How rayhan1705/4911-resikokredit works
This is a small Streamlit credit-risk prediction app. Users enter demographic and loan details, the app validates employment length and displays a summary, then scales the features and predicts default risk using a selected XGBoost or Random Forest model. The graph reflects the single sampled runtime file; the notebook, dataset, and serialized assets are present in the tree, but their contents and use are not established by the excerpts.
Main entry points: app.py Prediction flow · scaler.pkl Feature scaler · model_xgboost.pkl XGBoost model · model_random_forest.pkl Random Forest model
Components 9
User interface
- Input and validation
app.py - Applicant summary
app.py - Risk results
app.py
Prediction runtime
- Prediction flow
app.py
Model assets
- Feature scaler
scaler.pkl - XGBoost model
model_xgboost.pkl - Random Forest model
model_random_forest.pkl - Label encoders
label_encoders.pkl
Other components
- Applicant
Connections 7
- Applicant → Input and validation: enters data
app.py - Input and validation → Applicant summary: displays summary
app.py - Input and validation → Prediction flow: provides validated data
app.py - Prediction flow → Feature scaler: scales features
app.py - Prediction flow → XGBoost model: predicts with
app.py - Prediction flow → Random Forest model: predicts with
app.py - Prediction flow → Risk results: renders prediction
app.py