Mechanical refactor of the GR00T N1.7 policy to match the repo's architecture and
style standards. No change to policy algorithm/numerics; only UX/CLI and packaging
changes. Tests are intentionally left untouched (out of scope) and need updating
for the removed `model_version` field.
Cleanup & consolidation:
- Add `groot/utils.py` holding the pure, side-effect-free helpers (JSON I/O, value
coercion, stat flattening, rot6d/SE3 math, language/batch prep) shared by the
config and processor layers.
- Remove dead code: the unused `resolve_groot_n1_7_backbone_model` cache-resolver
cluster, `GR00TN17Config.to_filtered_dict/json`, and the `_copy_default` wrapper.
Imports & execution guards:
- Hoist nested imports to module top; relative imports within the package, absolute
for external modules. The version-gated Qwen3-VL classes import under the single
`_transformers_available` guard (transformers is pinned >=5.4, which ships them).
- No import-time side effects: `_register_with_transformers()` now runs in
`GR00TN17.__init__` (idempotent via `register(exist_ok=True)`), and the N1.5 step
stubs register lazily before pipeline deserialization (idempotent via the
registry, no run-once globals).
- Gate optional deps at the point of use with `require_package(..., extra="groot")`.
Dependencies & docs:
- Drop `flash-attn` (and its build-only dep `ninja`) from the `groot` extra; default
to SDPA (numerically equivalent) with opt-in via `--policy.use_flash_attention`.
Un-comment `lerobot[groot]` in the `all` extra and regenerate `uv.lock`.
- Rewrite the `groot.mdx` install section: flash-attn is a purely optional,
user-managed optimization that LeRobot neither installs nor requires.
Config & CLI:
- Surface previously-frozen knobs on `GrootConfig` (plumbed into `GR00TN17Config`;
no-ops at their defaults): inference — `num_inference_timesteps`, `rtc_ramp_rate`,
`use_flash_attention`; fine-tuning — `tune_top_llm_layers` (partial-LLM tuning)
and `tune_vlln` (previously hardwired to True).
- Convert the single-valued `model_version` and `n1_7_backbone_model` fields to
internal constants.
- Keep `base_model_path`: it is NOT equivalent to `pretrained_path` (raw NVIDIA
checkpoints have no LeRobot `type` field and load only via `base_model_path`) and
is genuinely user-tunable.
- Keep the deprecated Isaac-GR00T/N1.5 fields (and the dead LoRA fields) as a
back-compat block so a v0.5.1 N1.5 `config.json` still parses under draccus and is
rejected with the friendly N1.5 removal message instead of an opaque decode error.
GR00T training was dataloader-bound (0->100->0 GPU-utilization sawtooth).
GrootN17VLMEncodeStep ran the Qwen3-VL image processor per frame on PIL images
on the single CPU main-loop thread, and that cost is timed inside dataloading_s
(preprocessor(batch) runs in the main process, not the dataloader workers), so
adding workers cannot hide it.
- Feed the torchvision-backed Qwen3-VL processor (C,H,W) uint8 tensors instead
of a per-frame Image.fromarray PIL roundtrip, and run resize/normalize/patchify
on config.device (GPU) when available. Bit-identical on CPU when no resize is
configured; with a resize only the PIL->torchvision bicubic backend differs
(<2/255 per pixel). The use_albumentations path stays PIL/cv2; reload on a box
without the saved device falls back to CPU.
- Default image_target_size/crop to the N1.7 backbone's training geometry
(256x256 / 230x230) when a checkpoint ships no image sizing (checkpoint_assets
is None, e.g. finetuning nvidia/GR00T-N1.7-3B via repo-id with a new
embodiment). Previously image_target_size=None disabled the resize, so
full-resolution frames were patchified into ~4.7x more vision tokens than the
model was trained on -- inflating dataloading_s (patchify) and update_s (VLM
sequence) and skewing the input distribution. Checkpoints that pin their own
sizing are honored; the default constants are shared with GR00T_N1_7_DEFAULTS.
