jgrusewski 727ed0d43b refactor: remove use_qr_dqn — IQN always enabled via iqn_lambda ...

use_qr_dqn was a hacky toggle that gated the IQN dual-head behind
a num_atoms threshold. IQN is now always enabled when iqn_lambda > 0
(default 0.25). C51 remains the main loss; IQN is the auxiliary head
for CVaR risk quantification.

Removed use_qr_dqn from: DQNParams, DQNHyperparameters, fused_training,
hyperopt adapter, all tests, all examples.

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

2026-03-23 09:46:41 +01:00

..

baseline_common

fix(ml): wire spread_cost_bps into RL training — commission + spread = total tx cost

2026-03-01 22:10:43 +01:00

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

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

evaluate_baseline.rs

refactor: remove use_qr_dqn — IQN always enabled via iqn_lambda

2026-03-23 09:46:41 +01:00

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: Phase 3 reward tuning — fix dynamics+architecture, search 7 reward dims

2026-03-22 21:45:00 +01:00

hyperopt_baseline_supervised.rs

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

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

train_baseline_rl.rs

refactor: remove use_qr_dqn — IQN always enabled via iqn_lambda

2026-03-23 09:46:41 +01:00

train_baseline_supervised.rs

feat: TOML training profile system — config-driven hyperparameters

2026-03-21 23:35:41 +01:00