Commit graph

38ade3c1dd Upload files to "/" Scalperfaisal99 Faisal Pratama 2026-08-18 21:38:51 +00:00
10253b581b feat(inputs): private-build defaults = the drop-on-chart meta-pooling workflow 190552 AnimateDread 2026-08-13 16:31:15 -04:00
1bf3eba68a feat(meta): self-contained corpus - the META chart sweeps the real classic ladders over its own history AnimateDread 2026-08-13 16:19:43 -04:00
932c94e890 feat(research): offline meta-pool pipeline - loader, EA-mirrored splits, MLP, cov x (p-BE) eval AnimateDread 2026-08-13 16:09:14 -04:00
d20058fc1b feat(meta): dataset export for offline cross-sectional pooled training AnimateDread 2026-08-13 13:14:12 -04:00
4db223b40e docs: H4 experiment result - H3 (lift does not transfer; single-instrument well closed) AnimateDread 2026-08-13 13:06:46 -04:00
0fb53d8979 feat(meta): pin the corpus DB to the chart's symbol+timeframe; pre-register the H4 experiment AnimateDread 2026-08-13 11:11:38 -04:00
0adaea48b6 fix(resume): model reload stalled training - three hardenings on the resume path AnimateDread 2026-08-13 10:23:11 -04:00
df48c37f65 feat: per-family x per-side OOS breakdown in the META era report AnimateDread 2026-08-13 09:03:40 -04:00
444909d0a3 feat: S2 meta-labeling head - binary trade-quality model over the classic-candidate corpus AnimateDread 2026-08-13 06:52:31 -04:00
68459208bf fix(meta): make the stale-DB corpus warning unmissable in the tester AnimateDread 2026-08-12 23:56:26 -04:00
aeed4037d5 fix(db): add the missing FetchRecordCount passthrough on CDatabaseManager AnimateDread 2026-08-12 19:02:11 -04:00
6819bb4133 perf(db): targeted SQL lookups replace full-table fetches per signal AnimateDread 2026-08-12 18:53:04 -04:00
7d3c20d7f1 fix(meta): warn when a corpus backtest starts before the DB''s newest row AnimateDread 2026-08-12 16:20:53 -04:00
4507ea69a9 feat(meta): S1 - the signal DB becomes the meta-label training corpus AnimateDread 2026-08-12 15:20:33 -04:00
4937d4763b docs: meta-labeling design - the NN predicts trade quality, not direction AnimateDread 2026-08-12 15:08:41 -04:00
b06fdd2f0e research: point the seasonal screen at the five breadth instruments AnimateDread 2026-08-12 14:08:05 -04:00
83fa5e4d3d fix(db): a completed DML step is not a failure AnimateDread 2026-08-12 13:26:40 -04:00
195be2025b fix(db): the log no longer asks the decision layer for permission AnimateDread 2026-08-12 12:02:05 -04:00
c8ef478ce8 fix(db): reversing signals register their own trade (true stop-and-reverse) AnimateDread 2026-08-12 11:32:36 -04:00
652bf81112 fix(db): per-side pattern journaling + versioned journaling semantics AnimateDread 2026-08-12 10:37:57 -04:00
36e8463310 refactor: derive history bars for input sequences and update related configurations AnimateDread 2026-08-11 21:53:37 -04:00
923addf574 feat: pin the cross-asset pair set train->serve + warm the sync at init AnimateDread 2026-08-11 21:29:14 -04:00
ccc3dce69e feat: index-mode cross-asset encoding - base==quote wasted 3 of 6 slots AnimateDread 2026-08-11 21:07:52 -04:00
53ccc03453 fix: the trailing-incumbent count gate was unpassable by construction AnimateDread 2026-08-11 18:50:39 -04:00
bd1037975a fix: the trailing incumbent read the future across eras; cold AD blocks cached zeros as truth AnimateDread 2026-08-11 18:20:24 -04:00
77e8080cfe fix: four risk-layer holes a funded account would eventually find AnimateDread 2026-08-11 18:14:26 -04:00
0848c8a16c fix: live inference queried the 1-tick forming bar - a window training never built AnimateDread 2026-08-11 18:10:23 -04:00
ea9d86b3ee fix: dense backprop read the weight matrix transposed - on every backend AnimateDread 2026-08-11 18:06:09 -04:00
a9677bb538 perf: the CPU DLL's dot products were never vectorized - /fp:fast AnimateDread 2026-08-11 17:20:12 -04:00
c5acc5a7a8 perf: pass 1 forward-passed ~40% of bars that a later pass redid anyway AnimateDread 2026-08-11 16:50:14 -04:00
e2c959331f perf: the excursion head cost 3.6x era time - cut its dispatches ~250x AnimateDread 2026-08-11 16:38:29 -04:00
345a672500 fix: purge every EA object namespace on init and after deinit teardown AnimateDread 2026-08-11 16:19:26 -04:00
4cfbb82634 feat: race the excursion head against a trailing-quantile incumbent AnimateDread 2026-08-11 15:57:11 -04:00
06d4785e39 fix: the excursion gate would have passed Stage 2 on an artifact I made AnimateDread 2026-08-11 15:49:32 -04:00
25aca8367c fix: the excursion head was scored against a cap I gave it AnimateDread 2026-08-11 15:21:46 -04:00
f6150ee35b fix: cache only feature SUCCESSES - the cold-indicator poison came back through the guards ba13eef did not cover AnimateDread 2026-08-11 11:30:02 -04:00
950b0fdab0 diag: name the cause when every feature window fails, and enforce the width contract AnimateDread 2026-08-11 10:31:56 -04:00
2d28f6542b feat: excursion-size head (Stage 1, measurement only) AnimateDread 2026-08-11 07:40:01 -04:00
0c8b4dc30d fix: the deploy gate graded the un-thresholded model AnimateDread 2026-08-10 19:34:32 -04:00
2189316c35 fix: the operating point was fitted on bars the net had memorized AnimateDread 2026-08-10 15:58:18 -04:00
d919a4aea2 feat: 10-bar decluster window + alternation on every signal consumer AnimateDread 2026-08-10 14:26:12 -04:00
983a6a3de1 fix: the operating-point fit maximised precision, so a no-skill model traded everything AnimateDread 2026-08-10 14:05:18 -04:00
c474ab7ad3 ui: fold the break-even back onto one panel line AnimateDread 2026-08-10 13:51:53 -04:00
ccfbc62561 fix: the recall gate was unsatisfiable and the LR decay was a spiral AnimateDread 2026-08-10 13:28:58 -04:00
ece2154102 fix: flush the in-flight era on shutdown; sweep orphaned chart objects on attach AnimateDread 2026-08-10 13:08:40 -04:00
320f13253f feat: first-passage ladder + expectancy scan - price every geometry, not just the chosen one AnimateDread 2026-08-10 12:59:18 -04:00
f656914d62 fix: the panel presented a chance win rate as skill AnimateDread 2026-08-10 12:05:49 -04:00
ba13eefecc fix: a resumed model cached a cold ATR as permanent, so it never trained AnimateDread 2026-08-10 11:41:31 -04:00
464a0fe19d diag: an era that discards itself now says so instead of scanning forever AnimateDread 2026-08-10 11:29:33 -04:00
1a5157befc fix: training could only advance one 120ms chunk per bar AnimateDread 2026-08-10 10:06:49 -04:00
3855d4666a diag: the heartbeat could be outrun by the condition it watched for AnimateDread 2026-08-10 07:44:03 -04:00
b461844767 fix: prebuild and era sized different windows; diag: Train() names its branch AnimateDread 2026-08-10 07:32:08 -04:00
783fd9e7a6 fix: the panel showed "100%" for the whole of pass 1 AnimateDread 2026-08-10 07:22:29 -04:00
694b75686e diag: slow eras must explain themselves - heartbeat + era time split + pass-1 paint AnimateDread 2026-08-10 00:10:45 -04:00
0c85c54a5b fix: a restart no longer loses the measured geometry or the training window AnimateDread 2026-08-09 22:40:43 -04:00
199726f651 fix: a one-sided era can no longer become the best checkpoint AnimateDread 2026-08-09 21:19:44 -04:00
9a7c37f334 fix: live trades now use the geometry the gate certifies; perf: BN kernels AnimateDread 2026-08-09 17:51:40 -04:00
6f920fe378 perf: batch norm stops re-reading buffers it already has AnimateDread 2026-08-09 17:16:08 -04:00
899e0c66ca perf: mini-batch apply becomes a kernel - 8 weight-matrix transfers per batch become 0 AnimateDread 2026-08-09 17:09:52 -04:00
5cef0947f4 fix: the deploy gate was benchmarking a win rate against a label frequency AnimateDread 2026-08-09 16:00:51 -04:00
ce5265488e fix: both-won bars were labelled "do not trade" - resolve by first touch AnimateDread 2026-08-09 15:35:42 -04:00
19dfb91108 feat: fitted directional confidence threshold - selectivity gets a mechanism AnimateDread 2026-08-09 15:04:37 -04:00
217b9bc9bf feat: remove Min_Risk_Reward_Ratio - a guess was overriding a measurement AnimateDread 2026-08-09 14:51:59 -04:00
371f8aaecd fix: the Adam second moment was never Adam - all four tiers AnimateDread 2026-08-09 14:02:35 -04:00
3c8d67254b chore: update binary files for WarriorCPU and WarriorDML components AnimateDread 2026-08-09 12:43:49 -04:00
0c01dc279b feat: mini-batch gradient accumulation (F4), front-end-aware capacity budget (F6), split Wyckoff categoricals (N1) AnimateDread 2026-08-09 11:48:03 -04:00
