Object Detection
ONNX
food

Open Food Facts Nutriscore object detection model

This object detection model was trained on images from the Open Food Facts database to detect Nutri-score labels on food packaging.

The v1.1 of the nutrition-table-detection dataset was used, but with the nutrition-table class only (the other classes were ignored).

It was trained on 100 epochs using Ultralytics library (version 8.2.48), using yolov8n version, with images resized to 640x640. This model is licensed under the AGPLv3 license.

Weights

Weights are available in the weights/ directory. An ONNX export of the model is available in weights/model.onnx.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Examples
Unable to determine this model's library. Check the docs .

Dataset used to train openfoodfacts/nutrition-table-yolo