Skip to content

fix(quant): RL bridge onehot/features/clamp + drop dead tickers - #234

Draft
JoTalbot wants to merge 6 commits into
mainfrom
agent/20260814-paper-fix
Draft

fix(quant): RL bridge onehot/features/clamp + drop dead tickers#234
JoTalbot wants to merge 6 commits into
mainfrom
agent/20260814-paper-fix

Conversation

@JoTalbot

Copy link
Copy Markdown
Owner

Восстановленный orphan-PR: работа без PR с 2026-08-14. +6/−64 vs main.

Этот PR создан ИИ-агентом (OpenHands) от имени оператора при наведении порядка в ветках.

Deployed CatBoost direction model was degenerate (prob_up=0.433 for 30/35
assets, AUC 0.504, never reaches the 0.65 entry gate). Replace with a model
trained on scale-free features with a strict per-symbol walk-forward split:
AUC 0.533, hit-rate 81-83% at prob>=0.65 on two independent OOS windows,
positive net PnL with Directional v2 paper exit rules. Old model kept as
fallback; ml_predictor prefers catboost_price_dir_v2.cbm.
RLSignalBridge vs training env (kg_v8) mismatches:
1. onehot always marked BTC - now per-asset index in ASSET_ORDER (32 assets,
   alphabetical, as MultiAssetEnv); unknown asset -> None (no signal).
2. static feature 3 was vol_chg, training used vol_ratio - fixed.
3. action was not clamped to [-1,1] like in training rollout - pos could be
   -0.5 (outside [0,1]) - fixed.

Dead tickers MATIC/RNDR (renamed to POL/RENDER) removed from ML signals,
RL default universe and notebook generator template. rl_signals.json refreshed
(9 assets, honest FLAT verdicts). 61 quant/ml tests pass.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant