Food-Vision 🍜 📸

Built an end-to-end CNN Image Classification Model which identifies the food in your image.

I have used an pre-trained Image Classification Model, EfficientNetB0, previously pre-trained on ImageNet, that comes with Tensorflow and then retrained it on Food 101 Dataset dataset.

Fun Fact

This model actually beats the DeepFood Paper's model which also trained on the same dataset.

The Accuracy of DeepFood was 77.4% and our models is 79%. It is not much but, we take those.

How can you use it?

Finally after training the model, I have exported as .h5 files and then integrated it with Streamlit Web App

Try it out yourself, Click Here!

Once the app is loaded,

1. Choose the input type. Whether you want to upload an image or enter URL of an image
2. For upload an image option, choose image from your computer.
3. For URL option, erase the existing url and enter your own url and press enter
4. Once you have done the above, the model will predict the output and display the Top-5 Predictions

Installation

1. Clone the repo

   git clone https://github.com/mSounak/Food-Vision

2. install requirements.txt

   pip install -r requirements.txt

3. run app.py

   streamlit run app.py

Contributing

Contributions are what make the open source community such an amazing place to be learn, inspire, and create. Any contributions you make are greatly appreciated.

1. Fork the Project
2. Create your Feature Branch (git checkout -b feature/AmazingFeature)
3. Commit your Changes (git commit -m 'Add some AmazingFeature')
4. Push to the Branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

---