Index 23 HF-data retraining PI0.5 peaks (pi05_hf21_20261025) with 50-episode evaluations
Browse files- README.md +5 -3
- single_task/all_checkpoints_eval.csv +16 -0
- single_task/pi05_hf21_20261025_manifest.json +30 -0
- single_task/results.csv +2 -0
README.md
CHANGED
|
@@ -6,14 +6,14 @@ license: mit
|
|
| 6 |
|
| 7 |
## single_task
|
| 8 |
|
| 9 |
-
One PI0.5 LoRA checkpoint per ManiLadder task (
|
| 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 21 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 |
- The 3 fresh-data runs in `pi05_newdata_3tasks_20261028` use pinned dataset revision `d120457e637b01679fbb40fd4de1d2394f0d55c9`, 20 demos per task, 40,000 training steps, and all eight 5k–40k eval50 checkpoints with seeds 800000–800049. **All three scored 0/50 at every checkpoint**; step 5,000 is the earliest tied checkpoint. See [their manifest](single_task/pi05_newdata_3tasks_20261028_manifest.json).
|
|
|
|
| 17 |
|
| 18 |
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.
|
| 19 |
|
|
@@ -24,7 +24,7 @@ Each `single_task/<task>/` folder contains:
|
|
| 24 |
- `eval_result.json` — 50-episode evaluation of this checkpoint
|
| 25 |
|
| 26 |
`single_task/results.csv` lists the selected checkpoint per task; `single_task/all_checkpoints_eval.csv` has the
|
| 27 |
-
50-episode result of every evaluated checkpoint (
|
| 28 |
|
| 29 |
Evaluation: `exec_steps=16`, `sample_steps=10`. Tasks whose checkpoints all scored 0/50 are still included;
|
| 30 |
for them the step choice is arbitrary.
|
|
@@ -107,6 +107,7 @@ for them the step choice is arbitrary.
|
|
| 107 |
| HitBox-v1 | 3 | 25000 | 0.06 |
|
| 108 |
| LiquidMeasuring-v1 | 3 | 35000 | 0.32 |
|
| 109 |
| MobilePickPlace-v1 | 3 | 15000 | 0.08 |
|
|
|
|
| 110 |
| PushTargetPose-v1 | 3 | 30000 | 0.02 |
|
| 111 |
| SeparateTwoRopesEnv-v1 | 3 | 35000 | 0.26 |
|
| 112 |
| Stack6Blocks-v1 | 3 | 5000 | 0.00 |
|
|
@@ -135,6 +136,7 @@ for them the step choice is arbitrary.
|
|
| 135 |
| RodThroughHoles-v1 | 4 | 5000 | 0.02 |
|
| 136 |
| SpinPenHora-v1 | 4 | 5000 | 0.00 |
|
| 137 |
| StampRoll-v1 | 4 | 5000 | 0.00 |
|
|
|
|
| 138 |
| TennisRally-v1 | 4 | 15000 | 0.06 |
|
| 139 |
| WrapRope-v1 | 4 | 5000 | 0.02 |
|
| 140 |
|
|
|
|
| 6 |
|
| 7 |
## single_task
|
| 8 |
|
| 9 |
+
One PI0.5 LoRA checkpoint per ManiLadder task (108 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 |
|
| 15 |
- The 3 fresh-data runs in `pi05_newdata_3tasks_20261028` use pinned dataset revision `d120457e637b01679fbb40fd4de1d2394f0d55c9`, 20 demos per task, 40,000 training steps, and all eight 5k–40k eval50 checkpoints with seeds 800000–800049. **All three scored 0/50 at every checkpoint**; step 5,000 is the earliest tied checkpoint. See [their manifest](single_task/pi05_newdata_3tasks_20261028_manifest.json).
|
| 16 |
+
- The 23 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).
|
| 17 |
|
| 18 |
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.
|
| 19 |
|
|
|
|
| 24 |
- `eval_result.json` — 50-episode evaluation of this checkpoint
|
| 25 |
|
| 26 |
`single_task/results.csv` lists the selected checkpoint per task; `single_task/all_checkpoints_eval.csv` has the
|
| 27 |
+
50-episode result of every evaluated checkpoint (854 total).
|
| 28 |
|
| 29 |
Evaluation: `exec_steps=16`, `sample_steps=10`. Tasks whose checkpoints all scored 0/50 are still included;
|
| 30 |
for them the step choice is arbitrary.
