Hugging Face's logo

FidelityWM
/
planning-baselines

world-model
planning
dinowm
pldm
fast-lewm
Model card Files
xet
 Community
greycat111 commited on
Commit
2ffd52b
·
verified ·
1 Parent(s): 4c9f4e6

Complete PLDM PushT five-seed planning evaluation and report mean/std

Browse files
Files changed (30) hide show
  1. README.md +1 -0
  2. pldm/pusht/README.md +5 -0
  3. pldm/pusht/provenance.json +16 -2
  4. results/completed_seed_results.csv +4 -0
  5. results/completed_seed_results.md +4 -0
  6. results/planning_success.csv +20 -20
  7. results/planning_success.json +160 -160
  8. results/planning_success.xlsx +0 -0
  9. results/planning_summary.csv +4 -4
  10. results/planning_summary.json +36 -36
  11. results/pldm_pusht_100.json +306 -0
  12. results/pldm_pusht_100_seed43.json +306 -0
  13. results/pldm_pusht_100_seed44.json +306 -0
  14. results/pldm_pusht_100_seed45.json +306 -0
  15. results/pldm_pusht_100_seed46.json +306 -0
  16. results/pldm_pusht_25.json +306 -0
  17. results/pldm_pusht_25_seed43.json +306 -0
  18. results/pldm_pusht_25_seed44.json +306 -0
  19. results/pldm_pusht_25_seed45.json +306 -0
  20. results/pldm_pusht_25_seed46.json +306 -0
  21. results/pldm_pusht_50.json +306 -0
  22. results/pldm_pusht_50_seed43.json +306 -0
  23. results/pldm_pusht_50_seed44.json +306 -0
  24. results/pldm_pusht_50_seed45.json +306 -0
  25. results/pldm_pusht_50_seed46.json +306 -0
  26. results/pldm_pusht_75.json +306 -0
  27. results/pldm_pusht_75_seed43.json +306 -0
  28. results/pldm_pusht_75_seed44.json +306 -0
  29. results/pldm_pusht_75_seed45.json +306 -0
  30. results/pldm_pusht_75_seed46.json +306 -0
README.md CHANGED
@@ -15,6 +15,7 @@ This repository archives checkpoints and measured planning results for TwoRoom,
15
  | fast-lewm/tworoom, cube, pusht | Exact original `naiverer/fast-leworldmodel` checkpoints | All five-seed planning evaluations completed |
16
  | dinowm/pusht | Original DINO-WM OSF PushT checkpoint and frozen DINOv2-S/14 backbone | Published model trained on `pusht_noise`; evaluation incomplete |
17
  | pldm/cube | Final epoch-10 model, epoch-8 resume checkpoint, configuration and provenance | Training and all five-seed planning evaluations complete |
 
18
  | pldm/tworoom | Configuration and results only | The trained weights were deleted after evaluation under the earlier disk-saving instruction and are unavailable |
19
 
20
  Newly completed local checkpoints are archived under `<method>/<task>/model.pt`, with provenance alongside each file. Each local checkpoint uses training seed 42. Five evaluation seeds do not mean five independently trained models.
 
15
  | fast-lewm/tworoom, cube, pusht | Exact original `naiverer/fast-leworldmodel` checkpoints | All five-seed planning evaluations completed |
16
  | dinowm/pusht | Original DINO-WM OSF PushT checkpoint and frozen DINOv2-S/14 backbone | Published model trained on `pusht_noise`; evaluation incomplete |
17
  | pldm/cube | Final epoch-10 model, epoch-8 resume checkpoint, configuration and provenance | Training and all five-seed planning evaluations complete |
18
+ | pldm/pusht | Final epoch-10 model, configuration and provenance | Training and all five-seed planning evaluations complete |
19
  | pldm/tworoom | Configuration and results only | The trained weights were deleted after evaluation under the earlier disk-saving instruction and are unavailable |
20
 
21
  Newly completed local checkpoints are archived under `<method>/<task>/model.pt`, with provenance alongside each file. Each local checkpoint uses training seed 42. Five evaluation seeds do not mean five independently trained models.
pldm/pusht/README.md ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ # PLDM PushT checkpoints
2
+
3
+ `model.pt` is the final 10-epoch model trained with seed 42. All five evaluation seeds (42–46) and four goal offsets (25, 50, 75, 100) now have 200 completed episodes each. See ../../results/planning_success.xlsx for per-seed scores and mean ± sample standard deviation.
4
+
5
+ Trained on the supplied `pusht_expert_train.h5`. Final-model metadata is in `provenance.json`.
pldm/pusht/provenance.json CHANGED
@@ -3,6 +3,20 @@
3
  "epochs_completed": 10,
4
  "training_seed": 42,
5
  "variant": "PLDM stable-worldmodel recipe",
6
- "status": "Final epoch-10 model; five-seed evaluation queued",
7
- "training_dataset": "pusht_expert_train.h5"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
8
  }
 
3
  "epochs_completed": 10,
4
  "training_seed": 42,
5
  "variant": "PLDM stable-worldmodel recipe",
6
+ "status": "Final epoch-10 model; all 20 planning runs completed",
7
+ "training_dataset": "pusht_expert_train.h5",
8
+ "evaluation_seeds": [
9
+ 42,
10
+ 43,
11
+ 44,
12
+ 45,
13
+ 46
14
+ ],
15
+ "goal_offsets": [
16
+ 25,
17
+ 50,
18
+ 75,
19
+ 100
20
+ ],
21
+ "samples_per_run": 200
22
  }
results/completed_seed_results.csv CHANGED
@@ -7,6 +7,10 @@ pldm,cube,25,55.0,52.0,56.0,56.5,66.0,5,57.1,5.272570530585627,57.10 ± 5.27
7
  pldm,cube,50,49.5,39.0,41.5,44.5,50.5,5,45.0,4.9749371855331,45.00 ± 4.97
8
  pldm,cube,75,54.5,58.5,57.0,54.5,51.5,5,55.2,2.6832815729997477,55.20 ± 2.68
9
  pldm,cube,100,57.5,58.0,51.5,54.0,53.0,5,54.8,2.8416544476765644,54.80 ± 2.84
 
 
 
 
10
  fast-lewm,tworoom,25,88.0,90.0,86.5,88.5,91.0,5,88.8,1.753567791675018,88.80 ± 1.75
11
  fast-lewm,tworoom,50,64.0,58.0,59.0,63.5,66.0,5,62.1,3.4351128074635335,62.10 ± 3.44
12
  fast-lewm,tworoom,75,44.5,33.5,35.5,34.0,38.0,5,37.1,4.491658936295141,37.10 ± 4.49
 
7
  pldm,cube,50,49.5,39.0,41.5,44.5,50.5,5,45.0,4.9749371855331,45.00 ± 4.97
8
  pldm,cube,75,54.5,58.5,57.0,54.5,51.5,5,55.2,2.6832815729997477,55.20 ± 2.68
9
  pldm,cube,100,57.5,58.0,51.5,54.0,53.0,5,54.8,2.8416544476765644,54.80 ± 2.84
10
+ pldm,pusht,25,76.5,75.0,72.0,77.0,76.0,5,75.3,1.9874606914351791,75.30 ± 1.99
11
+ pldm,pusht,50,28.0,20.5,30.5,24.0,29.5,5,26.5,4.168333000133266,26.50 ± 4.17
12
+ pldm,pusht,75,13.5,9.5,14.0,11.0,13.5,5,12.3,1.9557607215607946,12.30 ± 1.96
13
+ pldm,pusht,100,6.5,6.5,4.5,6.0,7.5,5,6.2,1.0954451150103321,6.20 ± 1.10
14
  fast-lewm,tworoom,25,88.0,90.0,86.5,88.5,91.0,5,88.8,1.753567791675018,88.80 ± 1.75
15
  fast-lewm,tworoom,50,64.0,58.0,59.0,63.5,66.0,5,62.1,3.4351128074635335,62.10 ± 3.44
16
  fast-lewm,tworoom,75,44.5,33.5,35.5,34.0,38.0,5,37.1,4.491658936295141,37.10 ± 4.49
results/completed_seed_results.md CHANGED
@@ -12,6 +12,10 @@ Success percentages, 200 episodes per seed and offset. Sample standard deviation
12
  | pldm | cube | 50 | 49.5 | 39.0 | 41.5 | 44.5 | 50.5 | 45.00 ± 4.97 |
13
  | pldm | cube | 75 | 54.5 | 58.5 | 57.0 | 54.5 | 51.5 | 55.20 ± 2.68 |
14
  | pldm | cube | 100 | 57.5 | 58.0 | 51.5 | 54.0 | 53.0 | 54.80 ± 2.84 |
 
