c398a146b3
use constants for indexing into batches and remove env state references
2025-12-11 09:13:38 -08:00
9b47c5fac9
move policy to top of file
2025-12-11 09:04:38 -08:00
56dbeed89f
add processor tests to multitask dit tests
2025-12-11 09:04:23 -08:00
67b1a9eeb1
update typo in test instruction comment
2025-12-11 08:28:35 -08:00
86e0ee787d
remove environment state conditioning
2025-12-11 08:26:21 -08:00
ba968e84f1
fix bugs when testing on hardware
2025-12-10 16:26:57 -08:00
d49d3390f6
Merge branch 'main' into feature/add-multitask-dit
2025-12-10 14:44:04 -08:00
f1ac454800
add tutorial to training with multi_task_dit
2025-12-10 14:43:27 -08:00
10cfc17705
remove redundant asserts
2025-12-10 14:43:27 -08:00
5524a0d7a7
split up select action return statement
2025-12-10 14:43:27 -08:00
3a16a002f8
add torch.no_grad decorators
2025-12-10 14:43:27 -08:00
3b2a4f548c
merge all modules files into the main modeling file
2025-12-10 14:43:27 -08:00
b92dc82ddd
add references to the modeling file comments
2025-12-10 14:43:27 -08:00
103230c64c
simplify config for multitask dit by merging and flattening everything, then adding comments to denote where some parameters are only used for specific objectives
2025-12-10 14:43:27 -08:00
cdacc090cd
update docstring for multitask dit policy processor file
2025-12-10 14:43:27 -08:00
6f856016c5
adjust factory comment
2025-12-10 14:43:27 -08:00
adabb37af6
remove dino vision encoder and simplify text and vision encoders by removing inheritance structure
2025-12-10 14:43:27 -08:00
55e19ff9a7
update readme and citations for multitask dit policy
2025-12-10 14:43:27 -08:00
ce348a3460
enable variable image sizes to pi0/pi0.5 ( #2609 )
...
* enable variable image sizes to pi0/pi0.5
* add square image assertion
2025-12-10 19:41:11 +01:00
22714af08d
Merge branch 'main' into feature/add-multitask-dit
2025-12-09 20:45:59 -08:00
cb920235c4
docs: update X-VLA training strategies/commands ( #2611 )
2025-12-09 19:08:09 +01:00
46ebcc2f7d
add RoPE attention module as this is shown to help training dynamics and generation quality for DiTs
2025-12-09 08:42:56 -08:00
a0d5a088e3
Merge branch 'main' into feature/add-multitask-dit
2025-12-09 07:44:05 -08:00
7f40b3bf82
feat(dataset): add tool to convert images to video datasets ( #2560 )
...
* add video encoding tool
* style
* make it work
* more fixes
2025-12-08 18:50:21 +01:00
2e9c9fd832
Replay while loop in sample actions with for loops ( #2600 )
2025-12-08 14:47:54 +01:00
f9cb5e659c
chore(ci): skip workflows if not lerobot repository ( #2601 )
...
Co-authored-by: Alex Tyshka <atyshka15@gmail.com >
2025-12-08 12:44:36 +01:00
0217e1e3ad
Fix dataset aggreagation for multi video datasets' ( #2550 )
2025-12-05 16:09:25 +01:00
d79dd6d31f
Add a documentation page with a brief intro to hw backends ( #2385 )
2025-12-05 13:32:58 +01:00
56b43cc888
fix(scripts): missing so101 import ( #2577 )
...
* fix(scripts): missing so101 import
Co-authored-by: Skyler <skylerwiernik@gmail.com >
* fix(scripts): move urdf to cli args
* refactor(scripts): improve find_joints_limits
---------
Co-authored-by: Skyler <skylerwiernik@gmail.com >
2025-12-03 18:20:26 +01:00
77fe5a09ed
fix(docs): argument typo ( #2361 )
...
Co-authored-by: Steven Palma <imstevenpmwork@ieee.org >
2025-12-03 17:57:18 +01:00
89ae7813a7
Reorganize assembly instructions setup before assembly ( #2333 )
...
Motors should be set up before the arm is assembled.
Moving the entire motor setup section before the part cleaning and assembly section.
Signed-off-by: Austin King <shout@ozten.com >
Co-authored-by: Steven Palma <imstevenpmwork@ieee.org >
2025-12-03 17:56:58 +01:00
e003108cf8
Fix link to lerobot-train script in documentation ( #2466 )
...
* Fix link to lerobot-train script in documentation
Signed-off-by: ./c² <cagataycali@icloud.com >
* Update link to lerobot record script
Signed-off-by: ./c² <cagataycali@icloud.com >
---------
Signed-off-by: ./c² <cagataycali@icloud.com >
Co-authored-by: Steven Palma <imstevenpmwork@ieee.org >
2025-12-03 15:46:26 +01:00
5766eea377
fix(docs): remove duplicated package in install instructions ( #2573 )
2025-12-03 15:45:56 +01:00
f8a4cf225b
feat(robots): add earth rover robot support ( #2575 )
...
