README.md

ml-food-classification

Food Classification using Convolutional Neural Network (CNN) and deployed using TensorFlow Serving.

Notebook: MamMates Food Classification

Dataset: MamMates Dataset

Features 💡

Using Mammates Food Classification, you can categorize food images into 10 classes.

Class	Label
0	bika_ambon
1	dadar_gulung
2	donat
3	kue_cubit
4	kue_klepon
5	kue_lapis
6	kue_lumpur
7	kue_risoles
8	putu_ayu
9	roti

Prerequisites 📋

• TensorFlow 2.14.0 or higher
• Docker 24.0.7 or higher

Usage ✨

If you already have Docker installed, you only need to run the following command:

• Pull the image from Docker Hub:

docker pull putuwaw/mammates-food-classification

• Run the image:

docker run -p 8501:8501 --name ml-clf putuwaw/mammates-food-classification

• You can check that the model is already running by opening the browser and go to http://localhost:8501/v1/models/food_clf

• To do prediction, you can use the following command:

curl -s https://raw.githubusercontent.com/MamMates/ml-food-classification/main/example.json | curl -X POST -d @- http://localhost:8501/v1/models/food_clf:predict

• You will get the following response:

{
  "predictions": [
    [
      6.28405522e-11, 7.40732e-6, 0.998946607, 1.49191326e-8, 0.000139753625,
      2.86315444e-5, 0.000863699941, 6.22894277e-7, 1.15933371e-5, 1.64414064e-6
    ]
  ]
}

Development 💻

If you want to develop this model, you can follow the steps below:

• Clone this repository:

git clone https://github.com/MamMates/ml-food-classification.git

• Update the model by changing the saved model in the model folder.

• Build the Docker image:

docker build -t mammates-food-classification .

• Run the image:

docker run -p 8501:8501 --name ml-clf mammates-food-classification

• You can check that the model is already running by opening browser and go to http://localhost:8501/v1/models/food_clf

• To do prediction, you can use the following command:

curl -d @example.json -X POST http://localhost:8501/v1/models/food_clf:predict

• To stop the container:

docker stop ml-clf

Note

If you want to learn more about TensorFlow Serving, you can read the REST API documentation here.

Acknowledgements 🙏

Our sincere gratitude goes to the creators and maintainers of these datasets. Their generosity in sharing these resources has been instrumental in driving the progress and success of this project.

• Deteksi Makanan Daerah Dataset
• Indonesian Food Dataset
• Food 101 Dataset
• Food-5K Dataset
• Food-11 Dataset

License 📝

This project is licensed under the MIT License. See the LICENSE file for details.