8a746fceb2
regime_velocity (EMA of ADX+CUSUM deltas), regime_disagreement (|norm_ADX - norm_CUSUM|), regime_transition_ema, regime_stability (1 - sigmoid(5*velocity)). Read from states buffer sample 0. isv_signal_update + isv_forward kernels updated. w_isv_fc1 grows 16*8→16*12. Pinned buffers resized automatically via ISV_DIM. Introduced ISV_EMB_DIM=8 to decouple embedding output from signal input dimension — FC2/gate/gamma stay at 8. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
ml
10-model ML ensemble for the Foxhunt HFT system, built on Candle v0.9.1.
Models
- DQN (Rainbow) — deep Q-network with prioritized replay, dueling heads, noisy nets
- PPO — proximal policy optimization with GAE, LSTM policies, clip-higher
- TFT — temporal fusion transformer for multi-horizon forecasting
- Mamba2 — state space model for sequence prediction
- Liquid Networks — biologically inspired networks for non-stationary data
- TLOB — transformer-based limit order book analysis
- KAN — Kolmogorov-Arnold networks
- xLSTM — extended LSTM architecture
- TGGN — temporal graph neural network
- Diffusion — diffusion-based generative model
Key Modules
ensemble— model ensemble coordination and confidence aggregationhyperopt— PSO-based hyperparameter optimization with per-model adapterstrainers— unified training loops (DQN, PPO, supervised)inference—InferenceAdaptertrait for predictioncheckpoint— model checkpointing and restorationevaluation— walk-forward evaluation pipeline
Usage
use ml::dqn::DQN;
use ml::ppo::PpoTrainer;