admin/lerobot
1
0
Fork 0
mirror of https://github.com/huggingface/lerobot.git synced 2026-07-21 17:01:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

docs(groot): drop WHY TWO ENVIRONMENTS block from parity test docstring

Browse Source
This commit is contained in:
nv-sachdevkartik
2026-06-12 08:06:33 +00:00
parent 45e273b806
commit bc4d0db8f4
1 changed files with 0 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/policies/groot/test_groot_vs_original.py
-9
View File
@@ -23,15 +23,6 @@ normalized flow-matching prediction before any action decoding) as NVIDIA's orig
flow-matching seed. The comparison is parametrized over every embodiment tag present flow-matching seed. The comparison is parametrized over every embodiment tag present
in the checkpoint. in the checkpoint.
WHY TWO ENVIRONMENTS
--------------------
The original ``gr00t`` package pins ``transformers==4.57.3`` (Python 3.10) and its
model-config dataclasses subclass ``PretrainedConfig``. Under the transformers 5.x
that the LeRobot GR00T N1.7 integration requires, ``PretrainedConfig`` is itself a
defaulted dataclass, so the original config classes fail to import ("non-default
argument follows default argument"). The two implementations therefore CANNOT be
imported in the same Python process.
To keep the comparison fair, the original outputs + the exact collated inputs are To keep the comparison fair, the original outputs + the exact collated inputs are
produced once per embodiment in the original ``gr00t`` env via the companion script produced once per embodiment in the original ``gr00t`` env via the companion script
``dump_original_n1_7.py`` (next to this file) and saved to per-tag ``.npz`` files. ``dump_original_n1_7.py`` (next to this file) and saved to per-tag ``.npz`` files.