Pulse · huggingface/lerobot

April 10, 2025 – May 10, 2025

143 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

Port HIL SERL
#644 commented on May 10, 2025 • 81 new comments
feat: adding the audio modality 🎤
#967 commented on May 9, 2025 • 5 new comments
Add support for cameras over ROS 2
#866 commented on May 5, 2025 • 4 new comments
Upgrade `OnlineBuffer` to `LeRobotDatasetV2` and use it in the train script.
#445 commented on Apr 17, 2025 • 0 new comments
Added depth frame support for dataset recording/loading
#455 commented on Apr 17, 2025 • 0 new comments
Asynchronous policy inference with TemporalQueue
#457 commented on Apr 17, 2025 • 0 new comments
Update lerobot visualizer to send videos directly to Rerun
#476 commented on Apr 17, 2025 • 0 new comments
Add transformer model to diffusion policy
#481 commented on Apr 17, 2025 • 0 new comments
Add UFACTORY 850 as one of the robots supported, introduce various HRI utilities for teleop and next episode hinting while reseting
#493 commented on Apr 14, 2025 • 0 new comments
Add support for Elephant Robotics MyArm M&C
#506 commented on Apr 14, 2025 • 0 new comments
`configure_motor.py --keep-id`: allow reconfiguring a motor without changing its id
#507 commented on Apr 17, 2025 • 0 new comments
Enable control of the follower arm with PS4 joystick
#516 commented on Apr 23, 2025 • 0 new comments
Add `docker-compose` support for easier and more portable environment set-up
#519 commented on Apr 17, 2025 • 0 new comments
TD-MPC2
#523 commented on Apr 17, 2025 • 0 new comments
[viz tool] add policy pred column
#529 commented on Apr 17, 2025 • 0 new comments
Add HopeJrRobot
#531 commented on Apr 14, 2025 • 0 new comments
Add manage_dataset script
#543 commented on Apr 17, 2025 • 0 new comments
Add HIT policy
#548 commented on Apr 17, 2025 • 0 new comments
enabling single/multi video feed to the lerobot over lan with ip:port
#551 commented on Apr 17, 2025 • 0 new comments
Port HIL-SERL
#565 commented on Apr 17, 2025 • 0 new comments
some questions about real world env
#432 commented on Apr 11, 2025 • 0 new comments
Add RLBench
#290 commented on Apr 17, 2025 • 0 new comments
Add distributed training with `accelerate`
#317 commented on Apr 29, 2025 • 0 new comments
Adds MultiEpochsDataLoader to optimize first batch loading perfomance
#329 commented on Apr 17, 2025 • 0 new comments
Add support for recording a policy without a leader arm.
#344 commented on Apr 17, 2025 • 0 new comments
WIP : Add possibility for multitask training on ACT policy
#350 commented on Apr 17, 2025 • 0 new comments
Add inference to visualize_dataset_html.py
#353 commented on Apr 17, 2025 • 0 new comments
⚡️add florence+linear action head policy, and adapt it to pusht env
#356 commented on Apr 17, 2025 • 0 new comments
feat(arx): support arx arm
#364 commented on Apr 14, 2025 • 0 new comments
update find_available_ports to use serial.tools.list_ports
#372 commented on Apr 17, 2025 • 0 new comments
Add script to remove episodes from dataset
#378 commented on Apr 17, 2025 • 0 new comments
bug: Fix memory leakage cause by permanent storage of decoded video frame and not closed video stream
#380 commented on Apr 17, 2025 • 0 new comments
optimization: Speed up image saving procedure
#381 commented on Apr 17, 2025 • 0 new comments
Add code to communicate with Hiwonder motors
#391 commented on Apr 17, 2025 • 0 new comments
Add mujoco support for trace recording using Simulation of Dynamixel bus
#392 commented on Apr 17, 2025 • 0 new comments
Move function `make_optimizer_and_scheduler` to policy
#401 commented on Apr 17, 2025 • 0 new comments
Adding aggregated logs for training run
#411 commented on Apr 17, 2025 • 0 new comments
Adds a script for recording training data from a Gym simulation and replaying in sim
#437 commented on Apr 17, 2025 • 0 new comments
Visualize output of forward and select_action
#744 commented on Apr 17, 2025 • 0 new comments
xs: Add error message in homepage of dataset visualizer. (⚡️ UX Improvement)
#756 commented on Apr 17, 2025 • 0 new comments
Port Droid 1.01 dataset with SLURM
#758 commented on Apr 17, 2025 • 0 new comments
feat(policy): Add DexVLA policy [Community Contributed]
#767 commented on Apr 16, 2025 • 0 new comments
docs: update instructions for camera config
#768 commented on Apr 17, 2025 • 0 new comments
Refactor robots
#777 commented on May 8, 2025 • 0 new comments
Added a roarm_m3 for robotic arm adaptation, and modified the keyboard listening method to adapt to the screenless mode
#820 commented on Apr 17, 2025 • 0 new comments
feat: add finetuning analysis scripts
#824 commented on Apr 17, 2025 • 0 new comments
feat: add `remove_episodes` utility
#831 commented on Apr 25, 2025 • 0 new comments
fix scripts of v20 to v21
#834 commented on Apr 17, 2025 • 0 new comments
Enhance pi0 model inference
#872 commented on Apr 14, 2025 • 0 new comments
Fix yaml.github-actions.security.run-shell-injection.run-shell-injection
#873 commented on Apr 17, 2025 • 0 new comments
Fix generic.secrets.security.detected-aws-access-key-id-value.detected-aws-access-key-id-value
#874 commented on Apr 17, 2025 • 0 new comments
Improved Dataset Merge Script for Multiple Dataset Processing
#924 commented on Apr 24, 2025 • 0 new comments
Fix python.flask.security.injection.ssrf-requests.ssrf-requests--tmp-bebe03d0-da57-45d6-90f5-447e4352cd2c-lerobot-scripts-visualize_dataset_html.py
#931 commented on Apr 17, 2025 • 0 new comments
Add Hope Jr
#935 commented on Apr 16, 2025 • 0 new comments
(🐛 Bug) Fix losses unpadding bug
#965 commented on May 9, 2025 • 0 new comments
a refactored modeling_act for cpu and memory optimization
#569 commented on Apr 17, 2025 • 0 new comments
aloha_hd5 to LerobotDataset v2
#586 commented on Apr 17, 2025 • 0 new comments
Refact warmup waiting
#591 commented on Apr 17, 2025 • 0 new comments
Show friendly error when tries to record already existed dataset
#592 commented on Apr 17, 2025 • 0 new comments
[Visualizer] visualize int columns as well
#628 commented on Apr 17, 2025 • 0 new comments
[viz_html] Document in readme
#637 commented on Apr 17, 2025 • 0 new comments
Piper robot integration
#645 commented on Apr 14, 2025 • 0 new comments
Feat/ add scan motors tool for troubleshooting during assembly and calibration
#657 commented on Apr 17, 2025 • 0 new comments
Add port aloha hdf5 datasets
#659 commented on Apr 17, 2025 • 0 new comments
Add DiT policy with flow matching objective
#680 commented on Apr 17, 2025 • 0 new comments
fix: support multi repo datasets when training
#684 commented on Apr 17, 2025 • 0 new comments
Improve pi0
#691 commented on Apr 17, 2025 • 0 new comments
Raspberry pi client and remote inference support
#703 commented on Apr 17, 2025 • 0 new comments
feat: Implement Proximal Policy Optimization (PPO) policy
#704 commented on Apr 17, 2025 • 0 new comments
Optimize pi0 for torch.compile
#729 commented on Apr 17, 2025 • 0 new comments
Adding DOT policy
#739 commented on Apr 17, 2025 • 0 new comments
Add streaming to LeRobotDataset
#740 commented on Apr 17, 2025 • 0 new comments
Install Issues -> Add Devcontainer
#427 commented on Apr 11, 2025 • 0 new comments
Compatible off-the-shelf robots?
#413 commented on Apr 11, 2025 • 0 new comments
Multi-Image support for VQ-BeT
#407 commented on Apr 11, 2025 • 0 new comments
ERROR: Could not build wheels for mujoco, which is required to install pyproject.toml-based projects
#405 commented on Apr 11, 2025 • 0 new comments
Improve the way the video benchmark evaluates the accuracy of video decoders
