Fridge to Dish App is a web application that helps users find dish suggestions based on the ingredients they have in their fridge. Users can upload an image of their ingredients, and the app will identify the ingredients and provide dish suggestions along with their instructions. Users can also generate an image for each suggested dish.
- Ingredient recognition from an image
- Dish suggestions based on the ingredients
- Generating images for the suggested dishes
- Clone the repository
git clone https://github.com/yourusername/fridge-to-dish-app.git
- Change the working directory
cd fridge-to-dish-app
- Create a virtual environment
python -m venv venv
- Activate the virtual environment
# For Windows
venv\Scripts\activate
# For Linux/Mac
source venv/bin/activate
- Install the dependencies
pip install -r requirements.txt
- Start the Streamlit app
streamlit run app.py
-
Open the app in your browser using the provided URL
-
Upload an image of your ingredients
-
Review the detected ingredients and suggested dishes
-
Click on the "Generate Image" button to see an image of each suggested dish
- Python
- Streamlit
- OpenAI GPT-3 (API Key required)
- OpenAI CLIP
- Runwayml Stable Diffusion v1-5
- PIL (Python Imaging Library)
- Diffusers
This project is licensed under the MIT License.
You can find a version of this working on my huggingface space. Feel free to clone it however you will need to run it with atleast the nvidia T4 GPU becaus the diffusion model requires too much resources it cant run on the base CPU.
If you have any questions about this project or would like to contact the maintainer, please send an email to @samchakwera.