 
 
 
15
  | fast-lewm | tworoom | 25 | 88.0 | 90.0 | 86.5 | 88.5 | 91.0 | 88.80 ± 1.75 |
16
  | fast-lewm | tworoom | 50 | 64.0 | 58.0 | 59.0 | 63.5 | 66.0 | 62.10 ± 3.44 |
17
  | fast-lewm | tworoom | 75 | 44.5 | 33.5 | 35.5 | 34.0 | 38.0 | 37.10 ± 4.49 |
 
12
  | pldm | cube | 50 | 49.5 | 39.0 | 41.5 | 44.5 | 50.5 | 45.00 ± 4.97 |
13
  | pldm | cube | 75 | 54.5 | 58.5 | 57.0 | 54.5 | 51.5 | 55.20 ± 2.68 |
14
  | pldm | cube | 100 | 57.5 | 58.0 | 51.5 | 54.0 | 53.0 | 54.80 ± 2.84 |
15
+ | pldm | pusht | 25 | 76.5 | 75.0 | 72.0 | 77.0 | 76.0 | 75.30 ± 1.99 |
16
+ | pldm | pusht | 50 | 28.0 | 20.5 | 30.5 | 24.0 | 29.5 | 26.50 ± 4.17 |
17
+ | pldm | pusht | 75 | 13.5 | 9.5 | 14.0 | 11.0 | 13.5 | 12.30 ± 1.96 |
18
+ | pldm | pusht | 100 | 6.5 | 6.5 | 4.5 | 6.0 | 7.5 | 6.20 ± 1.10 |
19
  | fast-lewm | tworoom | 25 | 88.0 | 90.0 | 86.5 | 88.5 | 91.0 | 88.80 ± 1.75 |
20
  | fast-lewm | tworoom | 50 | 64.0 | 58.0 | 59.0 | 63.5 | 66.0 | 62.10 ± 3.44 |
21
  | fast-lewm | tworoom | 75 | 44.5 | 33.5 | 35.5 | 34.0 | 38.0 | 37.10 ± 4.49 |
results/planning_success.csv CHANGED
@@ -99,26 +99,26 @@ pldm,cube,100,43,200,200,116,58.0,complete,50,Locally trained 10 epochs: PLDM st
99
  pldm,cube,100,44,200,200,103,51.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
100
  pldm,cube,100,45,200,200,108,54.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
101
  pldm,cube,100,46,200,200,106,53.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
102
- pldm,pusht,25,42,0,200,,,unavailable: compatible checkpoint not found,50,,,,
103
- pldm,pusht,25,43,0,200,,,unavailable: compatible checkpoint not found,50,,,,
104
- pldm,pusht,25,44,0,200,,,unavailable: compatible checkpoint not found,50,,,,
105
- pldm,pusht,25,45,0,200,,,unavailable: compatible checkpoint not found,50,,,,
106
- pldm,pusht,25,46,0,200,,,unavailable: compatible checkpoint not found,50,,,,
107
- pldm,pusht,50,42,0,200,,,unavailable: compatible checkpoint not found,50,,,,
108
- pldm,pusht,50,43,0,200,,,unavailable: compatible checkpoint not found,50,,,,
109
- pldm,pusht,50,44,0,200,,,unavailable: compatible checkpoint not found,50,,,,
110
- pldm,pusht,50,45,0,200,,,unavailable: compatible checkpoint not found,50,,,,
111
- pldm,pusht,50,46,0,200,,,unavailable: compatible checkpoint not found,50,,,,
112
- pldm,pusht,75,42,0,200,,,unavailable: compatible checkpoint not found,50,,,,
113
- pldm,pusht,75,43,0,200,,,unavailable: compatible checkpoint not found,50,,,,
114
- pldm,pusht,75,44,0,200,,,unavailable: compatible checkpoint not found,50,,,,
115
- pldm,pusht,75,45,0,200,,,unavailable: compatible checkpoint not found,50,,,,
116
- pldm,pusht,75,46,0,200,,,unavailable: compatible checkpoint not found,50,,,,
117
- pldm,pusht,100,42,0,200,,,unavailable: compatible checkpoint not found,50,,,,
118
- pldm,pusht,100,43,0,200,,,unavailable: compatible checkpoint not found,50,,,,
119
- pldm,pusht,100,44,0,200,,,unavailable: compatible checkpoint not found,50,,,,
120
- pldm,pusht,100,45,0,200,,,unavailable: compatible checkpoint not found,50,,,,
121
- pldm,pusht,100,46,0,200,,,unavailable: compatible checkpoint not found,50,,,,
122
  fast-lewm,tworoom,25,42,200,200,176,88.0,complete,50,naiverer/fast-leworldmodel,,,
123
  fast-lewm,tworoom,25,43,200,200,180,90.0,complete,50,naiverer/fast-leworldmodel,,,
124
  fast-lewm,tworoom,25,44,200,200,173,86.5,complete,50,naiverer/fast-leworldmodel,,,
 
