jgrusewski 6414f57e34 docs: eval cleanup spec + plan — remove dead 5-action path, fix train/eval mismatch ...

3 fixes: (1) Delete GpuActionSelector + branching_action_select (zero callers,
hardcoded 5 actions vs production 9), (2) Add hold enforcement to backtest
env_step kernel (same Layer 2 as training), (3) Dynamic trade insufficiency
threshold = max(window_bars / (min_hold * 20), 5).

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

2026-03-26 00:43:58 +01:00

..

infra

feat(infra): GPU training launcher with local/cloud routing

2026-02-23 11:19:58 +01:00

plans

feat(dqn): enable Branching DQN with 45 factored actions (5×3×3)

2026-03-11 22:00:13 +01:00

superpowers

docs: eval cleanup spec + plan — remove dead 5-action path, fix train/eval mismatch

2026-03-26 00:43:58 +01:00