admin/lerobot
1
0
Fork 0
mirror of https://github.com/huggingface/lerobot.git synced 2026-07-07 18:11:50 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix EVO1 LIBERO eval action postprocessing

Browse Source
This commit is contained in:
javadcc_mac
2026-06-13 10:18:34 +08:00
parent f9b8f297b4
commit fa984990c0
4 changed files with 82 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/source/evo1.mdx
+7 -3
View File
@@ -159,14 +159,18 @@ pixel embeddings, VLM fused tokens, normalized actions, and denormalized actions
(`max_abs_diff=0.0`).
The published checkpoint expects the raw LIBERO camera feature names
`observation.images.agentview_image` and `observation.images.robot0_eye_in_hand_image`. To run the converted
checkpoint with LeRobot LIBERO evaluation for the same one-episode-per-task setting, keep those camera names
instead of the default `image`/`image2` mapping:
`observation.images.agentview_image` and `observation.images.robot0_eye_in_hand_image`. The official EVO1 LIBERO
rollout protocol also replans every 14 actions and binarizes the gripper command before stepping the simulator.
The LIBERO environment postprocessor applies the gripper binarization automatically for EVO1 policies. To run the
converted checkpoint with LeRobot LIBERO evaluation for the same one-episode-per-task setting, keep the raw camera
names instead of the default `image`/`image2` mapping and override `policy.n_action_steps` to 14:
```bash
lerobot-eval \
--policy.path=javadcc/evo1-libero-lerobot \
--policy.vlm_model_name=OpenGVLab/InternVL3-1B \
--policy.device=cuda \
--policy.n_action_steps=14 \
--env.type=libero \
--env.task=libero_object \
--env.camera_name_mapping="{agentview_image: agentview_image, robot0_eye_in_hand_image: robot0_eye_in_hand_image}" \