99
  pldm,cube,100,44,200,200,103,51.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
100
  pldm,cube,100,45,200,200,108,54.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
101
  pldm,cube,100,46,200,200,106,53.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
102
+ pldm,pusht,25,42,200,200,153,76.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
103
+ pldm,pusht,25,43,200,200,150,75.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
104
+ pldm,pusht,25,44,200,200,144,72.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
105
+ pldm,pusht,25,45,200,200,154,77.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
106
+ pldm,pusht,25,46,200,200,152,76.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
107
+ pldm,pusht,50,42,200,200,56,28.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
108
+ pldm,pusht,50,43,200,200,41,20.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
109
+ pldm,pusht,50,44,200,200,61,30.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
110
+ pldm,pusht,50,45,200,200,48,24.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
111
+ pldm,pusht,50,46,200,200,59,29.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
112
+ pldm,pusht,75,42,200,200,27,13.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
113
+ pldm,pusht,75,43,200,200,19,9.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
114
+ pldm,pusht,75,44,200,200,28,14.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
115
+ pldm,pusht,75,45,200,200,22,11.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
116
+ pldm,pusht,75,46,200,200,27,13.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
117
+ pldm,pusht,100,42,200,200,13,6.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
118
+ pldm,pusht,100,43,200,200,13,6.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
119
+ pldm,pusht,100,44,200,200,9,4.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
120
+ pldm,pusht,100,45,200,200,12,6.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
121
+ pldm,pusht,100,46,200,200,15,7.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
122
  fast-lewm,tworoom,25,42,200,200,176,88.0,complete,50,naiverer/fast-leworldmodel,,,
123
  fast-lewm,tworoom,25,43,200,200,180,90.0,complete,50,naiverer/fast-leworldmodel,,,
124
  fast-lewm,tworoom,25,44,200,200,173,86.5,complete,50,naiverer/fast-leworldmodel,,,
results/planning_success.json CHANGED
@@ -1604,320 +1604,320 @@
1604
  "dataset": "pusht",
1605
  "offset": 25,
1606
  "seed": 42,
1607
- "samples_completed": 0,
1608
  "samples_required": 200,
1609
- "successes": null,
1610
- "success_percent": null,
1611
- "status": "unavailable: compatible checkpoint not found",
1612
  "execution_budget": 50,
1613
- "checkpoint_release": null,
1614
- "training_epochs": null,
1615
- "training_seed": null,
1616
- "model_variant": null
1617
  },
1618
  {
1619
  "model": "pldm",
1620
  "dataset": "pusht",
1621
  "offset": 25,
1622
  "seed": 43,
1623
- "samples_completed": 0,
1624
  "samples_required": 200,
1625
- "successes": null,
1626
- "success_percent": null,
1627
- "status": "unavailable: compatible checkpoint not found",
1628
  "execution_budget": 50,
1629
- "checkpoint_release": null,
1630
- "training_epochs": null,
1631
- "training_seed": null,
1632
- "model_variant": null
1633
  },
1634
  {
1635
  "model": "pldm",
1636
  "dataset": "pusht",
1637
  "offset": 25,
1638
  "seed": 44,
1639
- "samples_completed": 0,
1640
  "samples_required": 200,
1641
- "successes": null,
1642
- "success_percent": null,
1643
- "status": "unavailable: compatible checkpoint not found",
1644
  "execution_budget": 50,
1645
- "checkpoint_release": null,
1646
- "training_epochs": null,
1647
- "training_seed": null,
1648
- "model_variant": null
1649
  },
1650
  {
1651
  "model": "pldm",
1652
  "dataset": "pusht",
1653
  "offset": 25,
