500_chess_moves_smolvla_100k

SmolVLA policy trained on 500 chess moves dataset for 100k steps

Model Details

• Checkpoint: 100000
• Architecture: SmolVLA Policy
• Observation Steps: 1
• Action Steps: 50
• Training Dataset: dopaul/500_chess_moves

Usage

from lerobot.common.policies.factory import make_policy

# Load the policy
policy = make_policy.from_pretrained("dopaul/500_chess_moves_smolvla_100k")

# Use for inference
action = policy.select_action(observation)

Training Details

This model was trained using the LeRobot framework and uploaded from checkpoint 100000.

For more information about LeRobot, visit: https://github.com/huggingface/lerobot