274630f802 fix: training-stability audit fixes F1/F2/F3/F5 - unbiased shuffle, real plateau escapes, fresh optimizer state on restore, pure OOS metric AnimateDread 2026-08-09 10:54:09 -04:00
ee48381cbd fix: NMS gates the TRADE, not just the arrow - one arrow is now one trade AnimateDread 2026-08-09 10:22:31 -04:00
c690a73901 docs: remove outdated note from AutoTuneIndicators comment AnimateDread 2026-08-09 09:51:27 -04:00
fcb69d8efb research: test whether across-time structure exists at all AnimateDread 2026-08-08 20:42:49 -04:00
922484e8d9 feat: expose the AD/Wyckoff parameters; default the indicator tuner off AnimateDread 2026-08-08 20:37:21 -04:00
1df305431d feat: gate deployment on the null of the MAXIMUM, not the per-era null AnimateDread 2026-08-08 18:01:04 -04:00
33f106d99d fix: the indicator re-init leaked a terminal handle per candidate AnimateDread 2026-08-08 15:42:03 -04:00
bfc1da9de1 fix: the sequence models were reading the window backwards AnimateDread 2026-08-07 18:28:44 -04:00
da54639996 feat: expectancy stop - halt when the measured result says the strategy loses AnimateDread 2026-08-07 14:20:00 -04:00
b3b7e7bceb fix: excursion window must not depend on the barrier it sizes AnimateDread 2026-08-07 13:57:23 -04:00
32ffeb99f3 fix: normalise the asymmetry target - the raw one is confounded by volatility AnimateDread 2026-08-07 13:04:13 -04:00
a7701f032b feat: derive the ATR multiples from measured excursions - no hardcoded geometry AnimateDread 2026-08-07 12:06:25 -04:00
2c78f3b90d diag: is "optimal SL/TP" learnable? Score the features against excursions AnimateDread 2026-08-07 10:22:41 -04:00
3482b6c238 feat: entry/SL/TP stop being inputs - the barrier geometry is measured AnimateDread 2026-08-07 09:39:30 -04:00
9e1c72aacc fix: make the indicator tuner actually measure, and gate what it installs AnimateDread 2026-08-07 09:31:06 -04:00
e5ceed6466 fix: MI diagnostics never ran on a resumed model - the stated intent was never achieved AnimateDread 2026-08-07 09:08:01 -04:00
0d5892357b diag: report the indicator tuner's candidate spread - "no improvement" is ambiguous AnimateDread 2026-08-06 21:59:10 -04:00
cccf94f9ca fix: correct the lag profile across lags too - it contradicted itself AnimateDread 2026-08-06 16:28:57 -04:00
04ee2e113a fix: gate the barrier-geometry winner on a family-wise null, not its own AnimateDread 2026-08-06 14:04:16 -04:00
3271f1ea93 diag: MI feature-lag profile - close the blind spot in every MI verdict so far AnimateDread 2026-08-06 13:51:39 -04:00
8c5ea639ee feat: extend ADWyckoffEventStream with new range-lifecycle parameters and update related features AnimateDread 2026-08-02 17:08:48 -04:00
8ccbddb051 Add new research scripts for trading strategy analysis AnimateDread 2026-08-02 12:25:20 -04:00
f1b7dcf7f3 fix: correct MI sample alignment and improve BN weight diagnostic report AnimateDread 2026-08-02 08:12:47 -04:00
5f647ba5db fix: improve error messages and suppress false sharing-violation logs AnimateDread 2026-08-02 01:09:18 -04:00
3459db887b infra: bid/ask M1 bars and an order engine that makes the fill bug unavailable AnimateDread 2026-08-02 00:08:30 -04:00
82cdbb40df research: Wyckoff context REPLICATES on a second, independent trigger AnimateDread 2026-08-01 23:57:30 -04:00
c998d655ee research: Wyckoff CONTEXT has real predictive content - 8/8 positive slopes AnimateDread 2026-08-01 23:56:05 -04:00
9a4ae635e0 research: the two canonical Wyckoff trades tested whole - both negative AnimateDread 2026-08-01 23:34:35 -04:00
79f108ab84 research: RETRACTION - the retail-fade edge was a fill-timing artifact AnimateDread 2026-08-01 23:17:55 -04:00
821f16df8b research: model-selected retail fades hold up out of sample (after killing a big leak) AnimateDread 2026-08-01 23:15:00 -04:00
c9b489e5d8 research: the retail fade DOES clear cost on EURUSD - correcting the earlier verdict AnimateDread 2026-08-01 22:33:11 -04:00
c645c98f31 chore: keep the reference PDFs out of the repository AnimateDread 2026-08-01 22:18:18 -04:00
c2dd9eb6aa research: retail setups ARE anti-predictive - and the edge dies with the cost AnimateDread 2026-08-01 22:18:01 -04:00