1654
  "seed": 45,
1655
- "samples_completed": 0,
1656
  "samples_required": 200,
1657
- "successes": null,
1658
- "success_percent": null,
1659
- "status": "unavailable: compatible checkpoint not found",
1660
  "execution_budget": 50,
1661
- "checkpoint_release": null,
1662
- "training_epochs": null,
1663
- "training_seed": null,
1664
- "model_variant": null
1665
  },
1666
  {
1667
  "model": "pldm",
1668
  "dataset": "pusht",
1669
  "offset": 25,
1670
  "seed": 46,
1671
- "samples_completed": 0,
1672
  "samples_required": 200,
1673
- "successes": null,
1674
- "success_percent": null,
1675
- "status": "unavailable: compatible checkpoint not found",
1676
  "execution_budget": 50,
1677
- "checkpoint_release": null,
1678
- "training_epochs": null,
1679
- "training_seed": null,
1680
- "model_variant": null
1681
  },
1682
  {
1683
  "model": "pldm",
1684
  "dataset": "pusht",
1685
  "offset": 50,
1686
  "seed": 42,
1687
- "samples_completed": 0,
1688
  "samples_required": 200,
1689
- "successes": null,
1690
- "success_percent": null,
1691
- "status": "unavailable: compatible checkpoint not found",
1692
  "execution_budget": 50,
1693
- "checkpoint_release": null,
1694
- "training_epochs": null,
1695
- "training_seed": null,
1696
- "model_variant": null
1697
  },
1698
  {
1699
  "model": "pldm",
1700
  "dataset": "pusht",
1701
  "offset": 50,
1702
  "seed": 43,
1703
- "samples_completed": 0,
1704
  "samples_required": 200,
1705
- "successes": null,
1706
- "success_percent": null,
1707
- "status": "unavailable: compatible checkpoint not found",
1708
  "execution_budget": 50,
1709
- "checkpoint_release": null,
1710
- "training_epochs": null,
1711
- "training_seed": null,
1712
- "model_variant": null
1713
  },
1714
  {
1715
  "model": "pldm",
1716
  "dataset": "pusht",
1717
  "offset": 50,
1718
  "seed": 44,
1719
- "samples_completed": 0,
1720
  "samples_required": 200,
1721
- "successes": null,
1722
- "success_percent": null,
1723
- "status": "unavailable: compatible checkpoint not found",
1724
  "execution_budget": 50,
1725
- "checkpoint_release": null,
1726
- "training_epochs": null,
1727
- "training_seed": null,
1728
- "model_variant": null
1729
  },
1730
  {
1731
  "model": "pldm",
1732
  "dataset": "pusht",
1733
  "offset": 50,
1734
  "seed": 45,
1735
- "samples_completed": 0,
1736
  "samples_required": 200,
1737
- "successes": null,
1738
- "success_percent": null,
1739
- "status": "unavailable: compatible checkpoint not found",
1740
  "execution_budget": 50,
1741
- "checkpoint_release": null,
1742
- "training_epochs": null,
1743
- "training_seed": null,
1744
- "model_variant": null
1745
  },
1746
  {
1747
  "model": "pldm",
1748
  "dataset": "pusht",
1749
  "offset": 50,
1750
  "seed": 46,
1751
- "samples_completed": 0,
1752
  "samples_required": 200,
1753
- "successes": null,
1754
- "success_percent": null,
1755
- "status": "unavailable: compatible checkpoint not found",
1756
  "execution_budget": 50,
1757
- "checkpoint_release": null,
1758
- "training_epochs": null,
1759
- "training_seed": null,
1760
- "model_variant": null
1761
  },
1762
  {
1763
  "model": "pldm",
1764
  "dataset": "pusht",
1765
  "offset": 75,
1766
  "seed": 42,
1767
- "samples_completed": 0,
1768
  "samples_required": 200,
1769
- "successes": null,
1770
- "success_percent": null,
1771
- "status": "unavailable: compatible checkpoint not found",
1772
  "execution_budget": 50,
1773
- "checkpoint_release": null,
1774
- "training_epochs": null,
1775
- "training_seed": null,
1776
- "model_variant": null
1777
  },
1778
  {
1779
  "model": "pldm",
1780
  "dataset": "pusht",
1781
  "offset": 75,
1782
  "seed": 43,
1783
- "samples_completed": 0,
1784
  "samples_required": 200,
1785
- "successes": null,
1786
- "success_percent": null,
1787
- "status": "unavailable: compatible checkpoint not found",
