Graham McIntire a6b89961f6 feat(ml): retrain propagation model and add per-prediction explainability ... Retrained on 60k month-balanced HRRR profiles through 2026-04-05 (test RMSE 1.7 score points, R² 0.97). Added explain_prediction/4 returning ranked feature contributions via batched finite-difference attribution, so the UI can show users which weather factors drove each prediction.		2026-04-28 14:14:22 -05:00
..
fixtures	NarrClient.extract_profile_from_file/3 + fetch_profile_at/2	2026-04-15 18:54:02 -05:00
microwaveprop	feat(ml): retrain propagation model and add per-prediction explainability	2026-04-28 14:14:22 -05:00
microwaveprop_web	feat(rover-locations): add /rover-locations page with CRUD	2026-04-26 13:08:27 -05:00
mix/tasks	refactor(weather): make rebatch logic callable from release eval	2026-04-24 13:05:52 -05:00
support	feat(rainscatter): classify QSO propagation mechanism from common-volume radar	2026-04-17 15:57:59 -05:00
test_helper.exs	test: silence expected warnings/errors during test runs	2026-04-24 09:27:17 -05:00