# pi0.5 LoRA — cab task (real teleop, 63 demos)

**Base weights**: `gs://openpi-assets/checkpoints/pi05_droid/params` (openpi pi0.5 DROID checkpoint)
**Training config**: `pi05_droid_finetune_lora` (openpi)
**Dataset**: [`IDEAS-Lab-Northwestern/real-cab-60-droid`](https://huggingface.co/datasets/IDEAS-Lab-Northwestern/real-cab-60-droid) — real Franka teleop, 63 episodes / 53,635 frames / 15 fps, DROID schema (LeRobot v2.1)
**Real or sim**: **Real** teleop on a Franka Panda
**Prompt (language conditioning)**:
> "Open the cabinet, place the wooden cube inside, and close it. Keep the area around the door clear so it can swing freely on the way in and out."

## 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 44m
- 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`
- Norm stats: reused from official DROID (`gs://openpi-assets/checkpoints/pi05_droid/assets/droid`) — appropriate since actions are Franka joint-velocity in the same regime

**wandb run**: https://wandb.ai/yiyanpeng2027-northwestern-university/openpi/runs/8tp8knks

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

## Use this checkpoint

```bash
# Download just this step
HF_HUB_DISABLE_XET=1 huggingface-cli download IDEAS-Lab-Northwestern/pi05-real-cab-60-droid-lora \
  --include "19999/**" \
  --local-dir vla_models/pi05-real-cab-60-droid-lora

# Serve with openpi
uv run scripts/serve_policy.py \
  --policy.config=pi05_droid_finetune_lora \
  --policy.dir=vla_models/pi05-real-cab-60-droid-lora/19999
```

## Sibling

This is one of two paired LoRA fine-tunes evaluating safety awareness in pi0.5-class VLA models:

- This repo: **cab task**
- Sibling: [`IDEAS-Lab-Northwestern/pi05-real-jar-60-droid-lora`](https://huggingface.co/IDEAS-Lab-Northwestern/pi05-real-jar-60-droid-lora) — jar task

Both share the same TrainConfig and warm-start; differ only in dataset + prompt.

## License

Apache-2.0