Co-authored-by: somthecoder <sbaner64@gmail.com >
Co-authored-by: randomSmarts <Aarshsmittal@gmail.com >
Co-authored-by: Hassoonu <halsae2@illinois.edu >
Co-authored-by: Saketh06 <saketh.kantipudi@gmail.com >
Co-authored-by: sairajshetye <sairajshetye2@gmail.com >
Co-authored-by: Khalil Meftah <kmeftah.khalil@gmail.com >
2025-12-03 15:36:22 +01:00
43b0f17eb9
feat(policies): Add X-VLA ( #2405 )
...
* first commit
* more fixes
* add franka action
* update testing script
* add changes
* update files
* logits matching
* add imagenet as a norm type
* logits matching atol1e-2
* more eval fixes
* more changes
* xvla works on libero
* remove seed
* more refactoring
* more fixes
* more changes
* more changes
* more fixes
* migrate policy revert
* major pre-commit cleanup
* renaming
* revert to self.transformer
* refactor
* new changes
* clean
* update libero
* more changes
* make it work
* more changes:
* remove imagenet dependency
* style
* more
* more refactor
* remove proprio
* add loss
* more
* more
* add freeze/unfreeze options
* add testing
* upgrade transformers version
* update testing
* add installation
* remove .sh file
* fix testing
* silent linter in xvlatest
* fix failing test
* upgrade test, fix failing
* fix testing
* more fixes to testing
* require cuda in tests
* temp check
* add xvla docs
* fix styling
* update libero doc
* remove timm dep
* add different dtype support
* remove timm skip
* remove white lines
* Enhance X-VLA finetuning documentation with optimizer details (#2537 )
Added detailed instructions for implementing a custom optimizer and modifying parameter retrieval for X-VLA finetuning.
Signed-off-by: Jinliang Zheng <54488861+2toinf@users.noreply.github.com >
* fix style
* iterate on review
* iterate on cpilot
* revert xvla dep
* free up ci
* test(xvla): remove main test (#2565 )
* Add xvla custom optim and dtype (#2567 )
* add custom optim
* add custom optim
* add auto mode
* more changes
* add identity to all
* add auto
* release
* add docs
* make image smaller docs
* smaller image in doc
* evan smaller image doc
* finalize doc
---------
Signed-off-by: Jinliang Zheng <54488861+2toinf@users.noreply.github.com >
Signed-off-by: Steven Palma <imstevenpmwork@ieee.org >
Co-authored-by: Jinliang Zheng <54488861+2toinf@users.noreply.github.com >
Co-authored-by: Michel Aractingi <michel.aractingi@huggingface.co >
Co-authored-by: Steven Palma <imstevenpmwork@ieee.org >
2025-12-03 15:29:14 +01:00
b0b755471b
Revert "Earth Rover Mini Plus integration ( #2544 )" ( #2574 )
...
This reverts commit 35c5a27352
2025-12-03 14:43:07 +01:00
35c5a27352
Earth Rover Mini Plus integration ( #2544 )
...
* feat: Add EarthRover Mini Plus robot integration with Frodobots SDK
* refactor: Clean up
* refactor: Remove VirtualCamera implementation for EarthRover Mini Plus integration
* fix: Reduce timeout for camera requests
* fix: Add empty cameras dict for compatibility with recording script
* refactor: Remove record.py script for EarthRover Mini Plus use lerobot_record instead
* refactor: Update documentation for EarthRover Mini Plus integration
* refactor keyboard teleoperation
* refactor: Remove angular velocity
* docs: Add documentation for EarthRover Mini Plus integration
* Add earthrover_mini_plus robot to replay and teleoperate scripts
* refactor: Update stop key from Space to X
* refactor: Implement caching for camera frames and robot telemetry data
* refactor
* refactor: Replace string literals with constants for action and observation keys
* Add Earth Rover Mini to robots section in documentation
Co-authored-by: somthecoder sbaner64@gmail.com
Co-authored-by: randomSmarts Aarshsmittal@gmail.com
Co-authored-by: Hassoonu halsae2@illinois.edu
Co-authored-by: Saketh06 saketh.kantipudi@gmail.com
Co-authored-by: sairajshetye sairajshetye2@gmail.com
2025-12-03 14:24:57 +01:00
afb90e17e7
doc: fix wrong package name in installation doc ( #2513 )
2025-12-03 13:36:59 +01:00
9ec9ee781a
feat(policies): Allow users to register 3rd party policies - pip install lerobot_policy_mypolicy ( #2308 )
...
