Datasets:

---

kowyo

/

pick-and-place

like 0

2026-03-25 14:47:34,825 INFO: Done.
Traceback (most recent call last):
File "/home/showlab/Users/zhijun/RLinf/toolkits/eval_scripts_openpi/zifeng_real_eval.py", line 911, in
main(args)
File "/home/showlab/Users/zhijun/RLinf/toolkits/eval_scripts_openpi/zifeng_real_eval.py", line 662, in main
action_chunk = policy.infer(observation)["actions"]
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/showlab/Users/zhijun/openpi/src/openpi/policies/policy.py", line 71, in infer
inputs = self._input_transform(inputs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/showlab/Users/zhijun/openpi/src/openpi/transforms.py", line 70, in call
data = transform(data)
^^^^^^^^^^^^^^^
File "/home/showlab/Users/zhijun/openpi/src/openpi/transforms.py", line 130, in call
return apply_tree(
^^^^^^^^^^^
File "/home/showlab/Users/zhijun/openpi/src/openpi/transforms.py", line 420, in apply_tree
return unflatten_dict({k: transform(k, v) for k, v in tree.items()})
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/showlab/Users/zhijun/openpi/src/openpi/transforms.py", line 420, in
return unflatten_dict({k: transform(k, v) for k, v in tree.items()})
^^^^^^^^^^^^^^^
File "/home/showlab/Users/zhijun/openpi/src/openpi/transforms.py", line 412, in transform
return fn(v, selector[k])
^^^^^^^^^^^^^^^^^^
File "/home/showlab/Users/zhijun/openpi/src/openpi/transforms.py", line 145, in _normalize_quantile
return (x - q01) / (q99 - q01 + 1e-6) * 2.0 - 1.0
~~^~~~~
ValueError: operands could not be broadcast together with shapes (32,) (8,)