admin/lerobot
1
0
Fork 0
mirror of https://github.com/huggingface/lerobot.git synced 2026-07-16 06:21:48 +00:00
Code Issues Packages Projects Releases Wiki Activity

feat(lingbot_va): RoboTwin eef-pose eval, single-file model, Hub checkpoints

Browse Source 
Make the LingBot-VA port runnable on both LIBERO and RoboTwin and clean up the
package to LeRobot conventions.

- Consolidate all vendored Wan2.2 model code (transformer, attention, VAE helpers,
  flow-matching scheduler, grid utils, flex-attention) into a single
  modeling_lingbot_va.py; remove the separate wan_*/schedulers modules.
- Move the fixed action (un)normalization quantiles out of the config and into the
  post-processor (LIBERO 7-DoF + RoboTwin 16-d eef); remove the conversion script in
  favour of ready-to-use LeRobot-format checkpoints on the Hub.
- Fixes found via on-sim validation: undo LIBERO's 180-degree image flip
  (image_hflip), encode obs as a multi-frame streaming-VAE clip, reset the streaming
  VAE cache between episodes, run the transformer in config.dtype, lazy-load frozen
  VAE/UMT5 by subfolder with the text encoder on CPU.
- RoboTwin: add an end-effector-pose action mode to RoboTwinEnv (16-d per-arm
  xyz+quat+gripper deltas composed onto the initial eef pose, executed via CuRobo IK)
  and the robotwin_tshape latent layout (full-res head + half-res wrists via a second
  streaming VAE) with the upstream RoboTwin action quantiles + camera mapping.
- Predicted-video saving works for both benchmarks; docs + tests updated.

Co-authored-by: Cursor <cursoragent@cursor.com>
This commit is contained in:
pepijn223
2026-06-06 15:20:51 +02:00
committed by Maxime Ellerbach
parent d600a52943
commit b81909fc28
20 changed files with 2372 additions and 2834 deletions
Download Patch File Download Diff File Expand all files Collapse all files
pyproject.toml
+1 -1
View File
@@ -228,7 +228,7 @@ vla_jepa = ["lerobot[transformers-dep]", "lerobot[diffusers-dep]", "lerobot[qwen
# LingBot-VA needs the Wan2.2 stack (AutoencoderKLWan z_dim=48 + WanTransformer3DModel config schema),
# which only exists in diffusers>=0.36. Pin the floor explicitly so a standalone `lerobot[lingbot_va]`
# install can't resolve to a pre-Wan2.2 diffusers via the looser diffusers-dep floor.
lingbot_va = ["lerobot[transformers-dep]", "diffusers>=0.36.0,<0.37.0", "lerobot[imageio-dep]"]
lingbot_va = ["lerobot[transformers-dep]", "diffusers>=0.36.0,<0.37.0", "lerobot[imageio-dep]", "accelerate>=1.10.0,<2.0.0", "ftfy>=6.0.0,<7.0.0"]
# Features
async = ["lerobot[grpcio-dep]", "lerobot[matplotlib-dep]"]