Help loading a Yolov5 model with toch.hub.load()

vision
1

I am trying to run inference on a model created via Yolov5. I am trying to load the model with weights using torch.hub.load() and am getting an error

Command

model = torch.hub.load('ultralytics/yolov5', 'custom', path='/yolov5/runs/train/exp9/weights/best.pt', force_reload=True)

Error

File ~/.cache/torch/hub/ultralytics_yolov5_master/hubconf.py:33, in _create(name, pretrained, channels, classes, autoshape, verbose, device)
     17 """Creates or loads a YOLOv5 model
     18 
     19 Arguments:
   (...)
     29     YOLOv5 model
     30 """
     31 from pathlib import Path
---> 33 from models.common import AutoShape, DetectMultiBackend
     34 from models.experimental import attempt_load
     35 from models.yolo import ClassificationModel, DetectionModel, SegmentationModel

File ~/.cache/torch/hub/ultralytics_yolov5_master/models/common.py:38
     35     os.system('pip install -U ultralytics')
     36     import ultralytics
---> 38 from ultralytics.utils.plotting import Annotator, colors, save_one_box
     40 from utils import TryExcept
     41 from utils.dataloaders import exif_transpose, letterbox

File /opt/conda/lib/python3.10/site-packages/ultralytics/utils/__init__.py:25
     22 import yaml
     23 from tqdm import tqdm as tqdm_original
---> 25 from ultralytics import __version__
     27 # PyTorch Multi-GPU DDP Constants
     28 RANK = int(os.getenv('RANK', -1))

ImportError: cannot import name '__version__' from 'ultralytics' (unknown location)

I am running in AWS Sagemaker with the Pytorch 2.0.0 Python 3.10 image and here are my versions
torch: 2.0.0
ultralytics 8.0.177

Any advice would be helpful.