训练时backward出错

[WEN466]

您好，最近在训练ocean时，loss.backward处报错：
Traceback (most recent call last):
File "C:\Python39\lib\site-packages\torch\autograd\function.py", line 204, in wrapper
outputs = fn(ctx, *args)
File "D:\work_datas\TracKit-master\tracking..\lib\models\dcn\deform_conv.py", line 85, in backward
deform_conv_cuda.deform_conv_backward_parameters_cuda(
RuntimeError: view size is not compatible with input tensor's size and stride (at least one dimension spans across two contiguous subspaces). Use .reshape(...) instead.

deform_conv_cuda这个好像是编译的库，我自己修改不了。请问这个报错是为什么呢？可以帮忙看一下吗