GreenFork

• 

• 

• GIF

  GreenFork

Inspiration

Hey there! Let me introduce you to GreenFork. I created it to combine my interest in tech and contribute in SDGs. The idea started pretty simply, what if we had an app that could break down what we eat, tell us if it’s healthy, and show how sustainable our choices are? The app aligns with United Nations Sustainable Development Goals (SDGs), particularly: Goal 2: Zero Hunger Goal 12: Responsible Consumption and Production Goal 13: Climate Action Goal 3: Good Health and Well-being

What it does

GreenFork identifies food items and provides detailed nutritional information, and offers actionable sustainability insights. Food Recognition: Upload or capture an image, and the app identifies the food item with precision. Nutrition Analysis: Get a detailed breakdown of calories, proteins, carbs, and fats, tailored for your dietary goals. Eco-Impact Awareness: Understand the environmental impact of your meals, including carbon footprint, water usage, and seasonality. Sustainability Tips: practical ways to reduce food waste and make greener choices. Healthier Alternatives: Recommendations for nutrient-rich, sustainable substitutes.

How we built it

• I used the SOTA Llama3 vision model to classify food images and Gemma to provide personalized insights.
• Crafted an interactive UI using Streamlit, making the app easy to use.
• To top it off, I added features like where we can track our sustainability impact. ## Challenges we ran into The biggest challenge was data, finding reliable and accurate sources for environmental stats was a real headache. I also hit a few walls with AI models; tweaking them to work efficiently for image classification. On the tech side, debugging JSON errors to get accurate results took some patience (and a lot of coffee).

What we learned

While building GreenFork, It was more about R&D, working with different Deeplearning and HuggingFace models for food classification. Along with that, I explored and experimented with multiple APIs for SDG. One of the coolest things I learned? The way technology can connect the dots between what we eat and how it impacts the planet. Each of these challenges pushed us to innovate, improve, and refine our solution, ultimately making it both robust and impactful.

Built With

• deep-learning
• gemma
• llama-vision
• natural-language-processing
• python