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

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
parent 4dfa8cea65
commit e3deff00ad
20 changed files with 1452 additions and 1732 deletions
Generated
+38 -5
View File
@@ -1172,10 +1172,11 @@ wheels = [
[[package]]
name = "diffusers"
version = "0.35.2"
version = "0.36.0"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "filelock" },
{ name = "httpx" },
{ name = "huggingface-hub" },
{ name = "importlib-metadata" },
{ name = "numpy" },
@@ -1184,9 +1185,9 @@ dependencies = [
{ name = "requests" },
{ name = "safetensors" },
]
sdist = { url = "https://files.pythonhosted.org/packages/03/68/288ca23c7c05c73e87ffe5efffc282400ac9b017f7a9bb03883f4310ea15/diffusers-0.35.2.tar.gz", hash = "sha256:30ecd552303edfcfe1724573c3918a8462ee3ab4d529bdbd4c0045f763affded", size = 3366711, upload-time = "2025-10-15T04:05:17.213Z" }
sdist = { url = "https://files.pythonhosted.org/packages/88/45/ccb2e2180ddf475a0f931dac6a50346310e4c464ce3cccb8a65d1fc1e16d/diffusers-0.36.0.tar.gz", hash = "sha256:a9cde8721b415bde6a678f2d02abb85396487e1b0e0d2b4abb462d14a9825ab0", size = 3795088, upload-time = "2025-12-08T10:14:34.255Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/2a/2e/38d9824f8c6bb048c5ba21c6d4da54c29c162a46b58b3ef907a360a76d3e/diffusers-0.35.2-py3-none-any.whl", hash = "sha256:d50d5e74fdd6dcf55e5c1d304bc52cc7c2659abd1752740d736d7b54078b4db5", size = 4121649, upload-time = "2025-10-15T04:05:14.391Z" },
{ url = "https://files.pythonhosted.org/packages/35/50/281f92cb1f83854dbd79b6e958b3bc5018607e2542971d41604ba7a14b2f/diffusers-0.36.0-py3-none-any.whl", hash = "sha256:525d42abc74bfc3b2db594999961295c054b48ef40a11724dacf50e6abd1af98", size = 4597884, upload-time = "2025-12-08T10:14:31.979Z" },
]
[[package]]
@@ -1636,6 +1637,18 @@ http = [
{ name = "aiohttp" },
]
[[package]]
name = "ftfy"
version = "6.3.1"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "wcwidth" },
]
sdist = { url = "https://files.pythonhosted.org/packages/a5/d3/8650919bc3c7c6e90ee3fa7fd618bf373cbbe55dff043bd67353dbb20cd8/ftfy-6.3.1.tar.gz", hash = "sha256:9b3c3d90f84fb267fe64d375a07b7f8912d817cf86009ae134aa03e1819506ec", size = 308927, upload-time = "2024-10-26T00:50:35.149Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/ab/6e/81d47999aebc1b155f81eca4477a616a70f238a2549848c38983f3c22a82/ftfy-6.3.1-py3-none-any.whl", hash = "sha256:7c70eb532015cd2f9adb53f101fb6c7945988d023a085d127d1573dc49dd0083", size = 44821, upload-time = "2024-10-26T00:50:33.425Z" },
]
[[package]]
name = "future"
version = "1.0.0"
@@ -2696,6 +2709,7 @@ all = [
{ name = "faker" },
{ name = "fastapi" },
{ name = "feetech-servo-sdk" },
{ name = "ftfy" },
{ name = "grpcio" },
{ name = "grpcio-tools" },
{ name = "gym-aloha" },
@@ -2704,6 +2718,7 @@ all = [
{ name = "hebi-py" },
{ name = "hf-libero", marker = "sys_platform == 'linux'" },
{ name = "hidapi" },
{ name = "imageio", extra = ["ffmpeg"] },
{ name = "ipykernel" },
{ name = "jsonlines" },
{ name = "jupyter" },
@@ -2877,6 +2892,9 @@ hopejr = [
{ name = "pygame" },
{ name = "pyserial" },
]
imageio-dep = [
{ name = "imageio", extra = ["ffmpeg"] },
]
intelrealsense = [
{ name = "pyrealsense2", marker = "sys_platform != 'darwin'" },
{ name = "pyrealsense2-macosx", marker = "sys_platform == 'darwin'" },
@@ -2901,6 +2919,13 @@ libero = [
{ name = "torchcodec", marker = "(platform_machine == 'arm64' and sys_platform == 'darwin') or (platform_machine == 'AMD64' and sys_platform == 'linux') or (platform_machine == 'aarch64' and sys_platform == 'linux') or (platform_machine == 'arm64' and sys_platform == 'linux') or (platform_machine == 'x86_64' and sys_platform == 'linux') or sys_platform == 'win32'" },
{ name = "transformers" },
]
lingbot-va = [
{ name = "accelerate" },
{ name = "diffusers" },
{ name = "ftfy" },
{ name = "imageio", extra = ["ffmpeg"] },
{ name = "transformers" },
]
matplotlib-dep = [
{ name = "contourpy" },
{ name = "matplotlib" },
@@ -3070,6 +3095,7 @@ xvla = [
[package.metadata]
requires-dist = [
{ name = "accelerate", marker = "extra == 'lingbot-va'", specifier = ">=1.10.0,<2.0.0" },
{ name = "accelerate", marker = "extra == 'smolvla'", specifier = ">=1.7.0,<2.0.0" },
{ name = "accelerate", marker = "extra == 'training'", specifier = ">=1.10.0,<2.0.0" },
{ name = "av", marker = "extra == 'av-dep'", specifier = ">=15.0.0,<16.0.0" },
@@ -3079,7 +3105,8 @@ requires-dist = [
{ name = "debugpy", marker = "extra == 'dev'", specifier = ">=1.8.1,<1.9.0" },
{ name = "decord", marker = "(platform_machine == 'AMD64' and extra == 'groot') or (platform_machine == 'x86_64' and extra == 'groot')", specifier = ">=0.6.0,<1.0.0" },
{ name = "deepdiff", marker = "extra == 'deepdiff-dep'", specifier = ">=7.0.1,<9.0.0" },
{ name = "diffusers", marker = "extra == 'diffusers-dep'", specifier = ">=0.27.2,<0.36.0" },
{ name = "diffusers", marker = "extra == 'diffusers-dep'", specifier = ">=0.27.2,<0.37.0" },
{ name = "diffusers", marker = "extra == 'lingbot-va'", specifier = ">=0.36.0,<0.37.0" },
{ name = "dm-tree", marker = "extra == 'groot'", specifier = ">=0.1.8,<1.0.0" },
{ name = "draccus", specifier = "==0.10.0" },
{ name = "dynamixel-sdk", marker = "extra == 'dynamixel'", specifier = ">=3.7.31,<3.9.0" },
@@ -3088,6 +3115,7 @@ requires-dist = [
{ name = "fastapi", marker = "extra == 'phone'", specifier = "<1.0" },
{ name = "feetech-servo-sdk", marker = "extra == 'feetech'", specifier = ">=1.0.0,<2.0.0" },
{ name = "flash-attn", marker = "sys_platform != 'darwin' and extra == 'groot'", specifier = ">=2.5.9,<3.0.0" },
{ name = "ftfy", marker = "extra == 'lingbot-va'", specifier = ">=6.0.0,<7.0.0" },
{ name = "grpcio", marker = "extra == 'grpcio-dep'", specifier = "==1.73.1" },
{ name = "grpcio-tools", marker = "extra == 'dev'", specifier = "==1.73.1" },
{ name = "gym-aloha", marker = "extra == 'aloha'", specifier = ">=0.1.2,<0.2.0" },
@@ -3098,6 +3126,7 @@ requires-dist = [
{ name = "hf-libero", marker = "sys_platform == 'linux' and extra == 'libero'", specifier = ">=0.1.3,<0.2.0" },
{ name = "hidapi", marker = "extra == 'gamepad'", specifier = ">=0.14.0,<0.15.0" },
{ name = "huggingface-hub", specifier = ">=1.0.0,<2.0.0" },
{ name = "imageio", extras = ["ffmpeg"], marker = "extra == 'imageio-dep'", specifier = ">=2.34.0,<3.0.0" },
{ name = "ipykernel", marker = "extra == 'notebook'", specifier = ">=6.0.0,<7.0.0" },
{ name = "jsonlines", marker = "extra == 'dataset'", specifier = ">=4.0.0,<5.0.0" },
{ name = "jupyter", marker = "extra == 'notebook'", specifier = ">=1.0.0,<2.0.0" },
@@ -3128,6 +3157,7 @@ requires-dist = [
{ name = "lerobot", extras = ["diffusers-dep"], marker = "extra == 'vla-jepa'" },
{ name = "lerobot", extras = ["diffusion"], marker = "extra == 'all'" },
{ name = "lerobot", extras = ["dynamixel"], marker = "extra == 'all'" },
{ name = "lerobot", extras = ["eo1"], marker = "extra == 'all'" },
{ name = "lerobot", extras = ["feetech"], marker = "extra == 'all'" },
{ name = "lerobot", extras = ["feetech"], marker = "extra == 'hopejr'" },
{ name = "lerobot", extras = ["feetech"], marker = "extra == 'lekiwi'" },
@@ -3139,10 +3169,12 @@ requires-dist = [
{ name = "lerobot", extras = ["hardware"], marker = "extra == 'core-scripts'" },
{ name = "lerobot", extras = ["hilserl"], marker = "extra == 'all'" },
{ name = "lerobot", extras = ["hopejr"], marker = "extra == 'all'" },
{ name = "lerobot", extras = ["imageio-dep"], marker = "extra == 'lingbot-va'" },
{ name = "lerobot", extras = ["intelrealsense"], marker = "extra == 'all'" },
{ name = "lerobot", extras = ["kinematics"], marker = "extra == 'all'" },
{ name = "lerobot", extras = ["lekiwi"], marker = "extra == 'all'" },
{ name = "lerobot", extras = ["libero"], marker = "sys_platform == 'linux' and extra == 'all'" },
{ name = "lerobot", extras = ["lingbot-va"], marker = "extra == 'all'" },
{ name = "lerobot", extras = ["matplotlib-dep"], marker = "extra == 'async'" },
{ name = "lerobot", extras = ["matplotlib-dep"], marker = "extra == 'sarm'" },
{ name = "lerobot", extras = ["matplotlib-dep"], marker = "extra == 'unitree-g1'" },
@@ -3199,6 +3231,7 @@ requires-dist = [
{ name = "lerobot", extras = ["transformers-dep"], marker = "extra == 'groot'" },
{ name = "lerobot", extras = ["transformers-dep"], marker = "extra == 'hilserl'" },
{ name = "lerobot", extras = ["transformers-dep"], marker = "extra == 'libero'" },
{ name = "lerobot", extras = ["transformers-dep"], marker = "extra == 'lingbot-va'" },
{ name = "lerobot", extras = ["transformers-dep"], marker = "extra == 'molmoact2'" },
{ name = "lerobot", extras = ["transformers-dep"], marker = "extra == 'multi-task-dit'" },
{ name = "lerobot", extras = ["transformers-dep"], marker = "extra == 'peft'" },
@@ -3276,7 +3309,7 @@ requires-dist = [
{ name = "transformers", marker = "extra == 'transformers-dep'", specifier = ">=5.4.0,<5.6.0" },
{ name = "wandb", marker = "extra == 'training'", specifier = ">=0.24.0,<0.25.0" },
]
provides-extras = ["dataset", "training", "hardware", "viz", "core-scripts", "evaluation", "dataset-viz", "av-dep", "pygame-dep", "placo-dep", "transformers-dep", "grpcio-dep", "can-dep", "peft-dep", "scipy-dep", "diffusers-dep", "qwen-vl-utils-dep", "matplotlib-dep", "pyserial-dep", "deepdiff-dep", "pynput-dep", "pyzmq-dep", "motorbridge-dep", "motorbridge-smart-servo-dep", "feetech", "dynamixel", "damiao", "robstride", "openarms", "gamepad", "hopejr", "lekiwi", "unitree-g1", "reachy2", "rebot", "kinematics", "intelrealsense", "phone", "diffusion", "wallx", "pi", "molmoact2", "smolvla", "multi-task-dit", "groot", "sarm", "robometer", "topreward", "xvla", "eo1", "hilserl", "vla-jepa", "async", "peft", "dev", "notebook", "test", "video-benchmark", "aloha", "pusht", "libero", "metaworld", "all"]
provides-extras = ["dataset", "training", "hardware", "viz", "core-scripts", "evaluation", "dataset-viz", "av-dep", "pygame-dep", "placo-dep", "transformers-dep", "grpcio-dep", "can-dep", "peft-dep", "scipy-dep", "diffusers-dep", "imageio-dep", "qwen-vl-utils-dep", "matplotlib-dep", "pyserial-dep", "deepdiff-dep", "pynput-dep", "pyzmq-dep", "motorbridge-dep", "motorbridge-smart-servo-dep", "feetech", "dynamixel", "damiao", "robstride", "openarms", "gamepad", "hopejr", "lekiwi", "unitree-g1", "reachy2", "rebot", "kinematics", "intelrealsense", "phone", "diffusion", "wallx", "pi", "molmoact2", "smolvla", "multi-task-dit", "groot", "sarm", "robometer", "topreward", "xvla", "eo1", "hilserl", "vla-jepa", "lingbot-va", "async", "peft", "dev", "notebook", "test", "video-benchmark", "aloha", "pusht", "libero", "metaworld", "all"]
[[package]]
name = "librt"