Views
No views yet
nut_thread)1# Requires openpi: https://github.com/Physical-Intelligence/openpi
2huggingface-cli download bhe1004/FIRe-VLA-PI05-NutThread-Real --local-dir ./FIRe-VLA-PI05-NutThread-Real1# Point the openpi trained-policy loader at the downloaded checkpoint dir.
2# The repo holds params/ + assets/ (norm stats); train_state/ was dropped
3# since it is only needed to resume training.
4from openpi.policies import policy_config
5policy = policy_config.create_trained_policy(config, "./FIRe-VLA-PI05-NutThread-Real")bhe1004/FIRe-VLA-PI05-NutThread-Real