How sayamalt/weather-prediction-using-machine-learning works
This is a small Flask weather-prediction app: a user submits weather measurements, the app scales them, predicts a weather class, and renders the decoded result. The sampled app source establishes this runtime flow; the notebooks and dataset are present, but their contents were not inspected, so no training or deployment workflow is inferred.
Main entry points: app.py Flask app · index.html Prediction page · scaler.pkl Feature scaler · model.pkl Weather model
Components 6
Web interface
- Prediction page
templates/index.html
Prediction runtime
- Flask app
app.py
Model artifacts
- Feature scaler
scaler.pkl - Weather model
model.pkl - Label encoder
label_encoder.pkl
Other components
- User
Connections 12
- User → Flask app: requests page
- Flask app → Prediction page: renders
- Prediction page → User: presents form
- User → Flask app: submits measurements
- Flask app → Feature scaler: transforms features
- Feature scaler → Flask app: returns scaled data
- Flask app → Weather model: predicts class
- Weather model → Flask app: returns prediction
- Flask app → Label encoder: loads classes
- Flask app → Prediction page: renders prediction
- Flask app → Feature scaler: loads artifact
- Flask app → Weather model: loads artifact