How keysarenull/prediction works
This is a Streamlit heart-disease prediction app. Users enter health indicators, request a prediction, and see an ensemble result with confidence, disagreement warnings, and individual model outputs. The sampled code shows the UI-to-ensemble calls and model loading and prediction; README content was not available to verify additional capabilities.
Main entry points: predictor.py Ensemble predictor · app.py Input and results UI · logistic_regression_heart_disease_model.joblib Logistic regression · random_forest_heart_disease_model.joblib Random forest
Components 7
User interface
- Input and results UI
app.py
Prediction engine
- Ensemble predictor
utils/predictor.py - Logistic regression
model/logistic_regression_heart_disease_model.joblib - Random forest
model/random_forest_heart_disease_model.joblib - Gradient boosting
model/gradient_boosting_heart_disease_model.joblib - XGBoost
model/xgboost_heart_disease_model.joblib
Other components
- User
Connections 7
- User → Input and results UI: enters health data
- Input and results UI → Ensemble predictor: requests prediction
- Ensemble predictor → Logistic regression: loads and predicts
- Ensemble predictor → Random forest: loads and predicts
- Ensemble predictor → Gradient boosting: loads and predicts
- Ensemble predictor → XGBoost: loads and predicts
- Ensemble predictor → Input and results UI: returns ensemble result