No Attribute 'recompute_scale_factor' on Detect/Train #5499
Description
Search before asking
- I have searched the YOLOv5 issues and found no similar bug report.
YOLOv5 Component
Training, Detection
Bug
Attempted to detect or train on Ubuntu 20.04 with all requirements installed and nightly pytorch installed for my gpu/cuda version.
I have rebuilt multiple venvs with fresh clone of YOLOv5 repo and packages reinstalled. I continue to get this error on device 0 or cpu:
(.trainTEST) user:/mnt/nvm/data/yolov5$ python3 detect.py --source /mnt/nvm/data/yolov5/data/photos_BUCKETS-MODELT
detect: weights=yolov5s.pt, source=/mnt/nvm/data/yolov5/data/photos imgsz=[640, 640], conf_thres=0.25, iou_thres=0.45, max_det=1000, device=, view_img=False, save_txt=False, save_conf=False, save_crop=False, nosave=False, classes=None, agnostic_nms=False, augment=False, visualize=False, update=False, project=runs/detect, name=exp, exist_ok=False, line_thickness=3, hide_labels=False, hide_conf=False, half=False, dnn=False
YOLOv5 🚀 v6.0-45-g042f02f torch 1.11.0.dev20211029+cu113 CUDA:0 (NVIDIA GeForce RTX 3080 Ti, 12051.3125MB)
Fusing layers...
Model Summary: 213 layers, 7225885 parameters, 0 gradients
Traceback (most recent call last):
File "detect.py", line 306, in <module>
main(opt)
File "detect.py", line 301, in main
run(**vars(opt))
File "/mnt/nvm/data/yolov5/.trainTEST/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 28, in decorate_context
return func(*args, **kwargs)
File "detect.py", line 131, in run
model(torch.zeros(1, 3, *imgsz).to(device).type_as(next(model.parameters()))) # run once
File "/mnt/nvm/data/yolov5/.trainTEST/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
File "/mnt/nvm/data/yolov5/models/yolo.py", line 127, in forward
return self._forward_once(x, profile, visualize) # single-scale inference, train
File "/mnt/nvm/data/yolov5/models/yolo.py", line 150, in _forward_once
x = m(x) # run
File "/mnt/nvm/data/yolov5/.trainTEST/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
File "/mnt/nvm/data/yolov5/.trainTEST/lib/python3.8/site-packages/torch/nn/modules/upsampling.py", line 154, in forward
recompute_scale_factor=self.recompute_scale_factor)
File "/mnt/nvm/data/yolov5/.trainTEST/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1185, in __getattr__
raise AttributeError("'{}' object has no attribute '{}'".format(
AttributeError: 'Upsample' object has no attribute 'recompute_scale_factor'
(.trainTEST) user:/mnt/nvm/data/yolov5$ pip uninstall torch
Found existing installation: torch 1.11.0.dev20211029+cu113I used this configuration on Friday with no issues. I did a YOLO rep pull over there weekend and it now displays this error.
Any help or assistance is greatly appreciated!
Environment
YOLOv5 🚀 v6.0-45-g042f02f torch 1.11.0.dev20211029+cu113 CUDA:0 (NVIDIA GeForce RTX 3080 Ti, 12051.3125MB) Ubunte 20.04 Python 3.8.10
Minimal Reproducible Example
python3 detect.py --source
Additional
No response
Are you willing to submit a PR?
- Yes I'd like to help by submitting a PR!