3 Commits

Author	SHA1	Message	Date
jgrusewski	03062f2401	fix(ml): use exposure index (0-4) in DQN experience storage and tracking ... FactoredAction::to_index() returns 0-44 (exposure×9 + order×3 + urgency), but DQN Q-network has 5 outputs. Storing the factored index in Experience caused train_step() to reject valid actions (e.g. Flat→index 19 > num_actions=5). Fix: use action.exposure as usize (0-4) for: - Experience storage (training loop + validation adapter) - Count bonus tracking (diversity metrics) - Safety action counts (diversity monitor) - Debug logging (action indices) - Test assertions (< 45 → < 5) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-03-06 16:27:17 +01:00
jgrusewski	512535076f	feat(ml): wire PPO to 45 factored actions via FactoredAction ... sample_action(), act(), act_with_log_prob(), greedy_action() now return FactoredAction instead of TradingAction. Fixes the architectural disconnect where num_actions=45 output neurons were sampled through a 3-action bottleneck. TrajectoryStep.action and TrajectoryBatch.actions now use FactoredAction. Added FactoredAction::from_legacy() for backward compatibility in tests. Updated all PPO consumers: trainers/ppo.rs, hyperopt/adapters/ppo.rs, validation/ppo_adapter.rs, benchmark/ppo_benchmark.rs. 2487 tests pass, 0 clippy warnings. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-03-02 15:41:23 +01:00
jgrusewski	9c3d741a08	refactor: restructure repo — crates/, bin/, testing/ layout ... Move 17 library crates into crates/, CLI binary into bin/fxt, consolidate 10 test crates into testing/, split config crate from deployment config files. Root directory reduced from 38+ to ~17 directories. All Cargo.toml paths and build.rs proto refs updated. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-25 11:56:00 +01:00