Sustainaplate

Project Description

Sustainaplate Backend is a critical component of the Sustainaplate project, developed for the Migros Challenge at HackZurich 2023. It is the backbone of our user-friendly Recipe Recommender Web Application, designed to encourage eco-conscious shopping and culinary exploration. The backend, built using FastAPI and various AI technologies, seamlessly integrates with the front end to provide users with personalized recipe recommendations based on their available ingredients. It also offers sustainability analysis to empower users to make environmentally responsible ingredient choices for their shopping lists.

---

1. Key features

• FastAPI Integration: Our backend utilizes FastAPI, a Python web framework, to handle HTTP requests and efficiently serve data to the front end. -Custom API: We've created a custom API that fetches and delivers recipe and sustainability data, enabling real-time recipe recommendations and sustainability analysis.
• AI-Powered Recommendation Engine: Leveraging AI algorithms, including BamLLM, our backend analyzes user-inputted ingredients and offers personalized recipe suggestions while considering sustainability factors.
• Sustainability Scoring: The backend includes a sustainability scoring system that rates the environmental impact of ingredients, helping users make eco-conscious choices.

2. Usage

• The backend is seamlessly integrated with the Sustainaplate web application and responds to user requests from the front end.
• It handles various endpoints for recipe queries, ingredient analysis, and real-time recipe generation based on user inputs.
• Users can input their available ingredients, dietary restrictions, meal type, or other preferences in English to receive personalized recipe recommendations.

Technologies Used

• FastAPI: A Python web framework used to build the backend server and handle HTTP requests.
• BamLLM: A language model used for text generation and natural language processing.
• Llama Index: A library for document indexing and retrieval.
• Pandas: A data manipulation library used for data analysis and processing.
• NumPy: A library for numerical computations, often used with Pandas.
• Difflib: A library for comparing sequences, used for comparing words in recipe ingredients.

3. Links

• The UI for this App was created by our team and it can be found here
• Further info about HackZurich
• Further info about Migros
• MaterialUI