Pretrained R(2+1)-D models

[fmthoker]

Hi,
Thanks for releasing your code.
I want to include your paper in my experimental study about video representation learning. As you are aware that it is computationally expensive to pre-train on a huge dataset like Kinetics given the meager university resources. I would really appreciate it if you can share your R(2+1)D-18 model pretrained on Kinetics-400.