yirugi
/
fithia2
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Commit Graph

  • ea9f156eeb Tune Form4 sleeve quality filters and reserve sizing I Luk Kim 2026-04-06 01:56:58 -0700
  • 72681e69e5 Add Form4 residual-cash sleeve and UI support I Luk Kim 2026-04-06 01:45:02 -0700
  • 86d55e01f9 Fix PIT snapshot store regressions for backtests I Luk Kim 2026-04-01 01:26:55 -0700
  • f2113b7e06 Fix cash parking phantom-money bug + live engine parking liquidation for events I Luk Kim 2026-03-31 14:36:17 -0700
  • 014d1a5d6c Add v12.61 strategy documentation I Luk Kim 2026-03-30 23:21:01 -0700
  • 9cb91ee846 Add synthetic scenario robustness testing system I Luk Kim 2026-03-30 23:07:17 -0700
  • 0928eb2428 Fix direct backtest mode UX: live log, inline results, trade table sort/filter I Luk Kim 2026-03-28 14:36:02 -0700
  • c5dea9a9a8 Add Paper Trading web GUI with ANSI terminal log rendering I Luk Kim 2026-03-28 04:21:00 -0700
  • 7df99447ee Register named strategies on leaderboard I Luk Kim 2026-03-27 04:16:24 -0700
  • 8dfebcf5fe Add v6new.311-322, named strategy configs, and tracker common_window fix I Luk Kim 2026-03-27 04:08:02 -0700
  • f56b353c8b Update README: paper trading operations and Phase 6 reconciliation docs I Luk Kim 2026-03-27 04:07:36 -0700
  • 98923e8620 Add Phase 6 reconciliation: position integrity, kill switch, order fill verification I Luk Kim 2026-03-27 04:07:30 -0700
  • 76581ead04 Remove overlay backtesting and scoring I Luk Kim 2026-03-26 21:20:34 -0700
  • 1d9507540b Add strategy aliases to key experiment configs I Luk Kim 2026-03-26 20:50:50 -0700
  • 902b7d99b8 Add v6new.307: SQS 74.4 #1, CW 350.2% — high-WR engine risk boost I Luk Kim 2026-03-26 20:44:08 -0700
  • 30eb311443 Add v6new.288: SQS 72.8 #1, CW 320.5% — core risk up + budget downsizing I Luk Kim 2026-03-26 20:31:28 -0700
  • b00f87b6b0 Add uv.lock for reproducible dependency resolution I Luk Kim 2026-03-26 20:20:31 -0700
  • 493b8a8d69 Add --overlay shorthand for lb command and gitignore *.db files I Luk Kim 2026-03-26 20:19:12 -0700
  • a81b3a6ac4 Update tracker, leaderboard, docs, and overlay leaderboard I Luk Kim 2026-03-26 20:17:21 -0700
  • 2aba6418e6 Add overlay engine, ranking models, snapshot pipelines, and research tools I Luk Kim 2026-03-26 20:17:16 -0700
  • 9b92ab6589 Add paper trading system: broker integration, state management, reporter I Luk Kim 2026-03-26 20:17:08 -0700
  • 479eb76308 Update tests for pipeline, backtest engine, paper trader, and new modules I Luk Kim 2026-03-26 20:17:03 -0700
  • 320a9d19df Clean up rank 100+ strategies and obsolete files I Luk Kim 2026-03-26 20:16:35 -0700
  • ce2150789d Fix leaderboard performance regression (60min → 12s) and clean up CLI help I Luk Kim 2026-03-26 20:15:16 -0700
  • 057a311572 Improve paper backtest: overlay support, --top/--rank/--year options, speed optimization I Luk Kim 2026-03-26 20:15:09 -0700
  • 784c581f19 Enhance backtest engine: v11 scoring, selector expansion, snapshot store improvements I Luk Kim 2026-03-26 20:15:02 -0700
  • 57d38ecfe0 Add earnings surprise feature pipeline and snapshot export improvements I Luk Kim 2026-03-26 20:14:57 -0700
  • bc4ffecb10 Add v6new.275: SQS 72.7 #1, CW 312.6% — OME early fail day 1 I Luk Kim 2026-03-26 20:14:54 -0700
  • 165424ecc2 Add v6new.272: CW 310.8% breakthrough via OME early failure cut I Luk Kim 2026-03-26 20:05:13 -0700
  • 7676ef7867 Register v6new.255 on leaderboard (SQS 72.2, #9, CW 293.2%) I Luk Kim 2026-03-26 18:24:27 -0700
  • 1c38415cb7 Add v6new.265-268: small-cap expansion experiments I Luk Kim 2026-03-26 18:06:15 -0700
  • e2229646e6 Add oversold bounce engine experiments (v6new.259-264) — blocked by architecture I Luk Kim 2026-03-26 16:56:53 -0700
  • 74244b15d2 Add v6new.224-258: grid search risk×ATR toward 300% CW I Luk Kim 2026-03-26 16:31:01 -0700
