KeyError: 'model.1.bn'

[qiaoyukeji]

剪枝时报错，KeyError: 'model.1.bn'，请问怎样解决呢

Test after prune ... 
YOLOv5   torch 1.10.0+cu102 CUDA:0 (NVIDIA GeForce RTX 2080 Ti, 11264MiB)

Suggested Gamma threshold should be less than 0.0703.
The corresponding prune ratio is 0.666, but you can set higher.
Gamma value that less than 0.0001 are set to zero!
==============================================================================================
|       layer name               |         origin channels     |         remaining channels  |
|       model.0.bn               |         16                  |         15                  |
|       model.1.conv.1           |         16                  |         15                  |
|       model.1.conv.5           |         16                  |         16                  |
|       model.2.conv.1           |         72                  |         72                  |
|       model.2.conv.4           |         72                  |         72                  |
|       model.2.conv.8           |         24                  |         24                  |
|       model.3.conv.1           |         88                  |         88                  |
|       model.3.conv.4           |         88                  |         88                  |
|       model.3.conv.8           |         24                  |         24                  |
|       model.4.conv.1           |         96                  |         68                  |
|       model.4.conv.4           |         96                  |         68                  |
|       model.4.conv.8           |         40                  |         40                  |
|       model.5.conv.1           |         240                 |         68                  |
|       model.5.conv.4           |         240                 |         68                  |
|       model.5.conv.8           |         40                  |         40                  |
|       model.6.conv.1           |         240                 |         42                  |
|       model.6.conv.4           |         240                 |         43                  |
|       model.6.conv.8           |         40                  |         40                  |
|       model.7.conv.1           |         120                 |         91                  |
|       model.7.conv.4           |         120                 |         91                  |
|       model.7.conv.8           |         48                  |         48                  |
|       model.8.conv.1           |         144                 |         68                  |
|       model.8.conv.4           |         144                 |         68                  |
|       model.8.conv.8           |         48                  |         48                  |
|       model.9.conv.1           |         288                 |         62                  |
|       model.9.conv.4           |         288                 |         62                  |
|       model.9.conv.8           |         96                  |         96                  |
|       model.10.conv.1          |         576                 |         68                  |
|       model.10.conv.4          |         576                 |         68                  |
|       model.10.conv.8          |         96                  |         96                  |
|       model.11.conv.1          |         576                 |         54                  |
|       model.11.conv.4          |         576                 |         54                  |
|       model.11.conv.8          |         96                  |         96                  |
|       model.12.bn              |         256                 |         236                 |
|       model.15.cv1.bn          |         128                 |         124                 |
|       model.15.cv2.bn          |         128                 |         109                 |
|       model.15.cv3.bn          |         256                 |         197                 |
|       model.15.m.0.cv1.bn      |         128                 |         127                 |
|       model.15.m.0.cv2.bn      |         128                 |         119                 |
|       model.16.bn              |         128                 |         128                 |
|       model.19.cv1.bn          |         64                  |         64                  |
|       model.19.cv2.bn          |         64                  |         64                  |
|       model.19.cv3.bn          |         128                 |         128                 |
|       model.19.m.0.cv1.bn      |         64                  |         64                  |
|       model.19.m.0.cv2.bn      |         64                  |         64                  |
    run_prune(**vars(opt))
  File "C:\Adata\soft\soft_install\anaconda\envs\deepsort\lib\site-packages\torch\autograd\grad_mode.py", line 28, in decorate_context
    return func(*args, **kwargs)
  File "prune.py", line 487, in run_prune
    pruned_model = ModelPruned(maskbndict=maskbndict, cfg=pruned_yaml, ch=3).cuda()
  File "C:\Adata\project\deeplearn\yolo\yolov5_6\yolov5_60\yolov5prune\models\yolo.py", line 263, in __init__
    self.model, self.save, self.from_to_map = parse_pruned_model(self.maskbndict, deepcopy(self.yaml), ch=[ch])  # model, savelist
  File "C:\Adata\project\deeplearn\yolo\yolov5_6\yolov5_60\yolov5prune\models\yolo.py", line 478, in parse_pruned_model
    bnc = int(maskbndict[named_m_bn].sum())
KeyError: 'model.1.bn'

[qiaoyukeji]

yolov5 的主干网络换成了 MobileNetV3 ，然后报错的

[huochen1]

yolov5 的主干网络换成了 MobileNetV3 ，然后报错的

请问解决了吗

[xinxin342]

我出现这个错误，你解决了吗？ Traceback (most recent call last): File "D:\ProgramData\Anaconda3\envs\yolov5\yolov5prune-6.0\prune.py", line 806, in main(opt) File "D:\ProgramData\Anaconda3\envs\yolov5\yolov5prune-6.0\prune.py", line 779, in main run_prune(*vars(opt)) File "D:\ProgramData\Anaconda3\envs\yolov5\lib\site-packages\torch\autograd\grad_mode.py", line 28, in decorate_context return func(args, **kwargs) File "D:\ProgramData\Anaconda3\envs\yolov5\yolov5prune-6.0\prune.py", line 537, in run_prune out_idx = np.squeeze(np.argwhere(np.asarray(maskbndict[layername[:-4] + "bn"].cpu().numpy()))) KeyError: 'model.0.act.depth_bn'

[huochen1]

收到，谢谢

[XsCai-sjtu]

请问你解决了嘛

[huochen1]

收到，谢谢

[gudugududu]

yolov5 的主干网络换成了 MobileNetV3 ，然后报错的

请问解决了吗

[huochen1]

收到，谢谢

[kkkkkid]

请问您是如何解决的呀

[huochen1]

收到，谢谢

[holdon1997]

请问楼主解决了？

[huochen1]

收到，谢谢