admin/lerobot
1
0
Fork 0
mirror of https://github.com/huggingface/lerobot.git synced 2026-05-20 11:09:59 +00:00
Code Issues Packages Projects Releases Wiki Activity

add to tree

Browse Source
This commit is contained in:
Jade Choghari
2026-03-02 13:42:00 +03:00
parent 3f80a52728
commit d861d97e87
2 changed files with 9 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/source/_toctree.yml
+2
View File
@@ -19,6 +19,8 @@
title: Multi GPU training title: Multi GPU training
- local: peft_training - local: peft_training
title: Training with PEFT (e.g., LoRA) title: Training with PEFT (e.g., LoRA)
- local: rename_map
title: Using Rename Map and Empty Cameras
title: "Tutorials" title: "Tutorials"
- sections: - sections:
- local: lerobot-dataset-v3 - local: lerobot-dataset-v3
docs/source/rename_map.mdx
+7 -7
View File
@@ -110,7 +110,7 @@ LeRobot supports this with **empty cameras**: the config declares extra “slots
</p> </p>
- In the policy config (e.g. [xvla-base config.json](https://huggingface.co/lerobot/xvla-base/blob/main/config.json)), `empty_cameras` is the number of these extra slots (default `0`). - In the policy config (e.g. [xvla-base config.json](https://huggingface.co/lerobot/xvla-base/blob/main/config.json)), `empty_cameras` is the number of these extra slots (default `0`).
- For each slot, the config adds an observation key of the form: - For each slot, the config adds an observation key of the form:
`observation.images.empty_camera_0`, `observation.images.empty_camera_1`, … `observation.images.empty_camera_0`, `observation.images.empty_camera_1`, …
Example: XVLA-base has three visual inputs and `empty_cameras=0`. Your dataset has only two images. Set **`empty_cameras=1`**. Then: Example: XVLA-base has three visual inputs and `empty_cameras=0`. Your dataset has only two images. Set **`empty_cameras=1`**. Then:
@@ -135,11 +135,11 @@ If you see a feature mismatch error (“Missing features” / “Extra features
## Quick reference ## Quick reference
| Goal | What to do | | Goal | What to do |
|------|------------| | ------------------------------------- | ---------------------------------------------------------------------------------------------------------- |
| Dataset keys ≠ policy keys (training) | `--rename_map='{"dataset_key": "policy_key", ...}'` | | Dataset keys ≠ policy keys (training) | `--rename_map='{"dataset_key": "policy_key", ...}'` |
| Env keys ≠ policy keys (eval) | `--rename_map='{"env_key": "policy_key", ...}'` | | Env keys ≠ policy keys (eval) | `--rename_map='{"env_key": "policy_key", ...}'` |
| Fewer cameras than policy expects | Set `empty_cameras` in the policy config (e.g. `1` when you have 2 real cameras and the policy expects 3). | | Fewer cameras than policy expects | Set `empty_cameras` in the policy config (e.g. `1` when you have 2 real cameras and the policy expects 3). |
| Avoid passing a rename map | Edit the policy’s `config.json` so its observation keys match your dataset or env. | | Avoid passing a rename map | Edit the policy’s `config.json` so its observation keys match your dataset or env. |
The rename map keeps your pipeline flexible: one policy, many data sources, no code changes—just a small dictionary on the command line or in your config. The rename map keeps your pipeline flexible: one policy, many data sources, no code changes—just a small dictionary on the command line or in your config.