Yolov5: torch.nn.modules.module.ModuleAttributeError: 'BatchNorm2d' object has no attribute '_non_persistent_buffers_set'

[123456789mojtaba]

Hi when I write
python train.py --img 640 --batch 2 --epochs 300 --data ./data/coco128.yaml --cfg ./models/yolov5s.yaml

I have no problem and can train model bu when I add weights to initialize it like buttom I have an error

python train.py --img 640 --batch 2 --epochs 300 --data ./data/coco128.yaml --cfg ./models/yolov5s.yaml --weights yolov5s.pt --noautoanchor

this is my error

train(hyp, tb_writer, opt, device)

File "train.py", line 127, in train
ckpt['model'] = {k: v for k, v in ckpt['model'].float().state_dict().items()
File "/home/mojtaba/virtualenvironment/project_1/lib/python3.6/site-packages/torch/nn/modules/module.py", line 900, in state_dict
module.state_dict(destination, prefix + name + '.', keep_vars=keep_vars)
File "/home/mojtaba/virtualenvironment/project_1/lib/python3.6/site-packages/torch/nn/modules/module.py", line 900, in state_dict
module.state_dict(destination, prefix + name + '.', keep_vars=keep_vars)
File "/home/mojtaba/virtualenvironment/project_1/lib/python3.6/site-packages/torch/nn/modules/module.py", line 900, in state_dict
module.state_dict(destination, prefix + name + '.', keep_vars=keep_vars)
[Previous line repeated 1 more time]
File "/home/mojtaba/virtualenvironment/project_1/lib/python3.6/site-packages/torch/nn/modules/module.py", line 897, in state_dict
self._save_to_state_dict(destination, prefix, keep_vars)
File "/home/mojtaba/virtualenvironment/project_1/lib/python3.6/site-packages/torch/nn/modules/module.py", line 860, in _save_to_state_dict
if buf is not None and name not in self._non_persistent_buffers_set:
File "/home/mojtaba/virtualenvironment/project_1/lib/python3.6/site-packages/torch/nn/modules/module.py", line 772, in __getattr__
type(self).__name__, name))
torch.nn.modules.module.ModuleAttributeError: 'BatchNorm2d' object has no attribute '_non_persistent_buffers_set'

what should I do?I have this problem in uubntu18 . in windows no problem.

[glenn-jocher]

@123456789mojtaba _non_persistent_buffers_set is a new module attribute that pytorch 1.6 requires. If you are seeing this message this means that you are attempting to load a model trained with an earlier version of pytorch. You should start with a blank slate: delete your older models and reclone a fresh copy of yolov5, and everything will work correctly.

NOTE: The YOLOv5 autodownload functionality will only download new models if it does not find a local copy. It does not check versions, so even if your models are out of date it will not automatically replace them.

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.