Skip to content

README.md

Latest commit

 

History

History
92 lines (66 loc) · 2.71 KB

README.md

File metadata and controls

92 lines (66 loc) · 2.71 KB

PlotSense Data Storytelling App

An interactive Streamlit web app that uses PlotSense AI to recommend and explain visualizations for climate data.

This project was built as part of the PlotSense AI Hackathon to demonstrate how storytelling with data can be made simple and extensible.

🚀 Features

  • Load a real climate dataset (climate.csv) with Temperature, Humidity, Wind Speed, Rainfall.

  • Filter data by City and choose a variable of interest.

  • Get AI-powered recommendations for visualizations using PlotSense.

  • Preview raw data (first 10 rows) before visualization.

  • Render visualizations directly inside the web app.

  • Generate AI explanations for each chart using your Groq API key.

📂 Project Structure

plotsense-data-storytelling/
├── app.py                # Main Streamlit app
├── data/
│   └── climate.csv       # Sample climate dataset
├── requirements.txt      # Python dependencies
├── README.md             # Project documentation
└── assets/               # Store screenshots for README

⚙️ Installation & Setup

  1. Clone this repository
git clone https://github.com/Havilah-Blockchain-Studios/PlotSenseAI-Hackathon-Demo-Projects.git
cd project_three
  1. Create a virtual environment (recommended)
python -m venv .venv
source .venv/bin/activate   # macOS/Linux
.venv\Scripts\activate      # Windows
  1. Install dependencies
pip install plotsense
pip install streamlit
pip install -r requirements.txt

This installs:

  • plotsense (AI visualization tool)
  • streamlit (for the web app)
  • pandas (for data handling)

▶️ Running the App

From the project root, run:

python -m streamlit run app.py

Then open the local server URL shown in your terminal (usually http://localhost:8501 ).

🔑 Groq API Key Setup

Some features (AI explanations) require a Groq API key:

  1. Create a free account at Groq Console.

  2. Copy your key (looks like gsk_xxxxx).

  3. Paste it into the sidebar input box inside the Streamlit app.

  • The input is hidden (password-style).
  • Each participant can use their own key without editing code.

🖼️ Screenshots

Recommendations + Visualization

screenshot_home

AI Explanation

ai-explainer

🛠️ Extending the App

This project is hackathon-extensible:

  • Replace climate.csv with your own dataset (data/yourdata.csv).
  • Add more variables or filters in app.py.
  • Customize the layout in Streamlit for different storytelling themes.