Caffe2 / R2Plus1D: "VideoInput is not a registered operator."

After having no luck with various manual build manuals for Caffe2 and OpenCV (within my anaconda python 2.7 environment, Ubuntu 16.04), I followed the advice to install the nightly build binaries via:

conda install pytorch-nightly torchvision -c pytorch

I want to use models from R2Plus1D for extracting action recognition features. However, when running their it will throw the following error message:

Traceback (most recent call last):
  File "lib/", line 333, in <module>
  File "lib/", line 328, in main
  File "lib/", line 131, in ExtractFeatures
  File "/anaconda2/lib/python2.7/site-packages/caffe2/python/", line 34, in Parallelize_GPU
    Parallelize(*args, **kwargs)
  File "/anaconda2/lib/python2.7/site-packages/caffe2/python/", line 219, in Parallelize
  File "lib/", line 106, in input_fn
  File "/models/c3d/R2Plus1D/lib/utils/", line 120, in AddVideoInput
    data, label, video_id =
  File "/anaconda2/lib/python2.7/site-packages/caffe2/python/", line 2171, in __getattr__
    ",".join(workspace.C.nearby_opnames(op_type)) + ']'
AttributeError: Method VideoInput is not a registered operator. Did you mean: []

Some issues mention that there is a simple solution. Does somebody know a solution?

i got the same problem, have you solve this?
please help me,