zwan1003/pickplace_skills_v3_2
Viewer • Updated • 239k • 187
How to use zwan1003/pickplace_skills_vla_v3_2 with LeRobot:
# See https://github.com/huggingface/lerobot?tab=readme-ov-file#installation for more details git clone https://github.com/huggingface/lerobot.git cd lerobot pip install -e .[smolvla]
# Launch finetuning on your dataset python lerobot/scripts/train.py \ --policy.path=zwan1003/pickplace_skills_vla_v3_2 \ --dataset.repo_id=lerobot/svla_so101_pickplace \ --batch_size=64 \ --steps=20000 \ --output_dir=outputs/train/my_smolvla \ --job_name=my_smolvla_training \ --policy.device=cuda \ --wandb.enable=true
# Run the policy using the record function
python -m lerobot.record \
--robot.type=so101_follower \
--robot.port=/dev/ttyACM0 \ # <- Use your port
--robot.id=my_blue_follower_arm \ # <- Use your robot id
--robot.cameras="{ front: {type: opencv, index_or_path: 8, width: 640, height: 480, fps: 30}}" \ # <- Use your cameras
--dataset.single_task="Grasp a lego block and put it in the bin." \ # <- Use the same task description you used in your dataset recording
--dataset.repo_id=HF_USER/dataset_name \ # <- This will be the dataset name on HF Hub
--dataset.episode_time_s=50 \
--dataset.num_episodes=10 \
--policy.path=zwan1003/pickplace_skills_vla_v3_2The final checkpoint of the training run. This is not the reported model.
Performance peaked at 60,000 steps and declined afterwards as the model began to overfit. For
the model behind the reported results, use
zwan1003/pickplace_skills_vla_v3_2_ckpt060k.
This repository is kept so that the comparison between the two checkpoints can be reproduced. See the reported model's card for task description, usage and limitations.