File "D:\ProgramData\Anaconda3\envs\GPU1\lib\site-packages\torch\nn\modules\module.py", line 1051, in _call_impl
return forward_call(*input, **kwargs)
File "D:\ProgramData\Anaconda3\envs\GPU1\lib\site-packages\torch\nn\modules\container.py", line 139, in forward
input = module(input)
File "D:\ProgramData\Anaconda3\envs\GPU1\lib\site-packages\torch\nn\modules\module.py", line 1051, in _call_impl
return forward_call(*input, **kwargs)
File "D:\ProgramData\Anaconda3\envs\GPU1\lib\site-packages\torch\nn\modules\batchnorm.py", line 167, in forward
return F.batch_norm(
File "D:\ProgramData\Anaconda3\envs\GPU1\lib\site-packages\torch\nn\functional.py", line 2279, in batch_norm
_verify_batch_size(input.size())
File "D:\ProgramData\Anaconda3\envs\GPU1\lib\site-packages\torch\nn\functional.py", line 2247, in _verify_batch_size
raise ValueError("Expected more than 1 value per channel when training, got input size {}".format(size))
ValueError: Expected more than 1 value per channel when training, got input size torch.Size([1, 16, 1, 1])
有哪位兄弟在yolo添加自己注意力或者其他模块的时候遇到过这个问题吗，能交流一下怎么解决的吗。
你debug一下，看是哪里出错了，input size torch.Size([1, 16, 1, 1]) ，前面肯定有些地方做了上面尺度缩减的操作，后面又没有恢复过来，或者少做了些操作。
已解决，原因来自自适应平均池化，因为代码输出[2,x,x,x]造成错误