Net: preprocessing leaves the CPU critical path and the VLM sees the resolution
it was trained on -- faster training/inference and a correct train/serve
distribution. Affects inference too (shared preprocessor); existing checkpoints
still load (backward compatible) but must be retrained to gain the benefits.
Covers the GR00T N1.7 source trio (configuration, processor, model wrapper).
Config:
- GrootConfig defaults are the N1.7 values; explicitly passed legacy N1.5-era
values (chunk_size=50, max_state_dim=64, ...) are remapped with a warning
instead of silently.
- action_decode_transform gains an 'auto' sentinel so an explicit 'none'
opt-out wins over the libero_sim default and survives save/load round-trips.
- action_delta_indices is cached on the inputs that determine it.
- Legacy N1.5 checkpoints/configs (tokenizer_assets_repo, model_type/
architectures/eagle backbone markers) are rejected with a single clear
error pointing to lerobot==0.5.1.
Processor:
- GrootN17ActionDecodeStep handles the 2-D (B, D) actions delivered by sync
select_action (relative eef/non-eef decode in eval/record flows).
- Postprocessor falls back to dataset stats when a raw checkpoint lacks the
configured embodiment tag; raw-state cache is per-instance, not
process-global; caller overrides (device, rename_map) are honored on the
raw-checkpoint branch.
- Camera/modality-key mismatches warn (including the zero-match fallback);
deprecated Qwen2VLImageProcessorFast replaced with Qwen2VLImageProcessor;
removed N1.5 processor steps are stubbed to raise the removal guidance and
the action-unpack step is re-registered as _v2.
Model:
- Flash-attention probe is diagnostic-only; forward raises on a missing loss;
print() replaced with logging; N1.5 base-path mismatch includes the
removal guidance.
- select_layer default tracks the N1.7-3B checkpoint value (16); real
checkpoint loads still override it from config.json.
- get_backbone_cls recognizes Cosmos-Reason2 / Qwen3-VL backbones by name and
warns (instead of silently assuming) when an unrecognized backbone is loaded
only on the strength of backbone_model_type='qwen'.
- 'revision' pins the GR00T checkpoint repo only and is no longer forwarded
into the unrelated backbone repo load; pin the backbone via
transformers_loading_kwargs instead.
- DiT / SelfAttentionTransformer parameter counts go through logging.debug
instead of print().
- groot.mdx: breaking-change warning and migration path (pin lerobot==0.5.1 to
keep N1.5, or move to N1.7); the dead `huggingface-cli download` is replaced
with `hf download`.
- policy_groot_README.md: N1.5 removal note, updated paper / model-card links,
and the two-comparison (model parity + preprocessor parity) description of
the original-vs-LeRobot test, including the raw-observation artifacts and
recorded seed.
The test loaded MODEL_PATH='aractingi/bimanual-handover-groot-10k', an N1.5
checkpoint (config base_model_path=nvidia/GR00T-N1.5-3B, no model_version). On
load, model_version defaults to n1.7 while the base path infers n1.5, so the
version-consistency guard in GrootConfig.__post_init__ raised ValueError and both
test_lerobot_groot_inference and test_lerobot_groot_forward_pass failed. N1.5 is no
longer a supported model_version.
Adopt the test for N1.7:
- MODEL_PATH -> nvidia/GR00T-N1.7-3B (root-level sharded safetensors; loads via
GrootPolicy.from_pretrained as a base N1.7 model).
- Embodiment tag 'gr1' (N1.5) -> 'gr1_unified' (valid N1.7 tag from the checkpoint
embodiment_id.json), via a single EMBODIMENT_TAG constant.
- DUMMY_ACTION_HORIZON 16 -> 40 to match N1.7's native action-chunk size.
- Docstrings/labels updated to 'GR00T N1.7'.
Both tests run and pass on CUDA; full tests/policies/groot/ suite is
73 passed / 0 failed / 0 skipped.
