Elmahi92/undp_poc

Interactive Streamlit dashboard for monitoring humanitarian interventions across Sudan. Features a login system, dynamic Plotly map with admin boundaries, KPI metrics, filterable project table, and AI-powered insights via Ollama. Filters by agency, sector, location, status, and date range. Sudan Intervention Dashboard An interactive web dashboard for monitoring and analysing humanitarian interventions across Sudan, built with Streamlit and Plotly. Features Secure Login — Password-protected access with SHA-256 hashed credentials via Streamlit secrets Interactive Map — Dark-themed Plotly map with Sudan admin2 boundaries, colour-coded project dots, and rich hover tooltips KPI Metrics — At-a-glance cards for total interventions, projects, states covered, and implementing agencies Dynamic Filters — Sidebar filters for date range, agency, sector, location, project, and status Project Table — Paginated, downloadable CSV table of all filtered projects AI Insights — On-demand analysis via built-in summary or local Ollama LLM (Llama 3.2 / Mistral) Requirements Python 3.9+ Streamlit Pandas Plotly Requests Install all dependencies: pip install -r requirements.txt Setup Clone the repository git clone github.com cd sudan-intervention-dashboard Add your credentials Create a .streamlit/secrets.toml file: [credentials] admin = "your\sha256\hashed\_password" To generate a SHA-256 hash of your password in Python: import hashlib print(hashlib.sha256("your\_password".encode()).hexdigest()) Add your data files Place the following files in the root directory: project\summary\v2.csv sudan\_admin2.geojson United-Nations\_logo.webp Run the app streamlit run app.py Data Format The CSV file must contain the following columns: ChfProjectCode — Unique project code ProjectTitle — Full project title OrganizationName — Implementing agency name ActualStartDate — Project start date (e.g. 10-Apr-14) ActualEndDate — Project end date (e.g. 31-Mar-15) ProjectStatus — Status (Ongoing, Completed, Pipeline, Cancelled) Cluster — Sector or intervention type AdminLocation1 — State name AdmLoc2 — Locality name Lat\_Admin2 — Latitude at locality level Lon\_Admin2 — Longitude at locality level Budget — Project budget in …