graham/prop - Forgejo: Beyond coding. We forge.

graham/prop

Fork 0

Commit graph

Author	SHA1	Message	Date
Graham McIntire	2fd88a94ea	fix(propagation): 7 pipeline bugs + validation harnesses + model improvements Some checks failed Build and Push / Build and Push Docker Image (push) Waiting to run Details Build prop-grid-rs / Test, build, push (push) Failing after 3m41s Details Phase A — 7 pipeline bugs: - retain_scores_window: fix timeline self-deletion (NOTIFY payload run_time\|valid_time) - Rust/Elixir weight divergence: Rust loads band_weights.json at startup - Aurora boost ported to Rust (Kp query + per-cell boost for bands <= 432 MHz) - Commercial-link boost applied in Rust per-cell scoring - f00 native gradient preferred over pressure-level gradient - HRDPS files: greedy regex fixed, now visible to timeline/prune/retain - GEFS/HRRR collision: GEFS namespace as .gefs.prop, merge at read Phase B — Validation harness: - scripts/validate_algo.py: out-of-sample Spearman rho(score,distance) + baselines - docs/algo-reports/validation-2026-08-01.{json,md} Phase C — Forecast-skill evaluation: - scripts/validate_forecast.py: skill degradation by lead time (0h-24h) - docs/algo-reports/forecast-2026-08-01.{json,md} Phase D — Calibration + model improvements: - recalibrate.py: validation gate before weight deploy - Recalibrator: Nx.max(0.0) replaces Nx.abs(), L2 regularization, val-set integrity - ML train/serve defaults unified; prop_compare null-handling skew fixed - Latitude-aware sunrise: solar-declination sunrise_hour(lat,month) (Elixir + Rust) - Path scoring: wind/sky/rain from HRRR (was ~30% of composite weight silent) - Region multiplier documented as unvalidated; PWAT/refractivity doc-vs-code noted - algo.md: synced scoring sections, marked retired features, D5/D6 changelog - Credo: path_compute cyclomatic complexity bumped (6->10 fields) — informational only	2026-08-01 19:28:41 -05:00
Graham McIntire	e7fb34eb6f	test: add /qsos→/contacts redirect + scorer conversion properties - PageController: cover the legacy `/qsos` → `/contacts` redirect pair (both the index and the id-preserving per-resource redirect), which had no coverage since the route rename. - Scorer property tests: - f_to_c / c_to_f are inverses up to float epsilon, both propagate nil, and the C→F slope is exactly 5/9. - wind_speed_kts is non-negative, propagates nil on either axis, and matches sqrt(u²+v²)·1.94384 (the 5-digit kts factor the scorer uses). - precip_to_rate_mmhr is non-negative, 0 for nil/zero/negative. - dbz_to_rain_rate_mmhr is 0 below 5 dBZ (and for nil), and monotonically non-decreasing above — guards against accidental inversion of the Marshall-Palmer lookup. Suite: 2,382 tests + 155 properties (was 2,380 + 148); coverage 71.08% → 71.12%; credo strict clean.	2026-04-23 14:03:02 -05:00
Graham McIntire	c60d56281f	test(property): add property tests for propagation scoring + CIDR matcher	2026-04-21 13:57:12 -05:00

Author

SHA1

Message

Date

Graham McIntire

2fd88a94ea

fix(propagation): 7 pipeline bugs + validation harnesses + model improvements

Build and Push / Build and Push Docker Image (push) Waiting to run

Details

Build prop-grid-rs / Test, build, push (push) Failing after 3m41s

Details

Phase A — 7 pipeline bugs:
- retain_scores_window: fix timeline self-deletion (NOTIFY payload run_time|valid_time)
- Rust/Elixir weight divergence: Rust loads band_weights.json at startup
- Aurora boost ported to Rust (Kp query + per-cell boost for bands <= 432 MHz)
- Commercial-link boost applied in Rust per-cell scoring
- f00 native gradient preferred over pressure-level gradient
- HRDPS files: greedy regex fixed, now visible to timeline/prune/retain
- GEFS/HRRR collision: GEFS namespace as .gefs.prop, merge at read

Phase B — Validation harness:
- scripts/validate_algo.py: out-of-sample Spearman rho(score,distance) + baselines
- docs/algo-reports/validation-2026-08-01.{json,md}

Phase C — Forecast-skill evaluation:
- scripts/validate_forecast.py: skill degradation by lead time (0h-24h)
- docs/algo-reports/forecast-2026-08-01.{json,md}

Phase D — Calibration + model improvements:
- recalibrate.py: validation gate before weight deploy
- Recalibrator: Nx.max(0.0) replaces Nx.abs(), L2 regularization, val-set integrity
- ML train/serve defaults unified; prop_compare null-handling skew fixed
- Latitude-aware sunrise: solar-declination sunrise_hour(lat,month) (Elixir + Rust)
- Path scoring: wind/sky/rain from HRRR (was ~30% of composite weight silent)
- Region multiplier documented as unvalidated; PWAT/refractivity doc-vs-code noted
- algo.md: synced scoring sections, marked retired features, D5/D6 changelog
- Credo: path_compute cyclomatic complexity bumped (6->10 fields) — informational only

2026-08-01 19:28:41 -05:00

Graham McIntire

e7fb34eb6f

test: add /qsos→/contacts redirect + scorer conversion properties

- PageController: cover the legacy `/qsos` → `/contacts` redirect pair
  (both the index and the id-preserving per-resource redirect), which
  had no coverage since the route rename.
- Scorer property tests:
  - f_to_c / c_to_f are inverses up to float epsilon, both propagate
    nil, and the C→F slope is exactly 5/9.
  - wind_speed_kts is non-negative, propagates nil on either axis,
    and matches sqrt(u²+v²)·1.94384 (the 5-digit kts factor the
    scorer uses).
  - precip_to_rate_mmhr is non-negative, 0 for nil/zero/negative.
  - dbz_to_rain_rate_mmhr is 0 below 5 dBZ (and for nil), and
    monotonically non-decreasing above — guards against accidental
    inversion of the Marshall-Palmer lookup.

Suite: 2,382 tests + 155 properties (was 2,380 + 148); coverage
71.08% → 71.12%; credo strict clean.

2026-04-23 14:03:02 -05:00

Graham McIntire

c60d56281f

test(property): add property tests for propagation scoring + CIDR matcher

2026-04-21 13:57:12 -05:00

3 commits