docs(g05): record real training smoke

This commit is contained in:
Pepijn
2026-07-28 21:43:47 +02:00
parent 0623dae3f6
commit 274ee585b4
+5
View File
@@ -173,6 +173,11 @@ uv run pytest tests/policies/g05 tests/runtime/test_g05_adapter.py -q
uv run ruff check src/lerobot/policies/g05 tests/policies/g05
```
An RTX 5090 smoke test loaded the real private `g05_libero` checkpoint and
completed a batch-size-one BF16 forward, backward, gradient clip, and AdamW
step. It produced finite loss `2.77356`, finite pre-clip gradient norm `54.38`,
all six author optimizer groups, and 23.59 GiB peak allocated CUDA memory.
A 50-episode LIBERO/RoboTwin success-rate comparison additionally requires the
matching simulator, task assets, reset seeds, and author evaluator; no task-level
benchmark number is claimed until that separate gate runs.