mimiminsoo/pi0_150

LoRA fine-tune of pi0_base (openpi) on the merged mimiminsoo/marker_150 SO101 marker-in-cup dataset (146 episodes, marker_50 + marker_100 combined).

config: pi0_150
checkpoint step: 29999
batch_size: 8, action_horizon: 30, fsdp_devices: 2 (2x RTX 3090)

Serving

uv run scripts/serve_policy.py --port=8000 \
    policy:checkpoint --policy.config=pi0_150 --policy.dir=<local path to this repo>

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support