How kavin553/temperature_prediction_forecast_model works
This Flask application serves a temperature-safety dashboard and a prediction endpoint. The main workflow is user input → timestamp/feature preparation → model forecast → unit conversion and threshold alert → dashboard results. A separate training script combines two temperature datasets, engineers features, trains XGBoost, and saves artifacts consumed by the app. README-described dashboard visualizations and risk presentation are included, but demo.html was not sampled, so their implementation details and wiring remain unverified.
Main entry points: app.py Flask Routes · demo.html Dashboard UI · train_xgb.py Training Workflow · xgb_model_combined.joblib XGBoost Model
Components 16
Dashboard
- Dashboard UI
demo.html - Flask Routes
app.py - Trend Visuals
demo.html
Forecasting
- Feature Preparation
app.py - XGBoost Model
xgb_model_combined.joblib - Model Features
feature_info_combined.joblib - Temperature Forecast
app.py
Model Training
- Training Workflow
train_xgb.py - Hourly Dataset
MLTempDataset1.csv - Power Dataset
MLTempDataset.csv - Trained Model
xgb_model_combined.joblib - Feature Metadata
feature_info_combined.joblib
Safety Results
- Unit Conversion
app.py - Threshold Alert
app.py - Risk Display
demo.html
Other components
- User
Connections 16
- User → Dashboard UI: enters readings
README.md - Dashboard UI → Flask Routes: requests forecast
- Flask Routes → Feature Preparation: builds features
app.py - Flask Routes → XGBoost Model: loads model
app.py - Flask Routes → Model Features: loads features
app.py - Flask Routes → Feature Preparation: uses feature list
app.py - Flask Routes → Temperature Forecast: predicts temperature
app.py - Temperature Forecast → Unit Conversion: converts units
app.py - Unit Conversion → Threshold Alert: checks threshold
app.py - Flask Routes → Dashboard UI: returns results
- Dashboard UI → Risk Display: shows risk
README.md - Dashboard UI → Trend Visuals: renders trends
README.md - Training Workflow → Hourly Dataset: reads data
train_xgb.py - Training Workflow → Power Dataset: reads data
train_xgb.py - Training Workflow → Trained Model: saves model
train_xgb.py - Training Workflow → Feature Metadata: saves features
train_xgb.py