and
claude
committed
14fbc6c
eval: friction_storm matrix -- surge sensitivity to the CF sweet spot

sdat2
and
claude
committed
50baf34
tidal diagnosis closed: HAMTIDE exonerated, surge vetoes CF change

sdat2
and
claude
committed
bfca70e