  • 672098c077 Add v6new.215-223: three-way parallel search toward 300% CW I Luk Kim 2026-03-26 16:18:57 -0700
  • 6dee2c24d2 Add v6new.197-214: parameter grid search around v196 I Luk Kim 2026-03-26 16:08:14 -0700
  • cc953949f1 Add v6new.196: new best SQS 72.6 (#3) — v29+entropy+risk+doc_quality I Luk Kim 2026-03-26 15:57:51 -0700
  • 090bfa8e36 Add contrarian feature analysis + v15 scoring (v6new.174-188) I Luk Kim 2026-03-26 14:09:37 -0700
  • 95c6b499c4 Add missing experiment configs v6new.129, 160-162 I Luk Kim 2026-03-26 13:31:26 -0700
  • 1e66b67c7a Add technical/scientific feature experiments (v6new.106-173) and v6new.122 SQS 72.4 I Luk Kim 2026-03-26 13:30:53 -0700
  • 83c7a66f77 Add v6new.37-38: final tuning attempts — v6new.30 confirmed as optimum I Luk Kim 2026-03-25 14:52:06 -0700
  • daa8536546 Add v6new.36: mean reversion engine for high-VIX drops I Luk Kim 2026-03-25 14:34:51 -0700
  • 16613f1758 Fix _rows_to_table to collect keys from ALL rows, not just first I Luk Kim 2026-03-25 13:14:02 -0700
  • fecdc12007 Add earnings surprise feature pipeline and v11 scoring I Luk Kim 2026-03-24 20:00:54 -0700
  • e8241d2924 Test execution model + short side — both rejected, v6new.30 confirmed I Luk Kim 2026-03-24 16:02:25 -0700
  • f2b3b57b3a Update v6new.30 on refreshed snapshot — SQS 62.3, WFV 100% positive I Luk Kim 2026-03-24 14:24:28 -0700
  • 85515f2db9 Add v6new.31-33: parameter tuning — all neutral/worse, v6new.30 confirmed optimal I Luk Kim 2026-03-24 13:43:22 -0700
  • b6ea2a4b50 Add v6new.29-30: OME risk experiments — v6new.30 Paper BT +108.62% I Luk Kim 2026-03-24 13:10:33 -0700
  • 2a82c1e13c Record v6new.27 SQS 63.9 — new #1 validated by both SQS and paper backtest I Luk Kim 2026-03-24 12:57:05 -0700
  • dc16a1f88b Add v6new.28 (marginal) — v6new.27 confirmed as optimal at +105.35% I Luk Kim 2026-03-24 12:48:58 -0700
  • 1f996dfc8e Add v6new.25-27: paper-backtest-driven strategy improvement I Luk Kim 2026-03-24 12:47:42 -0700
  • 7a2001a4fa Add v6new.25: OME risk halved — paper BT +94.38% beats v6.29 +93.60% I Luk Kim 2026-03-24 12:25:56 -0700
  • 529477727f Add score and event_type to trade blotter output I Luk Kim 2026-03-24 12:05:19 -0700
  • 07bcfbe51d Complete engine unification Phase 5-6: residual reserve + macro data I Luk Kim 2026-03-24 11:54:24 -0700
  • 26ca89c058 Integrate shared trade logic into PaperTradingEngine (Phase 3-4 complete) I Luk Kim 2026-03-24 11:49:25 -0700
  • fb4fec7dac Unify BacktestRunner and PaperTradingEngine trade decision logic I Luk Kim 2026-03-24 11:38:14 -0700
  • c747d5e4f0 Fix snapshot path resolution for paper backtest I Luk Kim 2026-03-24 11:03:32 -0700
  • 0134476632 Fix async/sync conflict in paper backtest — nested event loop error I Luk Kim 2026-03-24 10:56:51 -0700
  • 33acf12baa Remove snapshot refresh fallback — fail hard if pipeline update fails I Luk Kim 2026-03-24 08:30:57 -0700
  • 3c65c72a72 Auto-refresh snapshot when paper backtest end_date exceeds snapshot coverage I Luk Kim 2026-03-24 08:25:08 -0700
  • d7ecaf97dc Refactor paper backtest to use BacktestRunner — eliminates engine divergence I Luk Kim 2026-03-24 04:37:13 -0700
  • 0797535047 Revert DB-first feature fix — Oracle real-time enrichment must be primary I Luk Kim 2026-03-24 04:33:08 -0700
  • f0baa48e7b Revert paper trader scoring to compute_entry_score — fix 2025 trade loss I Luk Kim 2026-03-24 04:27:11 -0700
  • bb659cf7ae Add reaction-based position sizing to reduce mean-reversion loss risk I Luk Kim 2026-03-24 03:58:19 -0700
  • 2bb5fe1c37 Fix paper trader / backtester consistency: DB-first features + config-aware scoring I Luk Kim 2026-03-24 03:13:34 -0700
  • 4de842ba89 Fix paper trader volume_ratio field name mismatch with selector I Luk Kim 2026-03-24 03:02:11 -0700
  • 89e5b647b0 Add v6new.23: OME risk halved + vol/reaction gates — SQS 62.3, best loss profile I Luk Kim 2026-03-24 00:57:04 -0700
