我正在研究Google Colab。代码中有一个对Spatialsamplerbillinear函数的调用,该函数假定为spacesamplingbilinear.cu,并且显示未实现的错误
错误:
/content/drive/My Drive/sushanm1/sconv/functional/sconv.py in kernel_resampling_forward(kernel, grid) backend.SpatialGridSamplerBilinear_updateOutput(backend.library_state, kernel, grid, output, 0)
/usr/local/lib/python3.6/dist-packages/torch/_thnn/utils.py in __getattr__(self, name)
25 def register_method(self, name, ctypes_fn):
26 self.methods[name] = ctypes_fn
---> 27
28 @property
29 def library_state(self):
我尝试在colab上设置cuda(),当我在研究实验室的系统上进行设置时,它显示了此功能作为.cu文件的存在 我也尝试升级到pytorch 0.3.1,但仍然显示未实现。我不知道为什么在colab上未调用thcunn实现
代码:
backend = thnn.type2backend[kernel.type()]
backend.SpatialGridSamplerBilinear_updateOutput(backend.library_state,
kernel, grid, output, 0)
return output