sunmoon2018

Results 4 comments of sunmoon2018

dataset format versioning

I also encountered the same problem, where the dataset version 2.0 or 3.0 cannot be directly used, but version 2.1 has been tested and is acceptable.

Droid Dataset Finetuning action dimension mismatch

> > [@AsDeadAsADodo](https://github.com/AsDeadAsADodo) The pi-zero models is trained with a maximum action dimension of 32, which exceeds the actual DoF of most single-arm robots and it actually contains a lot...

error occurs when run remote inference

Change your command like this and try again: uv run scripts/serve_policy.py policy:checkpoint --policy.config=pi05_libero --policy.dir=gs://openpi-assets/checkpoints/pi05_libero

When the training reaches 1000 steps, the training program automatically stops and exits

I have found the cause, it is due to insufficient computer memory, and the problem has been resolved.