IDEAS-Lab-Northwestern/pi05-sim-stack-flat-30-libero-lora

pi0.5 LoRA — stack_flat (sim) (30 demos)

Base weights: gs://openpi-assets/checkpoints/pi05_base/params (openpi pi0.5 base — no robot prior) Training config: pi05_stack_flat_libero_lora (openpi, LIBERO-schema) Dataset: IDEAS-Lab-Northwestern/sim-stack-flat-30-libero (private) — OmniGibson sim teleop, 30 episodes, 30 fps, LIBERO v2.1 schema (image + wrist_image + 8D EEF state + 7D EEF-delta action) Real or sim: Sim — OmniGibson teleop on a desk-mounted Franka Panda Prompt (language conditioning):

"Pick up the flat object from under the stack and move it into the green goal sphere. Take care that the items resting on top remain stable and undisturbed."

Training

• 20,000 LoRA steps total — final checkpoint saved as step 19999 (openpi 0-indexed naming)
• Batch size 4 on a single A100-SXM4-40GB
• Wall time: ~2h 45m
• LoRA adapters: paligemma_variant="gemma_2b_lora", action_expert_variant="gemma_300m_lora" (PaliGemma 2B + 300M action expert base frozen; only adapter params updated)
• action_dim=32, action_horizon=16, pi05=True, discrete_state_input=False
• Norm stats: computed from this dataset (not reused — pi05_base ships no robot-specific norm stats). Computed via openpi's scripts/compute_norm_stats.py --config-name pi05_stack_flat_libero_lora

wandb run: https://wandb.ai/yiyanpeng2027-northwestern-university/openpi/runs/tess6u0n

Contents: only step 19999/ (params + assets + _CHECKPOINT_METADATA). train_state/ excluded — not needed for inference or LoRA-on-top resumption.

Use this checkpoint

# Download just this step
HF_HUB_DISABLE_XET=1 huggingface-cli download IDEAS-Lab-Northwestern/pi05-sim-stack-flat-30-libero-lora \
  --include "19999/**" \
  --local-dir vla_models/pi05-sim-stack-flat-30-libero-lora

# Serve with openpi (point at the step dir)
uv run scripts/serve_policy.py \
  --policy.config=pi05_stack_flat_libero_lora \
  --policy.dir=vla_models/pi05-sim-stack-flat-30-libero-lora/19999

Note: the pi05_stack_flat_libero_lora config defaults to its hardcoded repo_id for training. For inference you do not need the dataset — the ckpt is self-contained.

Family

This is one of three paired sim-LoRA fine-tunes evaluating safety awareness in pi0.5-class VLA models on OmniGibson sim teleop:

• This repo: stack_flat (sim)
• Sibling: IDEAS-Lab-Northwestern/pi05-sim-stack-same-30-libero-lora — stack_same (sim)
• Cousin: IDEAS-Lab-Northwestern/pi05-sim-lid-transport-food-30-libero-lora — lid_transport_food (sim)

All three share the same TrainConfig template (LIBERO schema, pi05_base warm-start, identical LoRA + optimizer hyperparams); they differ only in dataset + prompt.

Companion real-teleop pair (DROID schema): pi05-real-cab-60-droid-lora + pi05-real-jar-60-droid-lora.

License

Apache-2.0