|
|
|
|
| 107 |
| HitBox-v1 | 3 | 25000 | 0.06 |
|
| 108 |
| LiquidMeasuring-v1 | 3 | 35000 | 0.32 |
|
| 109 |
| MobilePickPlace-v1 | 3 | 15000 | 0.08 |
|
| 110 |
+
| PressFitSoftPeg-v1 | 3 | 20000 | 0.08 |
|
| 111 |
| PushTargetPose-v1 | 3 | 30000 | 0.02 |
|
| 112 |
| SeparateTwoRopesEnv-v1 | 3 | 35000 | 0.26 |
|
| 113 |
| Stack6Blocks-v1 | 3 | 5000 | 0.00 |
|
|
|
|
| 136 |
| RodThroughHoles-v1 | 4 | 5000 | 0.02 |
|
| 137 |
| SpinPenHora-v1 | 4 | 5000 | 0.00 |
|
| 138 |
| StampRoll-v1 | 4 | 5000 | 0.00 |
|
| 139 |
+
| SwingRing-v1 | 4 | 15000 | 0.04 |
|
| 140 |
| TennisRally-v1 | 4 | 15000 | 0.06 |
|
| 141 |
| WrapRope-v1 | 4 | 5000 | 0.02 |
|
| 142 |
|
single_task/all_checkpoints_eval.csv
CHANGED
|
@@ -601,6 +601,14 @@ MobilePickPlace-v1,level_3,25000,0.02,1,nan
|
|
| 601 |
MobilePickPlace-v1,level_3,30000,0.02,1,nan
|
| 602 |
MobilePickPlace-v1,level_3,35000,0.04,2,169.25979777826228
|
| 603 |
MobilePickPlace-v1,level_3,40000,0.04,2,nan
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 604 |
PushTargetPose-v1,level_3,5000,0.0,0,-348.17943846166133
|
| 605 |
PushTargetPose-v1,level_3,10000,0.0,0,-391.1001693305373
|
| 606 |
PushTargetPose-v1,level_3,15000,0.0,0,-293.1471811267734
|
|
@@ -821,6 +829,14 @@ StampRoll-v1,level_4,25000,0.0,0,0.0
|
|
| 821 |
StampRoll-v1,level_4,30000,0.0,0,0.0
|
| 822 |
StampRoll-v1,level_4,35000,0.0,0,0.0
|
| 823 |
StampRoll-v1,level_4,40000,0.0,0,0.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 824 |
TennisRally-v1,level_4,5000,0.02,1,357.17158117782327
|
| 825 |
TennisRally-v1,level_4,10000,0.02,1,582.0495989676378
|
| 826 |
TennisRally-v1,level_4,15000,0.06,3,631.176888383422
|
|
|
|
| 601 |
MobilePickPlace-v1,level_3,30000,0.02,1,nan
|
| 602 |
MobilePickPlace-v1,level_3,35000,0.04,2,169.25979777826228
|
| 603 |
MobilePickPlace-v1,level_3,40000,0.04,2,nan
|
| 604 |
+
PressFitSoftPeg-v1,level_3,5000,0.0,0,693.0195
|
| 605 |
+
PressFitSoftPeg-v1,level_3,10000,0.0,0,1621.6199599999998
|
| 606 |
+
PressFitSoftPeg-v1,level_3,15000,0.0,0,1761.0606
|
| 607 |
+
PressFitSoftPeg-v1,level_3,20000,0.08,4,1640.69166
|
| 608 |
+
PressFitSoftPeg-v1,level_3,25000,0.02,1,1553.72812
|
| 609 |
+
PressFitSoftPeg-v1,level_3,30000,0.08,4,1815.9183600000001
|
| 610 |
+
PressFitSoftPeg-v1,level_3,35000,0.08,4,1902.3642799999998
|
| 611 |
+
PressFitSoftPeg-v1,level_3,40000,0.06,3,1725.1256
|
| 612 |
PushTargetPose-v1,level_3,5000,0.0,0,-348.17943846166133
|
| 613 |
PushTargetPose-v1,level_3,10000,0.0,0,-391.1001693305373
|
| 614 |
PushTargetPose-v1,level_3,15000,0.0,0,-293.1471811267734
|
|
|
|
| 829 |
StampRoll-v1,level_4,30000,0.0,0,0.0
|
| 830 |
StampRoll-v1,level_4,35000,0.0,0,0.0
|
| 831 |
StampRoll-v1,level_4,40000,0.0,0,0.0
|
| 832 |
+
SwingRing-v1,level_4,5000,0.02,1,73.43362
|
| 833 |
+
SwingRing-v1,level_4,10000,0.02,1,77.27958
|
| 834 |
+
