pi05_droid_jointpos),
adapted to ManiSkill's PickCube-v1 task as part of the ExpoFT
reproduction project.pd_joint_delta_pos), matching pi0.5's DROID
pretraining distributiongs://openpi-assets/checkpoints/pi05_droid_jointpos/assets/droid)PickCube-v1 hides the goal marker from sensor cameras by
default. This checkpoint was trained and evaluated with a monkeypatch
(expo_ft/env/patches.py in the training repo below) that unhides it —
without applying the same patch, the environment will not expose the goal
position the policy was trained to rely on, and evaluation results will not
be representative.openpi training) and has
not been converted to PyTorch. openpi's official conversion script
(examples/convert_jax_model_to_pytorch.py) exists, but PyTorch-side LoRA
support is currently absent in openpi, and unresolved upstream issues
report output mismatches between JAX and converted PyTorch weights for
pi0.5 specifically -- so no PyTorch conversion is provided here.pi05_weight_loader_path at the downloaded params/ directory when
building the model via openpi's CheckpointWeightLoader, using the
expo_pi05_droid_lora_finetune_sft_joint_state training config as the base.