foxhunt

Files

jgrusewski bacaf2765a feat: leverage-based position cap (replaces hardcoded max_position)

max_position_absolute is now computed from max_leverage:
  max_position = floor(capital * max_leverage / (price * multiplier))

- Added max_leverage config field (default: 5.0)
- compute_max_position() derives position from leverage + median price
- Hyperopt risk_intensity scales max_leverage (not position directly)
- Updated all TOML configs: dqn-production, dqn-smoketest, dqn-localdev
- Hyperopt search space: max_leverage = [2.0, 10.0] (replaces [1.0, 4.0] contracts)
- GpuBacktestConfig wired with max_leverage for consistent eval

With $35K capital, ES at $5K, multiplier=50:
  5× leverage → floor(35000*5/250000) = 0.7 → 1 contract (safe)
  Old default 2.0 contracts → 14× leverage (dangerous)

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

2026-04-07 16:24:20 +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

feat: DQN gems — spectral decoupling, manifold mixup, regime replay, family hyperopt

2026-03-31 00:39:23 +02:00