File "from_para_mesh.py", line 44, in
return_verts=True, return_full_pose=True,use_pca=False
File "/home/renzhipeng/.conda/envs/venv_frankmocap/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, *kwargs)
File "/home/renzhipeng/.conda/envs/venv_frankmocap/lib/python3.7/site-packages/smplx/body_models.py", line 1204, in forward
'bi,ij->bj', [left_hand_pose,self.left_hand_components])
File "/home/renzhipeng/.conda/envs/venv_frankmocap/lib/python3.7/site-packages/torch/functional.py", line 325, in einsum
return einsum(equation, _operands)
File "/home/renzhipeng/.conda/envs/venv_frankmocap/lib/python3.7/site-packages/torch/functional.py", line 327, in einsum
return _VF.einsum(equation, operands) # type: ignore[attr-defined]
RuntimeError: einsum(): operands do not broadcast with remapped shapes [original->remapped]: [1, 45]->[1, 1, 45] [6, 45]->[1, 45, 6]
As shown above，When my left_hand_pose= torch.reshape(left_hand_pose, (1,45)) and is of the following form
tensor([[ 0.1554, 0.1760, -0.2391, 0.0918, -0.1128, -0.2956, -0.1144, 0.0386,
-0.0735, 0.0495, 0.0957, -0.3257, -0.1233, 0.0142, -0.4398, 0.0431,
0.0061, -0.1982, -0.1915, -0.2914, -0.4117, 0.0348, 0.0443, -0.3285,
-0.2318, 0.0015, 0.1677, 0.0638, -0.0788, -0.3995, -0.2018, 0.0199,
-0.3163, -0.0758, -0.1233, -0.2661, 0.9592, 0.2528, -0.1477, -0.5575,
-0.0189, 0.0124, 0.5102, -0.0274, -0.1530]]
I still get an error after changing the tensor dimension
File "from_para_mesh.py", line 44, in
return_verts=True, return_full_pose=True,use_pca=False
File "/home/renzhipeng/.conda/envs/venv_frankmocap/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, *kwargs)
File "/home/renzhipeng/.conda/envs/venv_frankmocap/lib/python3.7/site-packages/smplx/body_models.py", line 1204, in forward
'bi,ij->bj', [left_hand_pose,self.left_hand_components])
File "/home/renzhipeng/.conda/envs/venv_frankmocap/lib/python3.7/site-packages/torch/functional.py", line 325, in einsum
return einsum(equation, _operands)
File "/home/renzhipeng/.conda/envs/venv_frankmocap/lib/python3.7/site-packages/torch/functional.py", line 327, in einsum
return _VF.einsum(equation, operands) # type: ignore[attr-defined]
RuntimeError: einsum(): operands do not broadcast with remapped shapes [original->remapped]: [1, 45]->[1, 1, 45] [6, 45]->[1, 45, 6]
As shown above，When my left_hand_pose= torch.reshape(left_hand_pose, (1,45)) and is of the following form
tensor([[ 0.1554, 0.1760, -0.2391, 0.0918, -0.1128, -0.2956, -0.1144, 0.0386, -0.0735, 0.0495, 0.0957, -0.3257, -0.1233, 0.0142, -0.4398, 0.0431, 0.0061, -0.1982, -0.1915, -0.2914, -0.4117, 0.0348, 0.0443, -0.3285, -0.2318, 0.0015, 0.1677, 0.0638, -0.0788, -0.3995, -0.2018, 0.0199, -0.3163, -0.0758, -0.1233, -0.2661, 0.9592, 0.2528, -0.1477, -0.5575, -0.0189, 0.0124, 0.5102, -0.0274, -0.1530]] I still get an error after changing the tensor dimension