# Launch finetuning on your dataset
python lerobot/scripts/train.py \
--policy.path=pravsels/smolvla_transfer_cube \
--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=pravsels/smolvla_transfer_cubesmolvla_transfer_cube
Fine-tuned SmolVLA for bimanual SO101 (12-dim action/state).
| Base model | lerobot/smolvla_base |
| Dataset | villekuosmanen/armnetbench_transfer_cube |
| Checkpoints | 5,000 and 10,000 steps |
| Training | Vast.ai 4xH100, batch 128, ~8.5h total |
| Action dim | 12 (bimanual SO101, 6 per arm) |
| Cameras | top, left_wrist, right_wrist |
| W&B | smolvla_transfer_cube |
Checkpoints
checkpoints/005000/pretrained_model/checkpoints/010000/pretrained_model/
Usage
from lerobot.policies.smolvla.modeling_smolvla import SmolVLAPolicy
policy = SmolVLAPolicy.from_pretrained("pravsels/smolvla_transfer_cube", subfolder="checkpoints/010000/pretrained_model")
Training Logs
Full training curves and metrics: https://wandb.ai/pravsels/smolvla_transfer_cube/runs/w2j060oy
- Downloads last month
- 4
Model tree for pravsels/smolvla_transfer_cube
Base model
lerobot/smolvla_base
# 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]