Mirror the tests/policies/pi0_pi05/utils convention: move dump_original_n1_7.py into
a tests/policies/groot/utils/ package (with __init__.py) and update all path
references in the test docstring/skip-message and the policy README.
- Rename test_groot_n1_7_vs_original.py -> test_groot_vs_original.py
- Make the test self-contained: producer script (dump_original_n1_7.py) now lives
next to the test; default artifact dir is repo-relative
(tests/policies/groot/artifacts/), overridable via GROOT_N1_7_PARITY_DIR. The
test only reads artifacts and skips if absent -- it never creates external dirs.
- Heavy .npz artifacts (~6-9MB each) are gitignored and regenerated by the producer;
never committed.
- Drop the verbose 'MULTIPLE EMBODIMENTS' docstring block (kept a one-line note).
- Document the parity procedure in the groot policy README (docs/source/policy_groot_README.md).
- Rename test fn test_groot_n1_7_get_action_parity -> test_groot_get_action_parity.
9/9 embodiments still pass (max|diff| < 3e-6, fp32 eps).
Generalize the original-vs-LeRobot N1.7 output-parity test from a single
libero_sim case to every embodiment tag in the checkpoint (libero_sim, oxe_droid,
real_g1, the real_r1_pro_sharpa family, and the xdof family). Inputs are built
generically from checkpoint metadata; the test discovers per-tag .npz artifacts
and runs one parametrized case each, loading the LeRobot model once via a fixture.
All 9 embodiments match the original to fp32 epsilon (max|diff| < 3e-6), confirming
the integration is correct across the model's full embodiment space and not overfit
to libero_sim.
Verifies the LeRobot GR00T N1.7 integration produces equivalent raw
action_pred to NVIDIA Isaac-GR00T for the same checkpoint, inputs, seed,
precision (fp32) and attention kernel (SDPA): max|diff|=8.9e-7 on the
libero_sim embodiment (GR00T-N1.7-LIBERO/libero_10).
The two impls pin incompatible transformers majors (orig 4.57.3 vs
LeRobot 5.x) and cannot share a process, so the original outputs + exact
collated inputs are produced out-of-process and loaded from an .npz. The
test skips on CI / when the checkpoint or artifact are absent.
Addresses checker nit: processor_groot.py docstring still described the N1.5
Eagle VLM path with eagle_content/eagle_* keys that no longer exist in the code.
modeling_groot.py duplicated next(self.parameters()).device twice. LeRobot
ships get_device_from_parameters in policies/utils.py (used by diffusion,
vqbet, tdmpc, gaussian_actor). Reuse it for consistency with the framework.
N1.7 backbone is nvidia/Cosmos-Reason2-2B via Qwen3VLForConditionalGeneration,
not Eagle2 — eagle2_hg_model/ had zero refs outside its own dir.
GR00TN17ActionHead (groot_n1_7.py) re-implements MultiEmbodimentActionEncoder +
CategorySpecificLinear + swish + SinusoidalPositionalEncoding locally, so
flow_matching_action_head.py (N1.5 FlowmatchingActionHead) and its sole
dependency action_encoder.py are dead. Verified: no src/ or tests/ reference.
Removed (~2037 LOC):
- eagle2_hg_model/ (4 files, ~1575 LOC)
- action_head/flow_matching_action_head.py (408 LOC)
- action_head/action_encoder.py (54 LOC)
cross_attention_dit.py KEPT (DiT/AlternateVLDiT/SelfAttentionTransformer live in N1.7).
GR00T N1.7 emits gripper in [0,1] but LIBERO expects [-1,1]. The decode
transform existed but was never auto-enabled for embodiment_tag=libero_sim,
so the policy scored 0% on LIBERO eval. Auto-set it in __post_init__ (still
overridable). LIBERO Spatial eval: 0% -> 98%.
