Instructions to use Autobrik/SO-ARM100-dump-pocket-cleaning2-pi05 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LeRobot
How to use Autobrik/SO-ARM100-dump-pocket-cleaning2-pi05 with LeRobot:
- Notebooks
- Google Colab
- Kaggle
pi05 finetuned on SO-ARM100 dump-pocket-cleaning
This is a pi05 policy finetuned with LeRobot.
Training summary
- policy:
pi05 - pretrained base:
lerobot/pi05_base - dataset:
Autobrik/SO-ARM100-dump-pocket-cleaning2 - run_dir:
outputs/train/benchmark/20260215_075453_benchmark_pi05 - seed:
1000
Best offline validation
- best_val_step:
25774 - best_val_loss:
0.07445306489484109
Included artifacts
checkpoints/best(recommended for eval/deploy)checkpoints/last(resume/debug)best_val.json