Reinforcement Learning (RL) Commodity Trading Advisor (CTA) Strategy based on TQSDK
- Tick/Bar Level reinforcement learning environment
- Target position action schema
- Subscribing and presisting data from tqsdk to MongoDB sharding
- Ray RLlib supported distributed training
- Plugable factors and alphas as state
- Temporal Fusion Transformer as predictor model