jskumaar/act_ira_lego

Model Card for act

Action Chunking with Transformers (ACT) is an imitation-learning method that predicts short action chunks instead of single steps. It learns from teleoperated data and often achieves high success rates.

This policy has been trained and pushed to the Hub using LeRobot.

Learn how to train and run it in the LeRobot act guide, or browse the full documentation.

---

Model Details

• License: [More Information Needed]
• Robot type: so_follower
• Cameras: wrist_left, desk_view

Inputs & Outputs

The policy consumes these observation features and produces these action features.

Inputs

Feature	Type	Shape
observation.state	STATE	(6,)
observation.images.wrist_left	VISUAL	(3, 480, 640)
observation.images.desk_view	VISUAL	(3, 600, 800)

Outputs

Feature	Type	Shape
action	ACTION	(6,)

Training Dataset

• Repository: Project-IRA/lego
• Episodes: 300
• Frames: 298768
• Frame rate: 30 FPS
• Task(s): "Sort the lego by color", "Put each lego on the plate that's the same color as it", "Pick up the legos and sort them onto the right color plate", "Move all the legos to the plate that matches their color", "Take each lego and set it on the plate with the matching color", "Put red lego on red plate", "Move the red lego brick to the red plate", "Pick up the red lego and put it in the red area", "Sort the red lego bricks onto the red plate", "Take only all red bricks and place them on the red plate", "Put the green lego onto the green plate", "Move the green brick to the green plate", "Pick up the green lego and put it in the green area", "Sort the green lego bricks onto the green plate", "Take only all green bricks and place them on the green plate", "Put the yellow lego brick onto the yellow plate", "Move the yellow lego to the yellow plate", "Pick up the yellow lego and put it into the yellow area", "Sort the yellow lego bricks onto the yellow plate", "Take only all yellow bricks and place them on the yellow plate", "Put the blue lego brick onto the blue plate", "Move the blue legoto the blue plate", "Pick up the blue lego and put it on the blue area", "Sort the blue lego bricks onto the blue plate", "Take only all blue bricks and place them on the blue plate", "Place each lego brick on the plate that matches its color", "Pick up the lego bricks and sort them onto the correct color plate", "Move all lego bricks to their matching colored plate", "Organize the lego bricks by placing each one on its corresponding color plate", "Take each lego brick and put it on the plate with the same color"

Training Configuration

Setting	Value
Training steps	30000
Batch size	16
Optimizer	adamw
Learning rate	1e-05
Seed	1000
LeRobot version	0.6.2

---

How to Get Started with the Model

New to LeRobot? These guides cover the full workflow:

• Install LeRobot — set up the lerobot package.
• Hardware setup — assemble, wire, and calibrate your robot and cameras.
• Record data & train a policy — the end-to-end imitation-learning walkthrough.
• CLI cheat-sheet — quick reference for the lerobot-* commands.

The short version to run and train this policy:

Run the policy on your robot

lerobot-rollout \
  --strategy.type=base \
  --robot.type=so_follower \
  --robot.port=<your_robot_port> \
  --robot.cameras="{ <camera_1>: {type: opencv, index_or_path: <index_or_path>, width: 640, height: 480, fps: 30}, <camera_2>: {type: opencv, index_or_path: <index_or_path>, width: 640, height: 480, fps: 30}}" \
  --policy.path=jskumaar/act_ira_lego \
  --task="Sort the lego by color" \
  --duration=60

Replace the remaining <...> placeholders with your own values: --robot.port and the camera names/indices are specific to your machine, and the camera names must match the observation keys this policy was trained on.

When --strategy.type=base is used the script doesn't record the episodes. Skipping duration will make the policy run indefinitely. For more information look at rollout documentation.

Train your own policy

lerobot-train \
  --dataset.repo_id=${HF_USER}/<dataset> \
  --policy.type=act \
  --output_dir=outputs/train/<policy_repo_id> \
  --job_name=lerobot_training \
  --policy.device=cuda \
  --policy.repo_id=${HF_USER}/<policy_repo_id> \
  --wandb.enable=true

Writes checkpoints to outputs/train/<policy_repo_id>/checkpoints/.

---

Evaluation

No evaluation results have been provided for this policy yet.

---

Citation

If you use this policy, please cite the method linked in the description above, along with LeRobot:

@misc{cadene2024lerobot,
    author = {Cadene, Remi and Alibert, Simon and Soare, Alexander and Gallouedec, Quentin and Zouitine, Adil and Palma, Steven and Kooijmans, Pepijn and Aractingi, Michel and Shukor, Mustafa and Aubakirova, Dana and Russi, Martino and Capuano, Francesco and Pascal, Caroline and Choghari, Jade and Moss, Jess and Wolf, Thomas},
    title = {LeRobot: State-of-the-art Machine Learning for Real-World Robotics in Pytorch},
    howpublished = "\url{https://github.com/huggingface/lerobot}",
    year = {2024}
}