1788
  "execution_budget": 50,
1789
- "checkpoint_release": null,
1790
- "training_epochs": null,
1791
- "training_seed": null,
1792
- "model_variant": null
1793
  },
1794
  {
1795
  "model": "pldm",
1796
  "dataset": "pusht",
1797
  "offset": 75,
1798
  "seed": 44,
1799
- "samples_completed": 0,
1800
  "samples_required": 200,
1801
- "successes": null,
1802
- "success_percent": null,
1803
- "status": "unavailable: compatible checkpoint not found",
1804
  "execution_budget": 50,
1805
- "checkpoint_release": null,
1806
- "training_epochs": null,
1807
- "training_seed": null,
1808
- "model_variant": null
1809
  },
1810
  {
1811
  "model": "pldm",
1812
  "dataset": "pusht",
1813
  "offset": 75,
1814
  "seed": 45,
1815
- "samples_completed": 0,
1816
  "samples_required": 200,
1817
- "successes": null,
1818
- "success_percent": null,
1819
- "status": "unavailable: compatible checkpoint not found",
1820
  "execution_budget": 50,
1821
- "checkpoint_release": null,
1822
- "training_epochs": null,
1823
- "training_seed": null,
1824
- "model_variant": null
1825
  },
1826
  {
1827
  "model": "pldm",
1828
  "dataset": "pusht",
1829
  "offset": 75,
1830
  "seed": 46,
1831
- "samples_completed": 0,
1832
  "samples_required": 200,
1833
- "successes": null,
1834
- "success_percent": null,
1835
- "status": "unavailable: compatible checkpoint not found",
1836
  "execution_budget": 50,
1837
- "checkpoint_release": null,
1838
- "training_epochs": null,
1839
- "training_seed": null,
1840
- "model_variant": null
1841
  },
1842
  {
1843
  "model": "pldm",
1844
  "dataset": "pusht",
1845
  "offset": 100,
1846
  "seed": 42,
1847
- "samples_completed": 0,
1848
  "samples_required": 200,
1849
- "successes": null,
1850
- "success_percent": null,
1851
- "status": "unavailable: compatible checkpoint not found",
1852
  "execution_budget": 50,
1853
- "checkpoint_release": null,
1854
- "training_epochs": null,
1855
- "training_seed": null,
1856
- "model_variant": null
1857
  },
1858
  {
1859
  "model": "pldm",
1860
  "dataset": "pusht",
1861
  "offset": 100,
1862
  "seed": 43,
1863
- "samples_completed": 0,
1864
  "samples_required": 200,
1865
- "successes": null,
1866
- "success_percent": null,
1867
- "status": "unavailable: compatible checkpoint not found",
1868
  "execution_budget": 50,
1869
- "checkpoint_release": null,
1870
- "training_epochs": null,
1871
- "training_seed": null,
1872
- "model_variant": null
1873
  },
1874
  {
1875
  "model": "pldm",
1876
  "dataset": "pusht",
1877
  "offset": 100,
1878
  "seed": 44,
1879
- "samples_completed": 0,
1880
  "samples_required": 200,
1881
- "successes": null,
1882
- "success_percent": null,
1883
- "status": "unavailable: compatible checkpoint not found",
1884
  "execution_budget": 50,
1885
- "checkpoint_release": null,
1886
- "training_epochs": null,
1887
- "training_seed": null,
1888
- "model_variant": null
1889
  },
1890
  {
1891
  "model": "pldm",
1892
  "dataset": "pusht",
1893
  "offset": 100,
1894
  "seed": 45,
1895
- "samples_completed": 0,
1896
  "samples_required": 200,
1897
- "successes": null,
1898
- "success_percent": null,
1899
- "status": "unavailable: compatible checkpoint not found",
1900
  "execution_budget": 50,
1901
- "checkpoint_release": null,
1902
- "training_epochs": null,
1903
- "training_seed": null,
1904
- "model_variant": null
1905
  },
1906
  {
1907
  "model": "pldm",
1908
  "dataset": "pusht",
1909
  "offset": 100,
1910
  "seed": 46,
1911
- "samples_completed": 0,
1912
  "samples_required": 200,
1913
- "successes": null,
1914
- "success_percent": null,
1915
- "status": "unavailable: compatible checkpoint not found",
1916
  "execution_budget": 50,
1917
- "checkpoint_release": null,
1918
- "training_epochs": null,
1919
- "training_seed": null,
1920
- "model_variant": null
1921
  },
1922
  {
1923
  "model": "fast-lewm",
 