* feat(depth): add depth quantization helpers and tests
* feat(video): add ffv1 to supported codecs
* feat(depth): persist depth metadata
* feat(depth): extend quantization tools to better fit the encoding/decoding pipeline
* feat(depth): plumb DepthEncoderConfig through LeRobotDataset and DatasetWriter
* feat(depth): wire StreamingVideoEncoder + writer to depth encoder
* feat(depth): wire DatasetReader to decode_depth_frames
* feat(cameras/realsense): expose async depth in metric meters
* feat(features): route 2D camera shapes to observation.depth.<key>
* feat(robots/so_follower): emit + populate depth keys when use_depth
* feat(record): plumb DepthEncoderConfig through lerobot-record
* feat(viz): render depth observations as rr.DepthImage in Viridis
* feat(depth maps writer): adding support for raw depth maps recording with image writer
* chore(format): format code
* feat(depth shape): ensuring depth maps shape is always including the channel
* feat(is_depth): simplifying is_depth nested name + legacy support
* fix(stop_event): fixing stop_event race condition in camera classes
* fix(plumbing): fixing missing parts in the depth maps pipeline
* chore(typos): fixing typos
* test(fix): fixing exisiting tests to still work with latest features
* tests(depth): adding new tests for depth integration validation
* feat(pix_fmt channels): use PyAv to check get pixel formats number of channels
* feat(refactor): refactor DepthEncoderConfig quantization pipeline, so that the methods do not live in the config class. Add pixel format - channels validation.Move the default pixel format for depth in the config file.
* fix(pre-commit): fixing mutable defautl value
* fix(info): fixing info metadata update when is_depth_map was set
* tests(typos): fixing typos in tests
* fix(realsense): fixing typo in realsense serial number
* fix(normalization): restricting 255 normalization to non depth/uint8 images only
* fix(typo): fixing typo
* fix(TIFF): add missing quantization and cleanup for TIFF files
* feat(batched dequantization): optimizing dequantize_depth for torch based batched dequantization
* feat(tools): adding depth support in LeRobotDataset edition tools
* test(aggregate): extending aggregation tests to depth frames
* test(cleaning): cleaning up tests
* fix(from_video_info): fixing early validation issue in from_video_info
* fix(typo): fixing typo
* fix(is_depth): adding missing doctrings and is_depth arguments in video decoding functions
Co-authored-by: Wensi (Vince) Ai <59036629+wensi-ai@users.noreply.github.com>
* fix(depth units): fixing depth units output for the realsense cameras
* feat(output unit): adding support for output unit specification at dataset reading/training time
Co-authored-by: Wensi (Vince) Ai <59036629+wensi-ai@users.noreply.github.com>
* test(depth): cleaning up depth tests
* test(depth encoding): updating and cleaning video/depth encoding tests
* chore(format): formatting code
* docs(depth): improving depth maps docs
* test(fix): fixing depth tests
* test(dataset tools): adding missing tests for new dataset edition tools features
* chore(format): formatting code
* fix(pyav check): fixing PyAV option validation for integer codec options by normalizing
numeric values before calling `is_integer()`
Co-authored-by: Wensi (Vince) Ai <59036629+wensi-ai@users.noreply.github.com>
* docs(mermaid): fixing mermaid diagram
* fix(rebase): rebase follow up corrections
* feat(dataset tools): adding missing docstrings and features for depth fill support in dataset edition tools
* docs(docstring): updating docstrings
* docs(dataset tools): updating docs
* fix(save images): fixing image saving in dataset tools
* fix(update video info): fixing update video info logic to match the recording and editing use cases
* test(reencode): fixing reencoding monkeypatch
* fix(review): add Claude review
* chore(format): format code
* fix(update video info): ditching the differentiated approahces for video info update - video info are always updated unless for preserved keys.
* chore(rebase): fixing rebase merge conflicts
* test(visualization): fixing visualization tests
* feat(docstrings): adding explicit docstring for encoding parameters. Docstrigns will now show up as description in the CLI --help.
