21in7
30ddb2fef4
feat(ml): relax training thresholds for 5-10x more training samples
Add TRAIN_* constants (signal_threshold=2, adx=15, vol_mult=1.5, neg_ratio=3)
as dataset_builder defaults. Remove hardcoded negative_ratio=5 from all callers.
Bot entry conditions unchanged (config.py strict values).
WF 5-fold results (all symbols AUC 0.91+):
- XRPUSDT: 0.9216 ± 0.0052
- SOLUSDT: 0.9174 ± 0.0063
- DOGEUSDT: 0.9222 ± 0.0085
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-21 19:38:15 +09:00
..
2026-03-01 12:48:03 +09:00
2026-03-21 18:36:30 +09:00
2026-03-21 19:38:15 +09:00
2026-03-21 17:26:15 +09:00
2026-03-21 17:26:15 +09:00
2026-03-19 23:37:47 +09:00
2026-03-21 19:38:15 +09:00
2026-03-21 17:26:15 +09:00
2026-03-07 02:20:44 +09:00
2026-03-01 21:25:06 +09:00
2026-03-05 20:25:45 +09:00
2026-03-19 23:10:02 +09:00
2026-03-21 18:36:30 +09:00
2026-03-21 18:31:11 +09:00
2026-03-19 23:10:02 +09:00
2026-03-21 17:26:15 +09:00
2026-03-17 13:47:34 +09:00