HELT is an innovative web application designed to provide personalized food recommendations based on users' health profiles. By analyzing nutritional information from food images, HELT evaluates the suitability of food items for individual users, enabling informed dietary choices.
- Image Upload: Users can upload food images to extract nutritional information.
- Health Analysis: Users input their weight, height, age, and gender to receive a health classification.
- Personalized Recommendations: Tailored food recommendations based on the user's health profile.
- Dark Mode: A toggle option for light and dark themes for enhanced user experience.
- Flask: A lightweight web framework for building the application.
- HTML/CSS/JavaScript: For the frontend user interface and styling.
- RoBERTa: A transformer-based model for understanding the context of nutritional information.
- Facebook BART (facebook/bart-large-mnli): Utilized for classification tasks and generating personalized recommendations.
- EasyOCR: For Optical Character Recognition (OCR) to extract nutritional information from food images.
To set up the project locally, follow these steps:
- Clone the repository:
git clone https://github.com/pratapsdev11/helt