* feat(mm as default): adding a global DEFAULT_DEPTH_UNIT variable setting mm as default depth unit
* fix(RGB <-> camera): renaming camera_encoder to rgb_encoder for clarity
* chore(TODO): removing deprecated TODO
* doc(write_u16_plane): improving docstrings for write_u16_plane
* feat(units): adding constants for depth frames units (m and mm)
* fix(spam): replacing spamming warning but a debug log
* feat(leagcy metadata): adding automatic metadata update for legacy 'video.is_depth_map' feature
* fix(copy&reindex): fixing metadat reshaping for single channel frames
* fix(ImageNet): excluding dpeth frames from ImageNet stats
* fix(PyAV container seek): fixing initial PyAV container seek to be robust againsy codec choice
* feat(lerobot-dataset-viz): adding support for depth in lerobot-dataset-viz
* fix(compress): removing rerun compression for DepthImages
* fix(signle channel squeeze): fixing single channel squeezing
* chore(format): format code
* fix(streaming): adding support for dequantization in streaming_dataset.py
* refactor(read depth): factorizing depth reading methods for realsense camera and adding support for depth-only usage
* chore(renaming): fixing missed RGBEncoderConfig renamings
* docs(renaming): reflecting renamings in a clearer way in the docs
* chore(annotation): excluding depth from the annotation pipeline
* feat(robots): adding depth support in compatible follower robots
* feat(LeSadKiwi): excluding LeKiwi from depth support (for now)
* chore(fail): removing misplaced file
* chore(fail): removing misplaced file
* fix(remove ffv1): removing ffv1 as it does not support MP4
* docs(cheat sheet): adding depth and video encoding to the cheat sheet
* fix(lossless): tuning depth encoding parameters for lossless depth storage
* test(fix): fixing failing tests
* depth(ZMQ): excluding ZMQ from depth support
* Revert "depth(ZMQ): excluding ZMQ from depth support"
This reverts commit b95cf4e4c2.
* fix(image transforms): excluding depth frames from images transforms
* fix(typo): typo
* fix(stats): fixing stats computation for depth frames
* fix(TIFF vs. pytorch): adding an extra uint16 to float32 conversion for depth maps stored as raw TIFF images
* fix(typos): fixing typos
* test(dtype): fixing stats computation typing tests
---------
Signed-off-by: Steven Palma <imstevenpmwork@ieee.org>
Co-authored-by: Wensi (Vince) Ai <59036629+wensi-ai@users.noreply.github.com>
Co-authored-by: Steven Palma <imstevenpmwork@ieee.org>
Co-authored-by: Wensi Ai <wsai@stanford.edu>
* refactor(training): rename eval_freq to env_eval_freq
- Rename eval_freq to env_eval_freq to distinguish sim environment evaluation from offline loss evaluation.
* feat(training): add inline offline validation with train/eval split
- Add eval_split config for balanced per-task holdout
- Add eval_steps for periodic inline eval loss computation
- Add max_eval_samples to cap eval cost
* fix(datasets): remap absolute indices in __getitem__ for filtered datasets
* fix(train): vectorize eval subset selection for max_eval_samples
* fix(datasets): Move the remapping into EpisodeAwareSampler via absolute_to_relative_idx
* fix(validation): add eval_split range check and eval_steps warning
Validate eval_split is in [0.0, 1.0) to prevent garbage splits from
out-of-range values. Raise when eval_steps > 0 but eval_split is 0.0
since no offline eval will run.
* fix(train): prepare eval dataloader with accelerator for multi-GPU
Prepare eval_dataloader through accelerator.prepare() so eval data is
sharded across ranks instead of duplicated. Reduce eval_loss across
ranks with mean reduction for consistent logging.
