Views
No views yet
| Property | Value |
|---|---|
| OpenPI config | pi05_sir_droid_finetune |
| Checkpoint step | 4999 |
| Training data | N/A |
| Precision | bfloat16 |
| Parameter size | ~6.7 GB |
| Source checkpoint | /iris/u/ankile/openpi-pi05-real01c/checkpoints/pi05_sir_droid_finetune/sir_real01c_sobol_r0_5k_bs32_iris_h200/4999 |
| Hugging Face repo | ankile/openpi-pi05-real01c-insert-marker-d1-sobol-r0-ft-iris |
| W&B run | link |
| SLURM job ID | 15689480 |
1# Download checkpoint from HF Hub
2huggingface-cli download ankile/openpi-pi05-real01c-insert-marker-d1-sobol-r0-ft-iris --local-dir <local_path>
3
4# Run inference server
5cd deps/openpi
6uv run python scripts/serve_policy.py pi05_sir_droid_finetune \
7 --checkpoint-dir <local_path>1from openpi.training import config as openpi_config
2from openpi.policies import policy_config as openpi_policy_config
3
4train_config = openpi_config.get_config("pi05_sir_droid_finetune")
5policy = openpi_policy_config.create_trained_policy(
6 train_config, "<local_path>"
7)
8result = policy.infer(obs_dict)
9actions = result["actions"]├── _CHECKPOINT_METADATA
├── checkpoint_provenance.json
├── openpi_config.json
├── assets/
│ └── (normalization stats)
└── params/
└── (orbax checkpoint files)