[Bug]: Fail to read YOLOv8 Pytorch model

[openvino-book]

OpenVINO Version

2024.1.0-15008-f4afc983258-releases/2024/1

Operating System

Windows System

Device used for inference

CPU

Framework

PyTorch

Model used

yolov8n-cls.pt, yolov8n.pt and yolov8n-seg.pt

Issue description

I learnt OpenVINO 2024.1 can read the Pytorch model: xxx.pt directly as below:

However, I fail to read the yolov8n-cls.pt, yolov8n.pt and yolov8n-seg.pt model using the code as below

import openvino as ov

model_path = r".\yolov8n-cls.pt"

# Read the Pytorch model
print(ov.get_version())
core = ov.Core()

model = core.read_model(model_path)
compiled_model = core.compile_model(model, "CPU")

Error will happen as below, when read yolov8n-cls.pt, yolov8n.pt and yolov8n-seg.pt by read_model()

Step-by-step reproduction

Step 1: set up the environment

pip install openvino ultralytics

Step 2: export yolov8n.pt

yolo export model=yolov8n-cls.pt format=onnx

Step 3: execute the code

import openvino as ov

model_path = r".\yolov8n-cls.pt"

# Read the Pytorch model
print(ov.get_version())
core = ov.Core()

model = core.read_model(model_path)
compiled_model = core.compile_model(model, "CPU")

Relevant log output

No response

Issue submission checklist

• I'm reporting an issue. It's not a question.
• I checked the problem with the documentation, FAQ, open issues, Stack Overflow, etc., and have not found a solution.
• There is reproducer code and related data files such as images, videos, models, etc.

[andrei-kochin]

Hello @openvino-book,
Thank you for reaching the OpenVINO!

PyTorch native model representation is python so we only support the PyTorch model through the convert_model Python API(See docs).

Also there is a notebook on how to deal with the Yolov8. Check this out here.

Thank you!