#367 commented on Apr 11, 2025 • 0 new comments
[Feature request] Is it possible to add images to videos in a stream, and avoid converting them to png first?
#360 commented on Apr 11, 2025 • 0 new comments
rerun always begins playback at 30fps, even when the dataset prescribes a different fps.
#336 commented on Apr 11, 2025 • 0 new comments
where to set the initial joint (position + angle) information when controlling real aloha robot?
#334 commented on Apr 11, 2025 • 0 new comments
High Data Loading Times at the Beginning of Each Epoch with LeRobot Datasets
#324 commented on Apr 11, 2025 • 0 new comments
aloha arm do not move in sim eval env on my own task
#320 commented on Apr 11, 2025 • 0 new comments
Sometimes all actions are pad in the dataset
#315 commented on Apr 11, 2025 • 0 new comments
Add a separate `eval.n_envs` config for setting number of evaluation environments
#297 commented on Apr 11, 2025 • 0 new comments
after training using lerobot framework，how to infer the trained policy directly in real environment(ep. aloha code)? i have not found a solution yet
#294 commented on Apr 11, 2025 • 0 new comments
[internal] Environments run slow on the cluster
#287 commented on Apr 11, 2025 • 0 new comments
Reduce Repository Size by Guiding Users to Download Models from Hugging Face
#285 commented on Apr 11, 2025 • 0 new comments
Question - how to handle additional sensory input
#266 commented on Apr 11, 2025 • 0 new comments
how to use on wslcan not visulize
#238 commented on Apr 11, 2025 • 0 new comments
Removing episodes from LeRobotDataset
#630 commented on Apr 11, 2025 • 0 new comments
Diffusion Policy: Transformer backbone support
#624 commented on Apr 11, 2025 • 0 new comments
`control_robot.py` does not work with Windows
#509 commented on Apr 11, 2025 • 0 new comments
documentation improvment: linux brltty issue
#497 commented on Apr 11, 2025 • 0 new comments
Eval with mps gives bad results
#496 commented on Apr 11, 2025 • 0 new comments
Dynamixel port finding not working for Windows.
#478 commented on Apr 11, 2025 • 0 new comments
NaN values when running eval.py on M3 Mac
#475 commented on Apr 11, 2025 • 0 new comments
tdmpc xarm example not working
#474 commented on Apr 11, 2025 • 0 new comments
Broken link in 4_train_policy_with_script example
#469 commented on Apr 11, 2025 • 0 new comments
[Design Change] Enable Bringing Your Own Control Scheme
#462 commented on Apr 11, 2025 • 0 new comments
The instruction of using DynamixelWizard2 to update firmware in Readme will cause firmware update failure.
#456 commented on Apr 11, 2025 • 0 new comments
Venv isn't needed in docker
#454 commented on Apr 11, 2025 • 0 new comments
[Feature Request] use one pass to compute mean and variance of recorded data
#452 commented on Apr 11, 2025 • 0 new comments
Inquiry about Implementation of "Aloha Unleashed"
#451 commented on Apr 11, 2025 • 0 new comments
fail on calibration
#441 commented on Apr 11, 2025 • 0 new comments
Improve LeRobotDataset
#440 commented on Apr 11, 2025 • 0 new comments
Image storage format
#436 commented on Apr 11, 2025 • 0 new comments
Open-X datasets
#435 commented on Apr 11, 2025 • 0 new comments
pi0-fintune-performance
#960 commented on Apr 27, 2025 • 0 new comments
PI0FAST Evaluation results on LIBERO is very low
#955 commented on Apr 27, 2025 • 0 new comments
Do I need two SO 100 arms to get started?
#568 commented on Apr 27, 2025 • 0 new comments
All image features are missing from the batch
#895 commented on Apr 28, 2025 • 0 new comments
Hello, I would like to ask if I can use my ROS2 MoveIt2 robotic arm?
#851 commented on Apr 28, 2025 • 0 new comments
Low accuracy for diffusion policy+aloha env+sim_transfer_cude_human dataset
