graham/prop
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
1968 commits 2 branches 0 tags 10 MiB
Commit graph

6 commits

Author SHA1 Message Date
Graham McIntire
959f3a3f69
Fix LiveView crashes and stabilize test suite
Some checks failed
Build and Push / Build and Push Docker Image (push) Failing after 4m31s
Details
2026-08-01 09:26:47 -05:00
Graham McIntire
079346a1b9 fix: resolve all 281 credo issues across source and test files
Some checks failed
Build and Push / Build and Push Docker Image (push) Failing after 4m38s
Details 
- F-level (228): replace length/1 with Enum.count_until/2 or pattern
  matching; convert Enum.flat_map+if to Enum.filter+Enum.map; fix
  identity case in narr_client
- W-level (46): normalize dual atom/string key access in weather_layers,
  beacon_measurements, surface, skewt_live, contact_live/show; add
  :data_provider and weather-map assigns to ignored_assigns in credo
  config (consumed by child components credo can't trace); remove
  weak is_list assertion; remove explicit assert_receive timeout
- R-level (6): replace 'This module provides...' moduledocs with
  meaningful descriptions
- Also fix 4 compile-connected xref issues by deferring
  BandConfig.band_options() from module attribute to runtime

Co-Authored-By: Claude <noreply@anthropic.com>
 2026-07-29 09:04:21 -05:00
Graham McIntire
0d00d1777c
fix: 9 bugs across tests and production code 
- Fix Module.safe_concat -> Module.concat in tests with dynamic process names
  (safe_concat calls binary_to_existing_atom, but test names are newly generated)
- Fix PSKR AggregatorTest sandbox ownership by switching to async: false
- Fix MapLiveTest assert_patch: regex unsupported in LiveView 1.2, use string match
- Fix WeatherMapLiveTest toggle_grid: assert on checked attribute, not data-grid
- Fix BeaconLive.Index leaking unapproved beacons: wire data_provider to approved_beacons_query
- Fix ContactLive.Index leaking private contacts: wire data_provider with visibility filter
- Fix RecalibratorTest: train() expects factor vectors, not {vector, datetime, band} tuples
- Fix toggle_sort: compare current_field to new_field, not current_order
- Fix internal_network?: handle both atom and string session keys from Plug sessions

Test results: 3979/4001 -> 3997/4001 (18 previously-failing tests now pass)
 2026-06-21 12:13:58 -05:00
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