* feat: Register external policies
* ruff fix
* move policy util functions to policy factory
* refactor register_third_party_devices -> register_third_party_plugins
* feat: Update docs with bring your own policies
* Improve docs for new policies
* fix: Inconsistent quotation marks
* fix: Remove print statement
* fix: wrong base class name in documentation
* fix: Handle better how the models are parsed
* fix: precommit passing
* Update docs/source/bring_your_own_policies.mdx
Co-authored-by: Steven Palma <imstevenpmwork@ieee.org >
Signed-off-by: Daniel San José Pro <42489409+danielsanjosepro@users.noreply.github.com >
---------
Signed-off-by: Steven Palma <imstevenpmwork@ieee.org >
Signed-off-by: Daniel San José Pro <42489409+danielsanjosepro@users.noreply.github.com >
Co-authored-by: Steven Palma <imstevenpmwork@ieee.org >
2025-12-03 12:09:24 +01:00
0b497fc37d
Make transport module Mypy Compliant [issue#1731] ( #2433 )
...
* latest
* Delete =3.0.0
Signed-off-by: Md. Muhaimin Rahman <sezan92@gmail.com >
* Update src/lerobot/transport/utils.py
Signed-off-by: Md. Muhaimin Rahman <sezan92@gmail.com >
---------
Signed-off-by: Md. Muhaimin Rahman <sezan92@gmail.com >
Co-authored-by: Steven Palma <imstevenpmwork@ieee.org >
2025-12-02 22:12:15 +01:00
797cd2725a
fix pi05 forward compile ( #2551 )
2025-12-02 11:01:43 +01:00
af4766b602
fix(ci): move hub artifacts to /mnt to avoid runners' No space left on device ( #2564 )
...
* fix(ci): move hub & lerobot artefacts to /mnt to avoid No space left on device in the future
* chore(ci): remove dh -h steps
2025-12-01 20:14:51 +01:00
37f43df88a
Feat/add unitree g1 robot ( #2530 )
...
* add unitree_g1_robot_class
* finish locomotion loading code
* precommit
* separate groot locomotion logic
* remove leftover locomotion variable, unify kp kd
* format config
* properly comment config, example locomotion and unitree_g1 class
* ready to review
* download policy from the hub in `examples/unitree_g1/gr00t_locomotion`
* fix linter
* make precommit happy, add ignore flags
* linter pt3
* linter pt4
* [done] make precommit happy
* fix linter 5
* add docs
* push utils
* feat(robots): add Unitree G1 humanoid support with ZMQ bridge (#2539 )
* feat(robots): add Unitree G1 humanoid support with ZMQ bridge
- Use JSON + base64 serialization for secure communication instead of pickle
- Add documentation section
- Rename robot_server to run_g1_server
- Add dependecies to pyproject.toml
* nit in docs
* remove globals use
* cast robot data to int/float
* ensure robot is connected before changing mode
* temperature can be list, average in such case
---------
Co-authored-by: Martino Russi <nopyeps@gmail.com >
* style nit
* remove transform_imu_data
* remove scipy dependency
* modify toml, add external unitree_sdk2py dep
* return actions from send_action
* cleaning
* add instructions for local deployment
* Update src/lerobot/robots/unitree_g1/unitree_g1.py
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
Signed-off-by: Martino Russi <77496684+nepyope@users.noreply.github.com >
* update config and readme
* update docs
* update docs
* remove torch import
* fix docs
* remove ip from docs
* add licence header
---------
Signed-off-by: Martino Russi <77496684+nepyope@users.noreply.github.com >
Co-authored-by: Michel Aractingi <michel.aractingi@huggingface.co >
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
2025-12-01 16:10:13 +01:00
5f7b5f2817
remove the sampler cause the relative index is added ( #2521 )
...
Co-authored-by: Michel Aractingi <michel.aractingi@huggingface.co >
2025-11-30 22:28:32 +01:00
34499cbc1b
Merge branch 'main' into feature/add-multitask-dit
2025-11-28 16:45:07 -08:00
c55fbe1b3e
chore(dependencies): Bump lerobot to 0.4.3 ( #2540 )
2025-11-28 10:39:02 +01:00
58f70b6bd3
fix(scripts): better prints teleop ( #2538 )
2025-11-27 16:54:17 +01:00
b07160eb1b
feat(utils): precise_sleep() less CPU hungry without sacrificing accuracy ( #2526 )
2025-11-26 17:42:16 +01:00
648ea8f485
fix(benchmark) : fixing video benchmark ( #2094 )
...
* fix(time benchmark): removing deprecated TimeBenchmark dependency
* fix(typo): renaming frames in an up-to-date fashion
* feat(duets): rearanging crf and g parameters in a proper unique combination manner
* fix(segfault): fixing segfault by adding a lock in ThreadPoolExecutor
* chore(update) : update datasets, codecs and backends to the latest versions
* chore(unused files): removing unused files
* fix(dataset paths): fix datasets paths to live among lerobot datasets
2025-11-26 17:41:31 +01:00
581dd45eae
feat(parallel encoding): making parallel encoding the default choice over all platforms ( #2525 )
2025-11-26 14:57:34 +01:00
Bryson Jones