failed to run python scripts/data_process/fit_smpl_motion.py to produce retargetted motion file about g1/h1

[dbdxnuliba]

Hello,thanks for your greate work to train g1/h1
When I try to learn produce retargetted motion file about g1/h1 following #77 (comment) , I was failed .

Here is the detail:
(isaac) rob@rob:~/rl/PHC$ python scripts/data_process/fit_smpl_motion.py robot=[unitree_g1_fitting,unitree_h1_fitting] +fit_all=True
2024-11-10 13:12:17,796 - INFO - logger - logger initialized
2024-11-10 13:12:18,715 - DEBUG - Setting JobRuntime:name=UNKNOWN_NAME
2024-11-10 13:12:18,715 - DEBUG - Setting JobRuntime:name=fit_smpl_motion
Error executing job with overrides: ['robot=[unitree_g1_fitting,unitree_h1_fitting]', '+fit_all=True']
Traceback (most recent call last):
File "scripts/data_process/fit_smpl_motion.py", line 200, in main
jobs= [jobs[i:i + chunk] for i in range(0, len(jobs), chunk)]
ValueError: range() arg 3 must not be zero

Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace
and another try the command,following has also been failed
(isaac) rob @rob:~/rl/PHC$ python scripts/data_process/fit_smpl_motion.py robot=[unitree_g1_fitting, unitree_h1_fitting, etc] +fit_all=True
2024-11-10 13:14:47,051 - INFO - logger - logger initialized
2024-11-10 13:14:47,893 - DEBUG - Setting JobRuntime:name=UNKNOWN_NAME
2024-11-10 13:14:47,893 - DEBUG - Setting JobRuntime:name=fit_smpl_motion
no viable alternative at input '[unitree_g1_fitting,'
See https://hydra.cc/docs/1.2/advanced/override_grammar/basic for details

Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.

Could you please tell me the full command to retarget g1/h1, many thanks

[IlanElfen]

The issue is that in line 188 of fit_smpl_motion.py, you didn't change the directory of the AMASS Dataset.

amass_root = "/hdd/zen/data/ActBound/AMASS/AMASS_Complete",

so the number of chunks is 0 and the number of jobs becomes 0 (creating an error)
You have to change the amass_root to the directory of your unzipped AMASS dataset.

[dbdxnuliba]

The issue is that in line 188 of fit_smpl_motion.py, you didn't change the directory of the AMASS Dataset.

amass_root = "/hdd/zen/data/ActBound/AMASS/AMASS_Complete",

so the number of chunks is 0 and the number of jobs becomes 0 (creating an error) You have to change the amass_root to the directory of your unzipped AMASS dataset.

ok，I will have a try ,thanks