jgrusewski 108bb63fce feat: cost-driven hold timing — replace min_hold_bars with learned cost signals ...

Removed: enforce_hold(), min_hold_bars from config/kernels/backtest.
Added: holding_cost_rate (inventory penalty), churn_threshold_bars +
churn_penalty_scale (graduated flip penalty) to reward in
experience_env_step and backtest kernels.

The model learns optimal hold timing from cost signals:
- Per-trade tx cost prevents churning (existing)
- Inventory penalty makes large positions expensive to hold
- Churn penalty graduates cost for rapid flips
- Temporal attention learns when holding cost > expected profit

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

2026-04-18 01:04:11 +02:00

..

baseline_common

…

cuda_test.rs

fix: wire real GPU backprop + SPSA gradients, fix checkpoint loading, eliminate candle from examples

2026-03-20 15:41:42 +01:00

download_baseline.rs

fix: add retry with backoff to Databento download_quarter

2026-03-31 22:57:43 +02:00

evaluate_baseline.rs

fix: evaluate_baseline compile — make q_provider Optional, remove dead evaluate_dqn path

2026-04-13 21:05:01 +02:00

evaluate_supervised.rs

fix: wire real GPU backprop + SPSA gradients, fix checkpoint loading, eliminate candle from examples

2026-03-20 15:41:42 +01:00

hyperopt_baseline_rl.rs

feat: cost-driven hold timing — replace min_hold_bars with learned cost signals

2026-04-18 01:04:11 +02:00

hyperopt_baseline_supervised.rs

fix(clippy): resolve all clippy warnings across entire workspace (--all-targets)

2026-03-13 10:18:35 +01:00

precompute_features.rs

fix: fxcache version display + export FXCACHE_VERSION + cleanup dead code

2026-04-17 02:12:33 +02:00

train_baseline_rl.rs

feat: cost-driven hold timing — replace min_hold_bars with learned cost signals

2026-04-18 01:04:11 +02:00

train_baseline_supervised.rs

refactor: remove max_training_steps_per_epoch — always train full dataset

2026-04-02 14:41:16 +02:00