How vinee119/hdt works
This repository is a Streamlit Human Digital Twin for simulating behavioral tendencies from a user’s situation description, scenario, and difficulty. The main workflow detects an emotion from text, maps it to a behavioral state, combines it with contextual inputs for prediction, then presents an explanation and interpretation. The README also documents synthetic-data generation and model-training capabilities, represented here as the data and model-artifact branch rather than notebook workflows. Runtime connections are grounded in the sampled app and simulation engine; broader modeling details come from the README.
Main entry points: streamlit_app.py Streamlit Interface · simulation_engine.py Simulation Engine · models Trained Model Artifacts · data Behavioral Datasets
Components 12
User Interface
- Streamlit Interface
app/streamlit_app.py - Results and Charts
app/streamlit_app.py
Simulation Runtime
- Simulation Engine
core/simulation_engine.py - Emotion Detection
core/simulation_engine.py - Emotion Mapping
core/simulation_engine.py - Context Integration
core/simulation_engine.py - Behavior Prediction
core/simulation_engine.py - Behavior Explanation
core/simulation_engine.py - Behavior Interpretation
core/simulation_engine.py
Data and Models
- Trained Model Artifacts
models - Behavioral Datasets
data
Other components
- User
Connections 11
- User → Streamlit Interface: enters situation
README.md - Streamlit Interface → Simulation Engine: runs simulation
app/streamlit_app.py - Simulation Engine → Emotion Detection: processes text
core/simulation_engine.py - Emotion Detection → Emotion Mapping: maps emotion
core/simulation_engine.py - Emotion Mapping → Context Integration: combines context
core/simulation_engine.py - Context Integration → Behavior Prediction: predicts decision
core/simulation_engine.py - Behavior Prediction → Behavior Explanation: builds explanation
core/simulation_engine.py - Streamlit Interface → Behavior Interpretation: requests interpretation
app/streamlit_app.py - Streamlit Interface → Results and Charts: renders results
app/streamlit_app.py - Simulation Engine → Trained Model Artifacts: loads artifacts
core/simulation_engine.py - Behavioral Datasets → Trained Model Artifacts: supports training
README.md