Unable to accurately identify

[136323000]

I copied my mlpackage file into this demo . it can be accurately recognized in the mlpackage file of the x-code, just like the image below:

But when I run the code, these are displayed：

My steps are as follows:

1. Prepare my dataset
   Train them using: yolo detect train data=xxx. yaml model=yolov8n. yaml pretrained=ultrasonic/yolov8n. pt epoch=100 batch=4
2. I obtained a pt format file
   I ran the following code on PyCharm

from ultraviolet imports YOLO
Model=YOLO (f "best10. pt")
Model. export (format="coreml", int8=True, nms=True, imgsz=[640, 384])

I received a xxx. mlpackage file . I copy 5 copies ，and renamed this file as yolov8l. mlpackage, yolov8m. mlpackage, yolov8n. mlpackage, yolov8s. mlpackage, yolov8x. mlpackage, and copied it to the models file of the project you provided and run demo
get:

[pderrenger]

Hello! It looks like there might be an issue with how the model is being converted or loaded into your iOS project. Let's check a few things:

1. Confirm that the .mlpackage conversion has all necessary model parameters included and is properly configured for iOS.
2. Ensure that the .mlpackage files are correctly placed in the designated model directory of your iOS project.
3. Verify if the outputs from your training step closely resemble the .mlpackage expectations.

It's also helpful to ensure that the configuration used for training (data=xxx.yaml, model=yolov8n.yaml) matches the settings expected by your iOS application.

If everything above checks out, try testing one .mlpackage file at a time to isolate the variable causing the issue.

For further fine-tuning and configuration guidance, you might find the documentation useful at https://docs.ultralytics.com. Let's get your app accurately recognizing objects soon! 🚀