Problem with vid_demo.py with cpu

[wz7475]

device: intel x86_64; os: ubuntu 20.04; gpu: none

python tools/vid_demo.py -f exps/yolov++/v++_SwinTiny_decoupleReg.py -c weights/v++_swinTiny.pth  --path car-detection.mp4 --conf 0.25 --nms 0.5 --tsize 576 --save_result results --device cpu 


Traceback (most recent call last):
  File "tools/vid_demo.py", line 301, in <module>
    main(exp, args)
  File "tools/vid_demo.py", line 292, in main
    imageflow_demo(predictor, vis_folder, current_time, args,exp)
  File "tools/vid_demo.py", line 206, in imageflow_demo
    outputs.extend(predictor.inference(ele,lframe=lframe,gframe=gframe))
  File "/root/yolov/tools/val_to_imdb.py", line 147, in inference
    outputs,outputs_ori = self.model(img,lframe=lframe,gframe=gframe)
  File "/root/miniforge3/envs/yolov/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl
    return forward_call(*input, **kwargs)
  File "/root/yolov/yolox/models/yolov_plus.py", line 21, in forward
    fpn_outs = self.backbone(x)
  File "/root/miniforge3/envs/yolov/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl
    return forward_call(*input, **kwargs)
  File "/root/yolov/yolox/models/yolo_pafpn.py", line 215, in forward
    out_features = self.backbone(input)
  File "/root/miniforge3/envs/yolov/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl
    return forward_call(*input, **kwargs)
  File "/root/yolov/yolox/models/swin_transfomer.py", line 608, in forward
    x = self.patch_embed(x)
  File "/root/miniforge3/envs/yolov/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl
    return forward_call(*input, **kwargs)
  File "/root/yolov/yolox/models/swin_transfomer.py", line 448, in forward
    x = self.proj(x)  # B C Wh Ww
  File "/root/miniforge3/envs/yolov/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl
    return forward_call(*input, **kwargs)
  File "/root/miniforge3/envs/yolov/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 463, in forward
    return self._conv_forward(input, self.weight, self.bias)
  File "/root/miniforge3/envs/yolov/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 460, in _conv_forward
    self.padding, self.dilation, self.groups)
RuntimeError: Input type (float) and bias type (c10::Half) should be the same

[KozlovKY]

same error occurred