* fix(test): rename eval_freq to env_eval_freq for multi-GPU training
Align the MolmoAct2 implementation with lerobot codebase conventions:
- Rename hf_model/ to molmoact2_hf_model/
- Slim config: move all I/O and runtime logic to modeling
- Remove blanket from 8 vendored files, fix 66 lint issues
- Deduplicate _hf_token() and _resolve_checkpoint_location()
- Make huggingface_hub imports lazy
- Remove custom MolmoAct2CosineDecayWithWarmupSchedulerConfig, use base class
- Extract 13 static/classmethods from MolmoAct2Policy to free functions
- Replace print() with logger in vendored action_tokenizer
- Add module docstrings, class docstring, and key method docstrings
- Add module-level loggers to modeling and processor
- Fix docs: pip to uv install, deduplicate README symlink
- Remove shebangs from all files
placo pulls in pin (Pinocchio), whose binary wheels dlopen specific cmeel
sonames (liburdfdom_sensor.so.4.0, libtinyxml2.so.10) but declare only `>=`
floors on their cmeel packages. The 2026-05-21 major bumps (cmeel-urdfdom
6.0.0 -> .so.6, cmeel-tinyxml2 11.0.0 -> .so.11) ship newer sonames, so left
unpinned the resolver grabs them and `import placo` fails at load with
"liburdfdom_sensor.so.4.0: cannot open shared object file".
#3647 capped placo and hardened the kinematics import, but the guard only
defers the failure: constructing RobotKinematics still raises. Pin the cmeel
packages to the 4.x / 10.x ABI the placo/pin wheels are built against (there
is no cmeel-urdfdom 5.x; <5 selects 4.x). Regenerated uv.lock with uv 0.8.0
to match CI; the only resolution change is the two cmeel versions (plus a
deterministic decord platform-marker cascade from 4.0.1's wider wheel set).
Fixes#3755
* feat(eval): record eval rollouts as raw LeRobot datasets
- Record raw env observations inline during rollout(), before
preprocess_observation() transforms them. Uses LeRobotDataset.create()
with add_frame()/save_episode().
- Supports vectorized envs: each env in the batch records independently,
with save_episode() called per env on termination. Each task gets its
own dataset under output_dir/recordings/{task_group}_{task_id}/.
Enabled via --eval.recording=true; disabled by default.
* fix(eval): use FeatureType enum comparison instead of string value
* refactor(eval): per-env datasets recording, no double reset
- Extract _infer_shape_from_obs() to reduce nesting in feature conversion
- Move dataset creation into rollout() using its own env.reset() observation,
eliminating the extra reset in run_one()
- Replace deepcopy with _shallow_copy_obs() for raw observation stashing
- Support batch_size > 1: each parallel env records to its own dataset
(single env skips the env_0/ nesting for simplicity)
- One-time warning for env_features keys missing from observations
- Pass recording_dir + env_features through the call chain instead of
a pre-built recording_dataset object
* refactor(eval): remove shape inference and shallow copy helpers
* feat(eval): optionally push recorded eval datasets to the Hub
* fix(eval): address review comments
- Wrap rollout loop in try/finally so finalize() runs on crash/interrupt
- Guard push_to_hub with num_episodes > 0 to avoid pushing empty datasets
- Hoist loop-invariant multi_env and base_repo_id out of creation loop
* feat(train): FSDP checkpoint saving
* adding docs for FSDP
* adding a test for the fsdp checkpoint path
* cleanup
* fixing final upload to hub
* refactored initial implementation to use torch fsdp api and adding new tests
- Guard ACT's KL divergence computation against None latent params to
prevent crashes during eval when use_vae is set but the forward path
returns no VAE outputs.
- Add offline batch fallback to Diffusion's predict_action_chunk() so
it works with dataloader batches (empty queues) in addition to the
existing online rollout path (populated queues). This enables batched
action prediction for offline evaluation.
- Add pretrained_revision field to PreTrainedConfig (policies) and
RewardModelConfig (reward models), and thread it through make_policy(),
make_pre_post_processors(), and make_reward_model() so that weights and
processor configs can be loaded from a specific Hub commit, branch, or
tag. Defaults to None (latest version, preserving current behavior).
Dataset and env hub loading already supported revision pinning.
Co-authored-by: Steven Palma <imstevenpmwork@ieee.org>
Andy Wrenn