save ply from verts and faces

[nairouzshehata]

❓ Questions on how to use PyTorch3D

meshv is a torch Tensor of num of verts by 3
meshf is torch.long of num of faces by 3
save_obj(filename, meshv, meshf) works perfectly but save_ply(filename, meshv, meshf) produces the below error:

pytorch3d/io/ply_io.py in save_ply(f, verts, faces, verts_normals, ascii, decimal_places, path_manager)
   1157         path_manager = PathManager()
   1158     with _open_file(f, path_manager, "wb") as f:
-> 1159         _save_ply(
   1160             f,
   1161             verts=verts,

in _save_ply(f, verts, faces, verts_normals, verts_colors, ascii, decimal_places)
   1096                 # rows are 13 bytes: a one-byte 3 followed by three four-byte face indices.
   1097                 faces_uints = np.full((len(faces_array), 13), 3, dtype=np.uint8)
-> 1098                 faces_uints[:, 1:] = faces_array.astype(np.uint32).view(np.uint8)
   1099                 if isinstance(f, BytesIO):
   1100                     f.write(faces_uints.tobytes())

ValueError: could not broadcast input array from shape (3000,3) into shape (750,12)

NOTE: Please look at the existing list of Issues tagged with the label 'question](https://github.com/facebookresearch/pytorch3d/issues?q=label%3Aquestion) or ['how-to. Only open a new issue if you cannot find an answer there.

Also note the following:

1. If you encountered any errors or unexpected issues while using PyTorch3D and need help resolving them,
   please use the "Bugs / Unexpected behaviors" issue template.

2. We do not answer general machine learning / computer vision questions that are not specific to
   PyTorch3D, such as how a model works or what algorithm/methods can be
   used to achieve X.

[bottler]

Thanks for reporting this. I think I can see the problem. We are mistakenly assuming that faces is a contiguous tensor. As a workaround, I think you should be able to use save_ply(filename, meshv, meshf.contiguous()).