SwingRing-v1,level_4,15000,0.04,2,78.49958
|
| 835 |
+
SwingRing-v1,level_4,20000,0.0,0,76.71624
|
| 836 |
+
SwingRing-v1,level_4,25000,0.0,0,71.01814
|
| 837 |
+
SwingRing-v1,level_4,30000,0.02,1,75.07314000000001
|
| 838 |
+
SwingRing-v1,level_4,35000,0.0,0,73.35582000000001
|
| 839 |
+
SwingRing-v1,level_4,40000,0.0,0,74.07002
|
| 840 |
TennisRally-v1,level_4,5000,0.02,1,357.17158117782327
|
| 841 |
TennisRally-v1,level_4,10000,0.02,1,582.0495989676378
|
| 842 |
TennisRally-v1,level_4,15000,0.06,3,631.176888383422
|
single_task/pi05_hf21_20261025_manifest.json
CHANGED
|
@@ -155,6 +155,21 @@
|
|
| 155 |
"checkpoint": "single_task/DualArmThreading-v1/lora_step_0020000.npz",
|
| 156 |
"checkpoint_sha256": "7e815721a302d324c55dea8c9681332c939b65d432b3918e2a00d365828ffedd"
|
| 157 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 158 |
{
|
| 159 |
"task": "DualArmMakeLine-v1",
|
| 160 |
"level": "level_3",
|
|
@@ -290,6 +305,21 @@
|
|
| 290 |
"checkpoint": "single_task/CoverPattern-v1/lora_step_0030000.npz",
|
| 291 |
"checkpoint_sha256": "0f555c7b620fa9355b6f56ee378c7531d89e630c9b6383210cdac21cd2dae020"
|
| 292 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 293 |
{
|
| 294 |
"task": "StampRoll-v1",
|
| 295 |
"level": "level_4",
|
|
|
|
| 155 |
"checkpoint": "single_task/DualArmThreading-v1/lora_step_0020000.npz",
|
| 156 |
"checkpoint_sha256": "7e815721a302d324c55dea8c9681332c939b65d432b3918e2a00d365828ffedd"
|
| 157 |
},
|
| 158 |
+
{
|
| 159 |
+
"task": "PressFitSoftPeg-v1",
|
| 160 |
+
"level": "level_3",
|
| 161 |
+
"suite": "deformable/single_arm_dex",
|
| 162 |
+
"step": 20000,
|
| 163 |
+
"success_rate": 0.08,
|
| 164 |
+
"successes": 4,
|
| 165 |
+
"episodes": 50,
|
| 166 |
+
"episode_seed_base": 1000,
|
| 167 |
+
"ckpts_evaluated": 8,
|
| 168 |
+
"missing_eval_steps": [],
|
| 169 |
+
"campaign": "pi05_hf21_20261025",
|
| 170 |
+
"checkpoint": "single_task/PressFitSoftPeg-v1/lora_step_0020000.npz",
|
| 171 |
+
"checkpoint_sha256": "3f798221f49b418dee3fbf610282e8bbb29499d02705a072d0e4053cfcf051c4"
|
| 172 |
+
},
|
| 173 |
{
|
| 174 |
"task": "DualArmMakeLine-v1",
|
| 175 |
"level": "level_3",
|
|
|
|
| 305 |
"checkpoint": "single_task/CoverPattern-v1/lora_step_0030000.npz",
|
| 306 |
"checkpoint_sha256": "0f555c7b620fa9355b6f56ee378c7531d89e630c9b6383210cdac21cd2dae020"
|
| 307 |
},
|
| 308 |
+
{
|
| 309 |
+
"task": "SwingRing-v1",
|
| 310 |
+
"level": "level_4",
|
| 311 |
+
"suite": "deformable/single_arm_dex",
|
| 312 |
+
"step": 15000,
|
| 313 |
+
"success_rate": 0.04,
|
| 314 |
+
"successes": 2,
|
| 315 |
+
"episodes": 50,
|
| 316 |
+
"episode_seed_base": 1000,
|
| 317 |
+
"ckpts_evaluated": 8,
|
| 318 |
+
"missing_eval_steps": [],
|
| 319 |
+
"campaign": "pi05_hf21_20261025",
|
| 320 |
+
"checkpoint": "single_task/SwingRing-v1/lora_step_0015000.npz",
|
| 321 |
+
"checkpoint_sha256": "4ba44fb21dbbc9435859c93e0796840c75f7855ba3fc996e4b748f6d4360085e"
|
| 322 |
+
},
|
| 323 |
{
|
| 324 |
"task": "StampRoll-v1",
|
| 325 |
"level": "level_4",
|
single_task/results.csv
CHANGED
|
@@ -75,6 +75,7 @@ HammerNail-v1,level_3,35000,0.12,6,8,single_task/HammerNail-v1/lora_step_0035000
|
|
| 75 |
HitBox-v1,level_3,25000,0.06,3,8,single_task/HitBox-v1/lora_step_0025000.npz
|
| 76 |
LiquidMeasuring-v1,level_3,35000,0.32,16,8,single_task/LiquidMeasuring-v1/lora_step_0035000.npz
|
| 77 |
MobilePickPlace-v1,level_3,15000,0.08,4,8,single_task/MobilePickPlace-v1/lora_step_0015000.npz
|
|
|
|
| 78 |
PushTargetPose-v1,level_3,30000,0.02,1,8,single_task/PushTargetPose-v1/lora_step_0030000.npz
|
| 79 |
SeparateTwoRopesEnv-v1,level_3,35000,0.26,13,7,single_task/SeparateTwoRopesEnv-v1/lora_step_0035000.npz
|
| 80 |
Stack6Blocks-v1,level_3,5000,0.0,0,8,single_task/Stack6Blocks-v1/lora_step_0005000.npz
|
|
@@ -103,5 +104,6 @@ RingTossing-v1,level_4,5000,0.0,0,8,single_task/RingTossing-v1/lora_step_0005000
|
|
| 103 |
RodThroughHoles-v1,level_4,5000,0.02,1,8,single_task/RodThroughHoles-v1/lora_step_0005000.npz
|
| 104 |
SpinPenHora-v1,level_4,5000,0.0,0,8,single_task/SpinPenHora-v1/lora_step_0005000.npz
|
| 105 |
StampRoll-v1,level_4,5000,0.0,0,8,single_task/StampRoll-v1/lora_step_0005000.npz
|
|
|
|
| 106 |
TennisRally-v1,level_4,15000,0.06,3,8,single_task/TennisRally-v1/lora_step_0015000.npz
|
| 107 |
WrapRope-v1,level_4,5000,0.02,1,8,single_task/WrapRope-v1/lora_step_0005000.npz
|
|
|
|
| 75 |
HitBox-v1,level_3,25000,0.06,3,8,single_task/HitBox-v1/lora_step_0025000.npz
|
| 76 |
LiquidMeasuring-v1,level_3,35000,0.32,16,8,single_task/LiquidMeasuring-v1/lora_step_0035000.npz
|
| 77 |
MobilePickPlace-v1,level_3,15000,0.08,4,8,single_task/MobilePickPlace-v1/lora_step_0015000.npz
|
| 78 |
+
PressFitSoftPeg-v1,level_3,20000,0.08,4,8,single_task/PressFitSoftPeg-v1/lora_step_0020000.npz
|
| 79 |
PushTargetPose-v1,level_3,30000,0.02,1,8,single_task/PushTargetPose-v1/lora_step_0030000.npz
|
| 80 |
SeparateTwoRopesEnv-v1,level_3,35000,0.26,13,7,single_task/SeparateTwoRopesEnv-v1/lora_step_0035000.npz
|
| 81 |
Stack6Blocks-v1,level_3,5000,0.0,0,8,single_task/Stack6Blocks-v1/lora_step_0005000.npz
|
|
|
|
| 104 |
RodThroughHoles-v1,level_4,5000,0.02,1,8,single_task/RodThroughHoles-v1/lora_step_0005000.npz
|
| 105 |
SpinPenHora-v1,level_4,5000,0.0,0,8,single_task/SpinPenHora-v1/lora_step_0005000.npz
|
| 106 |
StampRoll-v1,level_4,5000,0.0,0,8,single_task/StampRoll-v1/lora_step_0005000.npz
|
| 107 |
+
SwingRing-v1,level_4,15000,0.04,2,8,single_task/SwingRing-v1/lora_step_0015000.npz
|
| 108 |
TennisRally-v1,level_4,15000,0.06,3,8,single_task/TennisRally-v1/lora_step_0015000.npz
|
| 109 |
WrapRope-v1,level_4,5000,0.02,1,8,single_task/WrapRope-v1/lora_step_0005000.npz
|