This project uses a neural network model to predict cognitive fatigue in simulated stock trading behavior based on decision-making patterns and reaction times. It combines time-series processing, deep learning, and trading simulation to explore how fatigue may manifest in behavioral data.
The goal was to build a real-time fatigue detection system that analyzes:
- Reaction time consistency
- Trading decisions over time (BUY, SELL, HOLD)
- Sequence patterns leading to mental fatigue
- FatigueLSTM model: Custom LSTM-based neural network built using PyTorch
- Real-time fatigue prediction using rolling windows
- Simulated trader environment for generating or testing behavioral data
- Behavioral encoding of decisions for model compatibility
- Configurable sequence length and model hyperparameters
| File | Description |
|---|---|
fatigue_model.pt |
Pretrained PyTorch model weights |
predict_live.py |
Real-time prediction using the trained model |
prepare_sequences.py |
Sequence preprocessing for time-series input |
train_model.py |
Training script for LSTM using session data |
trader_simulator.py |
Simulates trading sessions with reaction times and choices |
requirements.txt |
Python dependencies |
- Python
- PyTorch
- pandas, NumPy
- Matplotlib (optional for plotting, not shown in repo yet)