d9b7e0553120f5655070e66fae7fefea0de9ef23
Optimizations in _run_episode: - Pre-allocated ActionKind constants (avoid repeated attribute lookups) - Removed unnecessary max_pos_qty tracking (unused in scoring) - Simplified action kind checks (single comparison chain) - Reduced frozen dataclass allocations per step Result: same behavioral output, cleaner code path. Episode time: ~19ms/step sequential, ~13ms/step parallel (unchanged — bottleneck is MCTS planner + CWM, not Python orchestration).
Description
Sentiment analysis engine with ONNX FinBERT + LoRA adapters
Languages
Python
99.9%
Dockerfile
0.1%