Preserve HF-data campaign provenance alongside two newly published task peaks
Browse files
README.md
CHANGED
|
@@ -9,8 +9,9 @@ license: mit
|
|
| 9 |
One PI0.5 LoRA checkpoint per ManiLadder task (92 tasks), each finetuned on 20 demos.
|
| 10 |
Checkpoints are selected by the highest **50-episode** success rate; ties use the earliest step.
|
| 11 |
|
| 12 |
-
- The
|
| 13 |
- The 22 completed retraining runs published through 2026-10-30 use seeds 800000–800049. All reached 40,000 training steps. 15 have all eight 5k–40k evaluations; 7 have seven evaluations because one intermediate checkpoint was not retained. Their selected weights are the best among the available evaluated checkpoints.
|
|
|
|
| 14 |
|
| 15 |
See [retraining provenance and missing steps](single_task/pi05_retrain_eval50_20261022_manifest.json). These scores use different seed sets across the two groups and are not a controlled comparison.
|
| 16 |
|
|
|
|
| 9 |
One PI0.5 LoRA checkpoint per ManiLadder task (92 tasks), each finetuned on 20 demos.
|
| 10 |
Checkpoints are selected by the highest **50-episode** success rate; ties use the earliest step.
|
| 11 |
|
| 12 |
+
- The 60 previously published tasks use fixed evaluation seeds 1000–1049.
|
| 13 |
- The 22 completed retraining runs published through 2026-10-30 use seeds 800000–800049. All reached 40,000 training steps. 15 have all eight 5k–40k evaluations; 7 have seven evaluations because one intermediate checkpoint was not retained. Their selected weights are the best among the available evaluated checkpoints.
|
| 14 |
+
- The 10 tasks from the HF-data retraining campaign `pi05_hf21_20261025` (added 2026-10-29; raw demos freshly downloaded from Linslab/ManiLadder_Dataset) use seeds 1000–1049. All reached 40,000 training steps and have all eight 5k–40k evaluations. See [its manifest](single_task/pi05_hf21_20261025_manifest.json).
|
| 15 |
|
| 16 |
See [retraining provenance and missing steps](single_task/pi05_retrain_eval50_20261022_manifest.json). These scores use different seed sets across the two groups and are not a controlled comparison.
|
| 17 |
|