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

graham/prop

Fork 0

Commit graph

Author	SHA1	Message	Date
Graham McIntire	fd976b0cd5	fix: resolve 391 Credo issues across codebase - Add jump_credo_checks ~> 0.4 with all 20 checks enabled - Fix all standard Credo issues: 139 @spec (113 done, 26 remain), 4 refactoring, 3 alias usage, 9 System.cmd env, 5 unsafe_to_atom, 2 max line length, 9 assert_receive timeout - Fix 170+ jump_credo_checks warnings: - 117 TopLevelAliasImportRequire: move nested alias/import to module top - 32 UseObanProWorker: switch to Oban.Pro.Worker - 4 DoctestIExExamples: add doctests / create test file - ~20 WeakAssertion: strengthen type-check assertions - Various ConditionalAssertion, AssertReceiveTimeout fixes - Exclude vendor/ from Credo analysis - Remaining: 175 warnings (mostly opinionated WeakAssertion, AvoidSocketAssignsInTest), 26 @spec annotations	2026-06-12 13:51:32 -05:00
Graham McIntire	5fd37a17fd	feat(propagation): per-band weight calibration from full-corpus correlation Ran recalibrate_algo.py against the full local prop_dev (81,994 contacts, 18.6M HRRR rows) and derived per-band composite weights for the nine bands with >=200 matched contacts. Moisture (dewpoint/PWAT/surface N) is consistently beneficial through 5.76 GHz, reverses at 24 GHz; rain scales sqrt(rain_k) not linearly so 24 GHz gets 0.215 rain weight instead of the linear-ratio 0.95. 10 GHz stays as the reference band (defaults); 47+ GHz inherits defaults (n<200). - BandConfig.weights/1 returns per-band override or default fallback - @band_configs carries :weights on 222/432/902/1296/2304/3400/5760/24G - Recalibrator.compute_factors/3 + fit(band_mhz:) band-aware fitting - Scorer + ContactLive.Show pass band_config to weights/1 - algo.md Part 2d documents the 2026-04-18 analysis + derivation rule - scripts/derive_band_weights.py turns correlations into weight maps - report preserved at docs/algo-reports/2026-04-18-recalibration.md	2026-04-18 10:19:26 -05:00
Graham McIntire	e7a7ae073d	Phase 9.3, 9.4, and Phase 3 NEXRAD pipeline Task 9.3 - Weight recalibration via gradient descent: - Recalibrator module fits logistic regression weights using Nx - Trains on QSO positives vs random baseline negatives - Cross-validates by month, normalizes weights to sum to 1.0 - Mix task: mix recalibrate_scorer --sample 5000 --epochs 2000 Task 9.4 - Side-by-side scorer comparison: - ScorerDiff.compare/3 re-scores grid with old vs new weights - Reports mean diff, regressions, improvements, per-band breakdown - Mix task: mix scorer_diff --new-weights '{...}' Phase 3 - NEXRAD ingestion pipeline: - NexradClient fetches IEM n0q composite PNGs, extracts per-point box statistics (mean/max dBZ, texture variance) - NexradObservation schema with unique (lat, lon, observed_at) - NexradWorker on :nexrad queue for background processing - nexrad_texture backtest feature in Features module - mix nexrad_backfill --limit 200 All tasks added to AdminTaskWorker and Release for production use. 1116 tests, 0 failures.	2026-04-10 12:48:36 -05:00

Author

SHA1

Message

Date

Graham McIntire

fd976b0cd5

fix: resolve 391 Credo issues across codebase

- Add jump_credo_checks ~> 0.4 with all 20 checks enabled
- Fix all standard Credo issues: 139 @spec (113 done, 26 remain),
  4 refactoring, 3 alias usage, 9 System.cmd env, 5 unsafe_to_atom,
  2 max line length, 9 assert_receive timeout
- Fix 170+ jump_credo_checks warnings:
  - 117 TopLevelAliasImportRequire: move nested alias/import to module top
  - 32 UseObanProWorker: switch to Oban.Pro.Worker
  - 4 DoctestIExExamples: add doctests / create test file
  - ~20 WeakAssertion: strengthen type-check assertions
  - Various ConditionalAssertion, AssertReceiveTimeout fixes
- Exclude vendor/ from Credo analysis
- Remaining: 175 warnings (mostly opinionated WeakAssertion,
  AvoidSocketAssignsInTest), 26 @spec annotations

2026-06-12 13:51:32 -05:00

Graham McIntire

5fd37a17fd

feat(propagation): per-band weight calibration from full-corpus correlation

Ran recalibrate_algo.py against the full local prop_dev (81,994 contacts,
18.6M HRRR rows) and derived per-band composite weights for the nine
bands with >=200 matched contacts. Moisture (dewpoint/PWAT/surface N)
is consistently beneficial through 5.76 GHz, reverses at 24 GHz; rain
scales sqrt(rain_k) not linearly so 24 GHz gets 0.215 rain weight
instead of the linear-ratio 0.95. 10 GHz stays as the reference band
(defaults); 47+ GHz inherits defaults (n<200).

- BandConfig.weights/1 returns per-band override or default fallback
- @band_configs carries :weights on 222/432/902/1296/2304/3400/5760/24G
- Recalibrator.compute_factors/3 + fit(band_mhz:) band-aware fitting
- Scorer + ContactLive.Show pass band_config to weights/1
- algo.md Part 2d documents the 2026-04-18 analysis + derivation rule
- scripts/derive_band_weights.py turns correlations into weight maps
- report preserved at docs/algo-reports/2026-04-18-recalibration.md

2026-04-18 10:19:26 -05:00

Graham McIntire

e7a7ae073d

Phase 9.3, 9.4, and Phase 3 NEXRAD pipeline

Task 9.3 - Weight recalibration via gradient descent:
- Recalibrator module fits logistic regression weights using Nx
- Trains on QSO positives vs random baseline negatives
- Cross-validates by month, normalizes weights to sum to 1.0
- Mix task: mix recalibrate_scorer --sample 5000 --epochs 2000

Task 9.4 - Side-by-side scorer comparison:
- ScorerDiff.compare/3 re-scores grid with old vs new weights
- Reports mean diff, regressions, improvements, per-band breakdown
- Mix task: mix scorer_diff --new-weights '{...}'

Phase 3 - NEXRAD ingestion pipeline:
- NexradClient fetches IEM n0q composite PNGs, extracts per-point
  box statistics (mean/max dBZ, texture variance)
- NexradObservation schema with unique (lat, lon, observed_at)
- NexradWorker on :nexrad queue for background processing
- nexrad_texture backtest feature in Features module
- mix nexrad_backfill --limit 200

All tasks added to AdminTaskWorker and Release for production use.
1116 tests, 0 failures.

2026-04-10 12:48:36 -05:00

3 commits