Hi,
Thank you for this great work!
I'm struggling to run the retargeting pipeline with my own robot.
In grad_fit_shape.py, the h1_joint_names seems describe link names (the names end in _link), not joint names. I've noticed that your number of joints is the same as the number of links, but mine are different (74 links, 34 joints), which I believe is causing errors.
When I input my link names with the correct ROTATION_AXIS (of which many of the arrays are [0, 0, 0] since there are many more links than joints), grad_fit_shape.py works.
However, when running grad_fit.py, I get this error:
RuntimeError: The size of tensor a (74) must match the size of tensor b (34) at non-singleton dimension 2
If I change the h1_joint_names to describe the 34 joints instead of the 74 links, I get this error when running grad_fit_shape.py:
IndexError: index 0 is out of bounds for dimension 2 with size 0```
Does this mean that the number of joints must be identical? Thanks!
IlanElfen
commented
2 weeks ago
Hi, Thank you for this great work! I'm struggling to run the retargeting pipeline with my own robot. In grad_fit_shape.py, the h1_joint_names seems describe link names (the names end in _link), not joint names. I've noticed that your number of joints is the same as the number of links, but mine are different (74 links, 34 joints), which I believe is causing errors. When I input my link names with the correct ROTATION_AXIS (of which many of the arrays are [0, 0, 0] since there are many more links than joints), grad_fit_shape.py works. However, when running grad_fit.py, I get this error:
RuntimeError: The size of tensor a (74) must match the size of tensor b (34) at non-singleton dimension 2If I change the h1_joint_names to describe the 34 joints instead of the 74 links, I get this error when running grad_fit_shape.py: