|
Download README.md from hrl-imerse/doug-act-stack-80k: direct link, hf CLI and curl.
- Browser
- Download file 878 Bytes
-
https://huggingface.co/hrl-imerse/doug-act-stack-80k/resolve/main/README.md
- Command line
-
hf download hf://hrl-imerse/doug-act-stack-80k/README.md
-
curl -L -o README.md https://huggingface.co/hrl-imerse/doug-act-stack-80k/resolve/main/README.md
878 Bytes
| tags: [LeRobot, ACT, so101, robotics] | |
| datasets: [hrl-imerse/doug-stack-98-trimmed] | |
| # ACT block stacking, 80k steps (SO-101, wrist camera) | |
| ACT (chunk_size 100, batch 8), trained on `hrl-imerse/doug-stack-98-trimmed` for 170k steps (88 train / 10 val episodes). This is the 80k checkpoint. | |
| - Augmentation: colour jitter plus RandomAffine (±5°, ±5% shift). | |
| - Validation loss plateaued at about 0.57 after 10k steps. | |
| **Real-arm tests (2026-11-17, temporal ensembling coeff 0.01, one trial each):** | |
| | Checkpoint | Result | | |
| |---|---| | |
| | **80k** | **placed orange perfectly, then missed green**; the best of 20k/60k/80k/100k/170k | | |
| | 20k, 60k, 100k, 170k | 0 blocks | | |
| - Exec 15 was jerky and much worse than ensembling. | |
| - Grasps often missed by about 0.5 in on either side. | |
| Run with `lerobot-rollout ... --policy.n_action_steps=1 --policy.temporal_ensemble_coeff=0.01`. | |