KitchenTeacher is an interactive web application that predicts dishes from food images, provides YouTube recipe videos, and fetches detailed recipe instructions. This project utilizes Gradio for the user interface and leverages Hugging Face's model inference capabilities for dish prediction.
- Dish Prediction: Upload an image of a food item, and KitchenTeacher will predict the dish using a pre-trained model from Hugging Face.
- YouTube Recipe Videos: Once the dish is predicted, KitchenTeacher fetches YouTube recipe videos related to the predicted dish, allowing users to explore different recipes.
- Detailed Recipe Instructions: Additionally, KitchenTeacher fetches detailed recipe instructions from TheMealDB API, providing users with step-by-step instructions and ingredient lists.
To run KitchenTeacher locally, follow these steps:
-
Clone the repository:
git clone https://huggingface.co/spaces/Slfagrouche/KitchenTeacher
-
Navigate to the project directory:
cd KitchenTeacher -
Install dependencies:
pip install -r requirements.txt
-
Run the application:
python your_gradio_script.py
Or refer to setup instructions for more details on setting up KitchenTeacher on Windows or macOS.
- Open the application in your web browser.
- Upload an image of a food item.
- KitchenTeacher will predict the dish, fetch YouTube recipe videos, and provide detailed recipe instructions.
You can try out the live version of KitchenTeacher by visiting the following link:
Please note that the live app may have limited functionality compared to running the application locally.
Contributions are welcome! If you'd like to contribute to KitchenTeacher, please feel free!
This project is licensed under the MIT License - Hugging Face.
- This project makes use of Gradio for building the interactive interface.
- Model inference is powered by Hugging Face.
- Recipe data is fetched from TheMealDB.
For any questions or issues, please open an issue on GitHub.