How pooja30-ux/weather-predection-system works
This repository is a weather-temperature prediction project: Python scripts train and evaluate a Random Forest model, and a Flask endpoint accepts weather features and returns a temperature prediction. The request path is visible in app.py; both training scripts save the model artifact. The API also loads scaler.pkl, which is absent from the supplied tree, and no scaler-training flow is shown. HTML/CSS files are present, but their contents and integration are not evidenced by the sampled source.
Main entry points: app.py Flask API · temperature_model.pkl Random Forest · train.py Training and evaluation · weather.py Cleaned training pipeline
Components 8
Prediction API
- Flask API
app.py - Feature scaler
- Random Forest
temperature_model.pkl - Temperature result
app.py
Model Training
- Training and evaluation
train.py - Cleaned training pipeline
weather.py
Other components
- Prediction client
- Weather dataset
Connections 8
- Prediction client → Flask API: sends features
- Flask API → Feature scaler: scales input
- Flask API → Random Forest: predicts with
- Flask API → Temperature result: returns prediction
- Weather dataset → Training and evaluation: loads data
- Training and evaluation → Random Forest: saves model
- Weather dataset → Cleaned training pipeline: loads data
- Cleaned training pipeline → Random Forest: saves model