How vikas-rahangdale/predictive-maintenance-ml-model-rvd works
This is a small Streamlit predictive-maintenance app. A user enters machine type and sensor readings, submits them for encoding and scaling, and the saved classifier predicts whether failure is likely; the app displays the result. The source shows the model and scaler being loaded from local pickle files and used at prediction time. No README or other runtime modules are present, so the graph reflects only the supplied app and artifacts.
Main entry points: appp.py Encode and scale · scaler.pkl Saved scaler · best_model.pkl Saved classifier
Components 7
Prediction app
- Machine inputs
appp.py - Encode and scale
appp.py - Saved scaler
scaler.pkl - Failure prediction
appp.py - Saved classifier
best_model.pkl - Prediction message
appp.py
Other components
- Machine operator
Connections 7
- Machine operator → Machine inputs: enters readings
- Machine inputs → Encode and scale: submits
- Encode and scale → Saved scaler: transforms with
- Encode and scale → Failure prediction: passes scaled data
- Failure prediction → Saved classifier: predicts with
- Failure prediction → Prediction message: selects message
- Prediction message → Machine operator: displays result