Not working with yolov5 6.2 and yolov8

[AI-M-BOT]

import onnx
import torch
from onnx2torch import convert

from onnx import version_converter


# Load the ONNX model.
model = onnx.load("yolov5n.onnx")
# Convert the model to the target version.
target_version = 14
converted_model = version_converter.convert_version(model, target_version)
torch_model_1 = convert(converted_model)
torch_model_1.eval()
torch.save(torch_model_1, 'model.pt')

For yolov5n.onnx it will raise error:
NotImplementedError: Converter is not implemented (OperationDescription(domain='', operation_type='ArgMax', version=13))

For yolov8n.onnx the converted model goes wrong:

WARNING ⚠️ The file 'model.pt' appears to be improperly saved or formatted. For optimal results, use model.save('filename.pt') to correctly save YOLO models.
Traceback (most recent call last):
  File "D:\test\loader.py", line 3, in <module>
    model = YOLO("model.pt")
  File "D:\Python310\lib\site-packages\ultralytics\models\yolo\model.py", line 23, in __init__
    super().__init__(model=model, task=task, verbose=verbose)
  File "D:\Python310\lib\site-packages\ultralytics\engine\model.py", line 151, in __init__
    self._load(model, task=task)
  File "D:\Python310\lib\site-packages\ultralytics\engine\model.py", line 240, in _load
    self.model, self.ckpt = attempt_load_one_weight(weights)
  File "D:\Python310\lib\site-packages\ultralytics\nn\tasks.py", line 806, in attempt_load_one_weight
    ckpt, weight = torch_safe_load(weight)  # load ckpt
  File "D:\Python310\lib\site-packages\ultralytics\nn\tasks.py", line 760, in torch_safe_load
    ckpt = {"model": ckpt.model}
  File "D:\Python310\lib\site-packages\torch\nn\modules\module.py", line 1688, in __getattr__
    raise AttributeError(f"'{type(self).__name__}' object has no attribute '{name}'")
AttributeError: 'GraphModule' object has no attribute 'model'. Did you mean: 'code'?

While using code:

from ultralytics import YOLO

model = YOLO("model.pt")
model.export(format="onnx", imgsz=[416,416])

[advaitp]

Did you solve it ? I am also getting the same error while converting from onnx to torch for yolov8 model ?

[senysenyseny16]

Hi @AI-M-BOT, @advaitp

The ArgMax error should be resolved by this release:: https://github.com/ENOT-AutoDL/onnx2torch/releases/tag/v1.5.15

Also, could you provide ONNX files to reproduce the errors?