How to replace the model used in the Android flutter application

[khjang1991]

I have a question, so I leave a message.
We would like to replace the model used in your Android flutter application.
The YOLO11 model at the "https://github.com/ultralytics/ultralytics " address was changed to the tensorflow lite model using the command "!yolo export model=/content/runs/detect/train2/weights/best.pt format=tflite half=True int8=True".
If the application is executed after replacing the model, the application will be forcibly terminated. Similar results were obtained when the YOLO8 model was used.
Any advice on this part?
I will share the error log through the attached file.

error_log_device.txt

[UltralyticsAssistant]

👋 Hello @khjang1991, thank you for submitting a ultralytics/yolo-flutter-app 🚀 Issue. This is an automated response to gather initial information and assist our engineering team. An Ultralytics engineer will review and address your issue soon.

To help us address your concern efficiently, please make sure you've provided the following information:

1. For bug reports:

   • A clear and concise description of the issue you're facing
   • A minimum reproducible example (MRE) that demonstrates the problem. For guidance on creating an MRE, please refer to our documentation ✨
   • Detailed information about your environment (e.g., OS, Flutter version, and any relevant package versions)
   • What you expected to happen versus what actually occurred
   • Any error messages or logs related to the issue (thank you for attaching error_log_device.txt 😄)

2. For feature requests or questions:

   • A detailed description of your request or question, including the context
   • The specific problem you're trying to solve or information you're seeking
   • Any previous research or documentation you've consulted, such as our official documentation

Please ensure that you've searched for existing issues to avoid duplicates. If you have additional details or need to update your report, feel free to comment on this issue.

Thank you for helping us improve our project! 🍀

[pderrenger]

@khjang1991 to resolve the issue, ensure your TFLite model is compatible with the Flutter app's TensorFlow Lite version. Check the error log for specific compatibility issues and verify that the model conversion settings (e.g., half=True, int8=True) are supported by your app's environment. If the problem persists, try using the latest version of the Ultralytics package for model export.