You can deploy this model and start earning money today!

CursedRock17/
so101_remove_battery_full_act

Name: so101_remove_battery_full_act
Availability: InStock

CursedRock17/so101_remove_battery_full

2304.13705

apache-2.0

Views

No views yet

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. See the full documentation at LeRobot Docs.

How to Get Started with the Model

For a complete walkthrough, see the training guide. Below is the short version on how to train and run inference/eval:

Train from scratch

bash

export HF_USER='CursedRock17'
export HF_DATASET='so101_remove_battery_full'

lerobot-train \
  --policy.type=act \
  --policy.optimizer_lr=1e-05 \
  --policy.n_action_steps=25 \
  --policy.chunk_size=25 \
  --dataset.repo_id=${HF_USER}/${HF_DATASET} \
  --batch_size=8 \
  --steps=100000 \
  --output_dir=outputs/train/${HF_DATASET}_act \
  --job_name=${HF_DATASET}_act \
  --policy.device=cuda \
  --wandb.enable=false \
  --policy.repo_id=${HF_USER}/${HF_DATASET}_act \

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

Evaluate the policy/run inference