How gireesh2658/rising-water works
Rising Waters is a Flask flood-risk prediction application with account-based access, model inference, and prediction history. Users register or log in, submit meteorological inputs, and receive a risk classification and confidence score; authenticated predictions are persisted for later review. Offline training and synthetic-data generation support the deployed model, while evaluation reports are available through the application. Source excerpts are partial, so links for some documented capabilities rely on the README; unsampled implementation details are not inferred.
Main entry points: app.py Prediction endpoint · train.py Training pipeline · main.js Form interactions · flood_dataset_expanded.csv Expanded flood dataset
Components 16
Web application
- Web interface
templates - Form interactions
static/main.js - Account authentication
app.py - Prediction endpoint
app.py - History endpoints
app.py - Evaluation dashboard
app.py
Machine learning
- Flood classifier
models/floods.save - Feature scaler
models/transform.save - Feature names
models/feature_names.pkl - Training pipeline
train.py - Model evaluation reports
reports
Data preparation
- Synthetic data generation
data/generate_synthetic_data.py - Dataset analysis
data/analyze_dataset.py - Expanded flood dataset
data/flood_dataset_expanded.csv
Persistence
- PostgreSQL database
Other components
- Browser user
Connections 21
- Browser user → Web interface: opens pages
README.md - Web interface → Form interactions: loads interactions
README.md - Browser user → Account authentication: registers and logs in
README.md - Account authentication → PostgreSQL database: reads and writes accounts
README.md - Browser user → Prediction endpoint: submits weather
README.md - Form interactions → Prediction endpoint: posts inputs
README.md - Prediction endpoint → Feature scaler: scales features
README.md - Prediction endpoint → Feature names: uses feature order
README.md - Prediction endpoint → Flood classifier: classifies risk
README.md - Prediction endpoint → PostgreSQL database: logs prediction
app.py - Prediction endpoint → Form interactions: returns risk result
README.md - Browser user → History endpoints: reviews or clears history
README.md - History endpoints → PostgreSQL database: queries and deletes records
app.py - Browser user → Evaluation dashboard: views evaluation
README.md - Evaluation dashboard → Model evaluation reports: serves report files
app.py - Training pipeline → Flood classifier: exports classifier
README.md - Training pipeline → Feature scaler: exports scaler
README.md - Training pipeline → Model evaluation reports: generates evaluations
train.py - Synthetic data generation → Expanded flood dataset: generates dataset
data/generate_synthetic_data.py - Dataset analysis → Expanded flood dataset: analyzes dataset
- Training pipeline → Expanded flood dataset: trains from data