jgrusewski a400288ae0 feat: comprehensive TOML profiles + config consistency test
Task 7: Update all three DQN TOML profiles with every configurable parameter.
- dqn-smoketest: add [distributional], [advanced], [exploration] (noisy_sigma,
  entropy_coefficient, count_bonus), [risk] (max_position, loss_aversion),
  fix learning_rate 0.0003 -> 0.00003, add reward_scale + gamma
- dqn-production: add reward_scale, exploration params (noisy/entropy/count_bonus),
  advanced params (n_steps/tau/c51_warmup/her/iqn_lambda/spectral_norm),
  remove hardcoded v_min/v_max (now computed), remove duplicate tau from [training]
- dqn-hyperopt: add search space bounds for spectral_norm_sigma_max,
  c51_warmup_epochs, her_ratio

Profile system additions (training_profile.rs):
- TrainingSection: add reward_scale (recomputes v_min/v_max on apply)
- ExplorationSection: add noisy_sigma_init, entropy_coefficient,
  count_bonus_coefficient, q_gap_threshold
- AdvancedSection: add n_steps, tau, c51_warmup_epochs, her_ratio,
  iqn_lambda, spectral_norm_sigma_max, gradient_clip_norm
- RiskSection: add max_position (alias for max_position_absolute), loss_aversion
- RewardSection: add reward_scale
- SearchSpaceSection: add spectral_norm_sigma_max, c51_warmup_epochs, her_ratio
- apply_to: gamma change now recomputes v_min/v_max automatically

Task 8: Config consistency integration tests (5 tests):
- test_config_consistency_across_structs: v_range computed not hardcoded,
  fill simulation bounds, q_clip symmetry
- test_toml_profile_applies_all_fields: smoketest profile applies every
  new section field correctly
- test_production_profile_applies_all_sections: production profile end-to-end
- test_hyperopt_search_space_has_new_bounds: new search space fields parse
- test_reward_scale_recomputes_v_range: gamma override triggers v_range recomputation

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-25 10:27:45 +01:00

Foxhunt

Production HFT trading system in Rust.

Architecture

The workspace contains 32 crates organized as follows:

Core Libraries (16)

Crate Purpose
trading_engine Order processing, FIX 4.4, IB TWS, SIMD, RDTSC timing
risk VaR, Kelly, circuit breakers, kill switches, compliance
risk-data Risk data types and shared structures
trading-data Trading data types
ml DQN Rainbow, PPO, TFT, Mamba2, ensemble inference
ml-data ML data types and feature definitions
data Market data ingestion and storage
backtesting Replay engine, strategy tester
adaptive-strategy Ensemble execution, microstructure analysis
common Shared types, resilience, error handling
storage S3 and local model storage
model_loader Model serialization and loading
market-data Market data feed handlers
database PostgreSQL access layer (SQLx)
config Configuration management
tli CLI commands and tooling

Services (8)

Service Purpose
backtesting_service gRPC backtesting service
broker_gateway_service FIX routing, broker connectivity
trading_service Core trading operations
ml_training_service Model training orchestration
data_acquisition_service Market data acquisition
trading_agent_service Autonomous trading agents
api_gateway gRPC API gateway with auth
web-gateway Axum REST + WebSocket gateway

Frontend

web-dashboard/ -- React 19 + TypeScript + Vite + TradingView charts.

Building

# Check compilation (no PostgreSQL required)
SQLX_OFFLINE=true cargo check --workspace

# Run tests for a specific crate
SQLX_OFFLINE=true cargo test -p <crate> --lib

# Clippy
SQLX_OFFLINE=true cargo clippy --workspace

ML Models

Four production model architectures on Candle v0.9.1 with CUDA:

  • DQN Rainbow -- Deep Q-Network with prioritized replay, dueling heads, noisy nets
  • PPO -- Proximal Policy Optimization with GAE and LSTM policies
  • TFT -- Temporal Fusion Transformer for multi-horizon forecasting
  • Mamba2 -- State space model for sequence prediction

Each model has a standalone trainer and a UnifiedTrainable adapter for the hyperopt pipeline.

Infrastructure

  • Git: Gitea at git.fxhnt.ai (Tailscale-only), Scaleway DEV1-S
  • Observability: OpenTelemetry OTLP (env OTEL_EXPORTER_OTLP_ENDPOINT)
  • Database: PostgreSQL with SQLx offline mode for CI

License

Proprietary. All rights reserved.

Description
No description provided
Readme 849 MiB
Languages
Rust 88.2%
Cuda 7.7%
Python 1.3%
Shell 1.1%
PLpgSQL 0.8%
Other 0.8%