How searph-s0708023/taap_surakshak works
This repository is a SQLite-backed Mumbai heat and health data pipeline. Its principal workflow fetches daily weather, stores and derives climate measures, exposes joined daily features, then exports ML-ready datasets. The tree also includes population and vulnerability data files, but no sampled code shows them being loaded. The README is empty, and calculate_thermal.py is empty, so those capabilities and their wiring are not inferred.
Main entry points: mumbai_heat_health.db SQLite database · process_data.py Fill and clean data · create_view.py Daily ML feature view · create_database.py Create database
Components 19
Data Setup
- SQLite database
mumbai_heat_health.db - Database schema
schema.sql - Create database
create_database.py - Fetch weather
fetch_weather.py - Populate dates
populate_dates.py - Populate locations
populate_locations.py
Climate Processing
- Heat index
calculate_heat_index.py - Temperature anomaly
calculate_anomaly.py - Weather lags
calculate_lags.py - Thermal metrics
calculate_thermal.py - Climate aggregates
aggregate.py
Feature Assembly
- Build feature view
create_view.py - Daily ML feature view
create_view.py
Data Outputs
- Export ML-ready data
export.py - ML-ready CSV
mumbai_ml_ready.csv - Fill and clean data
process_data.py - Filled dataset
mumbai_ml_filled.csv - Strict cleaned dataset
mumbai_ml_cleaned_no_empty_cols.csv
Other components
- NASA POWER
Connections 18
- Create database → Database schema: reads schema
- Create database → SQLite database: creates schema
- NASA POWER → Fetch weather: returns weather
- Fetch weather → SQLite database: writes observations
- Populate dates → SQLite database: populates dates
- Populate locations → SQLite database: populates wards
- Heat index → SQLite database: updates heat index
- Temperature anomaly → SQLite database: updates anomalies
- Weather lags → SQLite database: updates lag values
- Thermal metrics → SQLite database
- Climate aggregates → SQLite database: refreshes aggregates
- Build feature view → SQLite database: creates view
- Build feature view → Daily ML feature view: defines
- Export ML-ready data → Daily ML feature view: queries
- Export ML-ready data → ML-ready CSV: writes CSV
- Fill and clean data → Daily ML feature view: queries
- Fill and clean data → Filled dataset: writes CSV
- Fill and clean data → Strict cleaned dataset: writes CSV