xjmaine/gis_agriculture_explorer_ghana

Streamlit dashboard for Ghana agriculture data # Ghana Agriculture Analytics Platform Self-contained data pipeline + FastAPI backend + Streamlit dashboard for Ghana agriculture data. ## Components - `scripts/` — download (FAOSTAT, World Bank, HDX), clean, merge, and analyse the data. - `app/` — FastAPI backend (DDD-style self-contained domains) exposing `/api/v1/*` endpoints. - `dashboard/` — Streamlit multi-page frontend consuming the API. - `data/raw`, `data/cleaned`, `data/merged`, `output/` — pipeline artifacts. ## Run Full setup, start/stop and troubleshooting instructions: see **START_SERVERS.md**. Quick start (Windows) — double-click or run: ``` run_app.bat ``` Standalone (no API required — the dashboard reads the data files directly): ``` streamlit run dashboard/app.py ``` Full stack (API + dashboard): ``` python -m uvicorn app.main:app --host 127.0.0.1 --port 8000 # terminal 1 streamlit run dashboard/app.py # terminal 2 ``` The dashboard is self-contained; it only talks to the API when `AGRI_API_URL` is set. ## Deploy to Streamlit Cloud (single app) The dashboard runs as a single self-contained app. Push this repo to GitHub, then create an app in Streamlit Cloud pointing at **`dashboard/app.py`**. The data under `data/cleaned`, `data/merged` and `data/geo` is committed (raw downloads are gitignored). Optionally set the `AGRI_API_URL` secret to use a remote API instead. ## Reproduce the data ``` python scripts/download_faostat.py python scripts/download_worldbank.py python scripts/download_hdx.py python scripts/clean_data.py python scripts/merge_data.py python scripts/analyse_data.py ``` ## API overview `GET /api/v1/analytics/overview` — dataset summary (also `/trends`, `/correlation`, `/search`). Domain routers: `geographical`, `crop-production`, `commodity-price`, `food-security`, `livestock`, `climate-agriculture`, `land-use`. Interactive docs at `/docs`.