converted core ml model doen't include classes in metadata

[dhgokul]

converted yolo5s.pt model to core ml model using cmd

!python models/export.py --weights weights/yolov5s.pt --img 640 --batch 1 --train

in previous tried cmd
!python models/export.py --weights weights/yolov5s.pt --img 640 --batch 1 but coreml model fails to convert

once the core ml model tried in ios below result got

here no class found.

Tried in mac x code

[github-actions]

👋 Hello @dhgokul, thank you for your interest in 🚀 YOLOv5! Please visit our ⭐️ Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution.

If this is a 🐛 Bug Report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom training ❓ Question, please provide as much information as possible, including dataset images, training logs, screenshots, and a public link to online W&B logging if available.

For business inquiries or professional support requests please visit https://www.ultralytics.com or email Glenn Jocher at glenn.jocher@ultralytics.com.

Requirements

Python 3.8 or later with all requirements.txt dependencies installed, including torch>=1.7. To install run:

$ pip install -r requirements.txt

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

• Google Colab and Kaggle notebooks with free GPU:
• Google Cloud Deep Learning VM. See GCP Quickstart Guide
• Amazon Deep Learning AMI. See AWS Quickstart Guide
• Docker Image. See Docker Quickstart Guide

Status

If this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training (train.py), testing (test.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

[dhgokul]

not able to convert yolo5s model to core ml model . as suggessted command
!python models/export.py --weights weights/yolov5s.pt --img 640 --batch 1

tried both linux cpu machine and in google colab. In both cases not showing the exact error on conversion

so just removed exception in coreml conversion part in model.py file , got the error as follows in google colab

just installed all in the requirement.txt

how to over come the issue.

[glenn-jocher]

@dhgokul good news 😃! Your original issue may now been fixed ✅ in PR #3262. This PR improves error handling for CoreML exports, displaying an error message that notifies the user that CoreML exports should be done when the model is in .train() mode with python models/export.py --train. To receive this update you can:

• git pull from within your yolov5/ directory
• git clone https://github.com/ultralytics/yolov5 again
• Force-reload PyTorch Hub: model = torch.hub.load('ultralytics/yolov5', 'yolov5s', force_reload=True)
• View our updated notebooks:

Thank you for spotting this issue and informing us of the problem. Please let us know if this update resolves the issue for you, and feel free to inform us of any other issues you discover or feature requests that come to mind. Happy trainings with YOLOv5 🚀!

[dhgokul]

thanks for the quick response

we just pulled the repo and tried conversion

!python models/export.py --weights weights/yolov5s.pt --img 640 --batch 1 --train
the generated core model is tested on x code, but we have not any class labels, generally in other core ml mode list in metadata

here is the converted ml model of yolov5s.pt
yolov5s_3.zip
we tried this ml model only in ios x code.

[yohandarosa]

anyone had find a solution to include class labels in the CoreML model ?

[jiwon-hae]

Same issue here... Was this solved?

[Sic-1]

Still seems to be an issue. Any solutions?

[MaxNazarov93]

I figured out this problem and what you need is create ClassifierConfig and pass it into convert function

from coremltools.converters import ClassifierConfig

classifier_config = ClassifierConfig(class_labels=['class1', 'class2', 'class3', 'class4'])
ct_model = ct.convert(ts, inputs=[ct.ImageType('image', shape=im.shape, scale=1 / 255, bias=[0, 0, 0])], classifier_config=classifier_config)

[dhgokul]

Thanks @MaxNazarov93 will check it. Anyone try this and post in this group

Thanks in advance!

[zjsjayce]

I also meet this problem

[BasanthVerma]

I figured out this problem and what you need is create ClassifierConfig and pass it into convert function

from coremltools.converters import ClassifierConfig

classifier_config = ClassifierConfig(class_labels=['class1', 'class2', 'class3', 'class4'])
ct_model = ct.convert(ts, inputs=[ct.ImageType('image', shape=im.shape, scale=1 / 255, bias=[0, 0, 0])], classifier_config=classifier_config)

@MaxNazarov93 this process adds the classes to the model and we can see the classes in the metadata. But when we run the model, we get:
Error Domain=com.apple.vis Code=3 "The VNCoreMLTransform request failed" UserInfo={NSLocalizedDescription=The VNCoreMLTransform request failed, NSUnderlyingError=0x280a81590 {Error Domain=com.apple.CoreML Code=0 "The size of the output layer 'var_887' in the neural network does not match the number of classes in the classifier." UserInfo={NSLocalizedDescription=The size of the output layer 'var_887' in the neural network does not match the number of classes in the classifier.}}}

We tried multiple Yolov5 that we had custom trained, we made sure that class_labels has the exact number of labels that we used for training. But we still keep getting this error.

So we ended up following the script from https://rockyshikoku.medium.com/convert-yolov5-to-coreml-also-add-a-decode-layer-113408b7a848 and it worked!