#502 commented on May 1, 2025 • 0 new comments
[Question] Dataset v2.0 to v2.1 error
#910 commented on May 2, 2025 • 0 new comments
PI0 Evaluation results on LIBERO is very low
#952 commented on May 7, 2025 • 0 new comments
Collecting human operated datasets in simulation
#477 commented on May 7, 2025 • 0 new comments
So100 follower's 2nd motor moves the opposite direction of the leader
#930 commented on May 9, 2025 • 0 new comments
ValueError: One or several timestamps unexpectedly violate the tolerance inside episode range. This might be due to synchronization issues with timestamps during data collection.
#933 commented on May 10, 2025 • 0 new comments
Log locally with tensorboard when wandb is disabled.
#154 commented on Apr 17, 2025 • 0 new comments
Minimal Port of Octo Model
#209 commented on Apr 17, 2025 • 0 new comments
Port LR Schedulers
#225 commented on Apr 17, 2025 • 0 new comments
Check we don't have too many Nan in conversion dataset
#245 commented on Apr 17, 2025 • 0 new comments
[Experimental] Decouple Experiment Tracking from Logger
#262 commented on Apr 17, 2025 • 0 new comments
Simulate latency v2 and use a policy wrapper to handle latency
#272 commented on Apr 17, 2025 • 0 new comments
Improvement: Update normalization in ACT + remove in-place modification + add test
#281 commented on Apr 17, 2025 • 0 new comments
ACT model returns constant actions
#951 commented on Apr 11, 2025 • 0 new comments
Keyerror: task when evaluating fine-tuned pi0 on local so100 arm
#898 commented on Apr 14, 2025 • 0 new comments
Problem in Validating of Features for Numpy Arrays
#916 commented on Apr 14, 2025 • 0 new comments
gstreamer error
#717 commented on Apr 14, 2025 • 0 new comments
_get_query_indices function gives an error when LeRobotDataset initialized with episodes
#959 commented on Apr 14, 2025 • 0 new comments
Bug: VideoFrame column overwritten by dict in LeRobotDataset.__getitem__
#961 commented on Apr 14, 2025 • 0 new comments
Modify camera async_read/read API to return a dictionary instead of tuple for better compatability?
#860 commented on Apr 14, 2025 • 0 new comments
How to train diffusion policy in only state space, no images?
#854 commented on Apr 15, 2025 • 0 new comments
没有物理机器人我如何进行仿真机器人，我应该如何学习
#968 commented on Apr 15, 2025 • 0 new comments
Improve documentation for resuming training runs
#888 commented on Apr 15, 2025 • 0 new comments
ACT is working, but not Diffusion
#515 commented on Apr 16, 2025 • 0 new comments
The inverse kinematic solution code of so-100
#678 commented on Apr 19, 2025 • 0 new comments
Unable to recognize ports when using a windows computer with 1 usb port
#932 comme 43F2 nted on Apr 22, 2025 • 0 new comments
support for running on Jetson?
#819 commented on Apr 22, 2025 • 0 new comments
Isaac Sim integration with LeRobot
#408 commented on Apr 25, 2025 • 0 new comments
maximum range of [-10, 110] % to account for some imprecision during calibration, but present value is inf %
#607 commented on Apr 25, 2025 • 0 new comments
pi0 multi gps train
#956 commented on Apr 26, 2025 • 0 new comments
[Pi0] The dataset_stats are not provided on the model weights
#699 commented on Apr 27, 2025 • 0 new comments

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

April 10, 2025 – May 10, 2025

Overview

Could not load contribution data

32 Pull requests merged by 16 people

37 Pull requests opened by 29 people

15 Issues closed by 11 people

44 Issues opened by 41 people

143 Unresolved conversations

Insights: huggingface/lerobot

April 10, 2025 – May 10, 2025

Overview

Could not load contribution data

32 Pull requests merged by 16 people

37 Pull requests opened by 29 people

15 Issues closed by 11 people

44 Issues opened by 41 people

143 Unresolved conversations