Dietary Remix is a web application that leverages AI to transform recipes based on different dietary requirements. It allows users to input recipes and convert them to various diets such as vegan, keto, gluten-free. The application uses MindsDB with Google's Gemini Pro model to perform intelligent recipe transformations while maintaining the essence and flavor profile of the original recipe.

• Features
• Technology Stack
• YouTube Demonstration
• Requirements
• Installation
• Usage
• Support

• Add new recipes to a local database
• Transform recipes to different diets:
  • Vegan
  • Keto
  • Gluten-free
• User-friendly web interface for recipe input and transformation
• Display of both original and transformed recipes
• Real-time recipe transformation using AI

• Backend: Python with Flask framework
• Database: SQLite for local storage
• AI Integration: MindsDB with Google Gemini Pro Model
• Frontend: HTML, CSS, JavaScript (jQuery)
• API: RESTful API endpoints for recipe management and transformation

Watch the video

• Python 3.7 or higher
• MindsDB SDK
• SQLite3

1. Clone the repository:

   git clone https://github.com/FaycalZM/dietary-remix
   cd dietary-remix

2. Create virtual env & activate it & Install required packages:

   python3 -m venv venv
   source venv/bin/activate
   pip install -r requirements.txt

3. Set up MindsDB:

   • Follow the MindsDB installation guide to install and run MindsDB locally.
   • Note the server address and port (default is https://127.0.0.1:47334).
   • make .env file add your Google Gemini api key

    API_KEY='YOUR_KEY'

4. Create and populate the SQLite database:

   python data.py

1. Run the Flask server:

   flask --app main run --debug

2. Add a recipe by providing the recipe name and recipe text in the 'add recipe' form:

3. Transform a recipe by providing the recipe name, recipe text and selecting the desired diet in the 'transform recipe' form, and visualize the transformed recipe.

4. Visit https://localhost:5000/get_recipes to get all the recipes in the local database.

If you like this project, please support it by upvoting on Quira and starring the GitHub repository!

Thank you for your support!