How cloud-strife-182000/carbon-finance-project-group-7 works
Project type and purpose
• This is a data‐visualization dashboard built as a single‐service Python app using Streamlit (a “full‐stack” in the sense of UI + backend in one process).
• Purpose: interactive exploration of ESG and green bond metrics in Indian capital markets.
File structure
• Top level:
• streamlit_app.py – main application entrypoint
• data/ – CSV data sources (ESG scores, green bond data)
• requirements.txt – Python dependencies
• .devcontainer/ – VS Code devcontainer config
• .github/ – repository governance (CODEOWNERS)
• No separate frontend vs backend directories: everything lives in one Python script.
README insights
• Tech stack: Python, Streamlit, Altair, Pandas (implied), CSV data files.
• Deployed live on Streamlit Community Cloud.
• Core functionality: data ingestion → transformation/filtering → Altair visualizations → Streamlit UI.
Main components & their relationships
a. Components
• User Browser (Client)
• Streamlit Server (Python process hosting streamlit_app.py)
• Data Storage (local CSV files under data/)
• Data Processing Layer (Pandas; filter, aggregate)
• Visualization Layer (Altair; charts, heatmaps, KPIs)
• Deployment/Dev Environment (Docker devcontainer, GitHub, Streamlit Cloud)
b. Interactions
1. User interacts with UI controls (year/sector filters) in Browser
2. Browser sends requests to Streamlit Server
3. Server loads CSVs from data/, applies Pandas transforms
4. Processed data fed into Altair charts
5. Streamlit sends rendered HTML/JS charts back to Browser
6. Code and environment config managed in GitHub & .devcontainer; deployed to Streamlit Cloud
c. Architectural patterns
• Monolithic, event‐driven data‐pipeline within a single process
• “MVC‐lite”: UI widgets (View) → Python callbacks (Controller) → Pandas (Model)
d. Key technologies
• Streamlit: handles routing of UI events and rendering
• Pandas: data transformation
• Altair: declarative visualization
• Docker devcontainer: reproducible local setup
• Streamlit Community Cloud: deployment host
Detail
• Every essential piece—data sources, code, UI, dev environment, and deployment endpoint
• A “Developer Workflow” lane (GitHub → Devcontainer → Local Streamlit Server)
• A “Production” lane (GitHub → CI/CD → Streamlit Community Cloud)
• Data stores (CSV files) feeding into the runtime block
• Browser client on the right receiving the rendered dashboard