Testing ACT on complex lego task
Description
Training ACT model on complex lego task to use with LLM Agent. Dataset involves picking up 1x1 block and placing onto 2x2 block.
Dataset
- Repo: airshop/Lego_task_complex
- Image augmentations: Disabled.
Training
- Model / Policy Type: ACT
- Steps: 500
- Batch size: 32
- Vision backbone: resnet18
- VAE enabled: True (latent dim: 32)
- Training time: 0 days, 0 hours, 45 minutes, 54 seconds
- Machine: DGX Spark
Inputs
4 visual streams, 1 state inputs
Results
[Insert results here]
How to Get Started with the Model
For a complete walkthrough, see the training guide.
Evaluate the policy / run inference
lerobot-record \
--robot.type=so100_follower \
--dataset.repo_id=<hf_user>/eval_<dataset> \
--policy.path=<hf_user>/<desired_policy_repo_id> \
--episodes=10
Prefix the dataset repo with eval_ and supply --policy.path pointing to a local or hub checkpoint.
Model Details
This policy has been trained and pushed to the Hub using LeRobot. See the full documentation at LeRobot Docs.
- License: apache-2.0
- Downloads last month
- 13