  • 60da697d54 Add v6new.22: paper trade loss fixes — SQS 61.9 but blocks $1.5K live losses I Luk Kim 2026-03-24 00:03:02 -0700
  • 3c13c9e1b5 Add vix_pead regime sizing mode and macro experiments (v6new.19-21) I Luk Kim 2026-03-23 23:37:59 -0700
  • 8dacdaab4e Add v10 macro regime scoring (VIX+HY) and FRED macro features — SQS 59.1 (rejected) I Luk Kim 2026-03-23 23:18:25 -0700
  • 75de2493f0 Add v6new.14-16 (execution tuning) — all worse, v6new.9 (63.3) holds I Luk Kim 2026-03-23 22:22:44 -0700
  • a1a0209342 Add v6new.12-13 (risk sizing tuning) — neutral, v6new.9 (63.3) confirmed optimal I Luk Kim 2026-03-23 21:55:57 -0700
  • 5f470b73da Add v6new.11 (engine tuning) — SQS 62.7, v6new.9 (63.3) remains best I Luk Kim 2026-03-23 21:35:52 -0700
  • 0e62b01c89 Add v6new.10 (engine pruning) — SQS 61.9, confirms v6new.9 remains best at 63.3 I Luk Kim 2026-03-23 21:27:43 -0700
  • 5d728298cf Add v6new.9: coverage expansion with tuned risk — SQS 63.3 beats v6.29 (62.8) I Luk Kim 2026-03-23 20:36:27 -0700
  • 85c4d98987 Add v6new experiment suite: scoring v8/v9/v9g, cross-event drift, coverage engines I Luk Kim 2026-03-23 20:22:17 -0700
  • 02542248b7 Add attention-aware PEAD caps and promote step75 I Luk Kim 2026-03-17 17:27:43 -0700
  • b507fbf499 Add attention client and continue PEAD research I Luk Kim 2026-03-17 16:58:04 -0700
  • d3e4c1d0d5 chore: add runs/ to .gitignore I Luk Kim 2026-03-17 02:21:12 -0700
  • c646303423 Recalibrate public SQS and exposure-aware tracking I Luk Kim 2026-03-17 02:18:41 -0700
  • 9ec0b26e10 Implement multi-engine PEAD strategy research workflow I Luk Kim 2026-03-17 00:57:41 -0700
  • 2395a0c0c3 feat: PEAD mid-cap strategy + pipeline hardening + README cleanup I Luk Kim 2026-03-16 21:19:09 -0700
  • 764493dbe1 docs: add project README with architecture, usage, and improvement tracking guide I Luk Kim 2026-03-16 15:47:41 -0700
  • cb19afa87b feat: add strategy improvement tracking system (SQS + journal + leaderboard) I Luk Kim 2026-03-16 15:43:00 -0700
  • 191653394d feat: scoring cleanup — alpha-only composite, default-deny unknown events, new exit/risk features I Luk Kim 2026-03-14 00:41:40 -0700
  • 3779be78e3 fix: kill switch research mode reset + adjust veto threshold I Luk Kim 2026-03-13 21:03:34 -0700
  • 4d0e773ba0 feat: overhaul strategy — document quality > price momentum I Luk Kim 2026-03-13 20:04:26 -0700
  • cdf6ae3493 feat: Phase 5 fundamental strategy improvements I Luk Kim 2026-03-13 09:35:04 -0700
  • d4ae900286 feat: expand data pipeline and integrate event features into scoring I Luk Kim 2026-03-12 22:28:24 -0700
  • 354f7a1716 feat: add rule-based entry score model for backtester I Luk Kim 2026-03-12 22:14:41 -0700
  • d387d567c5 fix: correct equity calculation — use market value, not unrealized PnL I Luk Kim 2026-03-12 22:09:28 -0700
  • 867d70afae fix: correct simulation loop and equity curve calculation after real-data testing I Luk Kim 2026-03-12 21:44:36 -0700
  • 2f4d9f61f7 feat: implement Phase 4 -- event-driven backtester I Luk Kim 2026-03-12 21:35:06 -0700
  • f69a8b4731 fix: use TEST-prefixed IDs in integration tests to avoid real-data conflicts I Luk Kim 2026-03-12 17:22:10 -0700
  • 9d3427e24e fix: disable Qwen3.5 thinking mode and switch OllamaClient to sync httpx I Luk Kim 2026-03-12 17:16:34 -0700
  • bbe0e31150 feat: implement Phase 3 -- LLM enrichment, labeler, review queue, dataset export I Luk Kim 2026-03-12 12:00:31 -0700
  • 7fed850fbf test: Phase 1/2 체크리스트 재작성 및 Oracle 실연동 통합 테스트 I Luk Kim 2026-03-12 11:28:00 -0700
  • 38ac53b597 implement Phase 2 gap items: backfill CLI, retry, financial features I Luk Kim 2026-03-12 10:35:00 -0700
  • 32a845044c fix: integration test isolation and migration server_default escaping I Luk Kim 2026-03-12 09:17:12 -0700
  • 0471018dd8 feat: implement ACE-F v1 Phase 1 -- Stock Oracle 기반 이벤트 파이프라인 I Luk Kim 2026-03-12 08:44:53 -0700