Skip to content

mudit108-code/ECOSENSE

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

🌿 EcoSense — Real-Time Sustainable Decision Engine

Make every decision count for the planet.

EcoSense is an AI-powered sustainability advisor that analyzes your real-world decisions — commuting, ordering food, shopping, energy use — and instantly returns an EcoScore, CO₂ estimate, 3 ranked alternatives, and a one-tap nudge, all powered by Google Gemini.

✨ Features

Feature Description
🌿 EcoScore (0–100) Composite sustainability score: CO₂ (50%) + water (30%) + waste (20%), normalised to 0–100
💨 CO₂ Estimate Per-decision carbon footprint in kg CO₂e
🔀 3 Smart Alternatives Ranked alternatives with cost delta, time delta, convenience rating
💡 One-Line Nudge Punchy, actionable advice at the moment of decision
🌍 Fun Fact A surprising eco-stat related to your specific scenario
💬 Eco Chat Persistent sidebar chatbot for follow-up sustainability questions
📋 History Last 4 analyses tracked per session with verdict icons
⚙️ Preference Sliders Location, convenience bias, budget sensitivity — Gemini adapts recommendations
6 Categories Transport · Food · Shopping · Energy · Travel · Home

🚀 Quick Start

1. Clone the repo

git clone repository URL
cd repo name

2. Create a virtual environment

python3.10 -m venv .venv
source .venv/bin/activate       

3. Install dependencies

pip install -r requirements.txt

4. Add your API key

Create .streamlit/secrets.toml:

5. Run locally

streamlit run app.py

Open http://localhost:8501


🧠 How the EcoScore Works

EcoScore = 0.5 × CO₂_normalized
         + 0.3 × water_normalized
         + 0.2 × waste_normalized
         
         × convenience_adjustment
  • 0 = worst possible environmental impact
  • 100 = best sustainable option available
  • Gemini computes the score contextually per scenario and location
  • Alternatives are always ranked highest-to-lowest EcoScore

Releases

No releases published

Packages

 
 
 

Contributors

Languages