1604
  "dataset": "pusht",
1605
  "offset": 25,
1606
  "seed": 42,
1607
+ "samples_completed": 200,
1608
  "samples_required": 200,
1609
+ "successes": 153,
1610
+ "success_percent": 76.5,
1611
+ "status": "complete",
1612
  "execution_budget": 50,
1613
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1614
+ "training_epochs": 10,
1615
+ "training_seed": 42,
1616
+ "model_variant": "PLDM stable-worldmodel recipe"
1617
  },
1618
  {
1619
  "model": "pldm",
1620
  "dataset": "pusht",
1621
  "offset": 25,
1622
  "seed": 43,
1623
+ "samples_completed": 200,
1624
  "samples_required": 200,
1625
+ "successes": 150,
1626
+ "success_percent": 75.0,
1627
+ "status": "complete",
1628
  "execution_budget": 50,
1629
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1630
+ "training_epochs": 10,
1631
+ "training_seed": 42,
1632
+ "model_variant": "PLDM stable-worldmodel recipe"
1633
  },
1634
  {
1635
  "model": "pldm",
1636
  "dataset": "pusht",
1637
  "offset": 25,
1638
  "seed": 44,
1639
+ "samples_completed": 200,
1640
  "samples_required": 200,
1641
+ "successes": 144,
1642
+ "success_percent": 72.0,
1643
+ "status": "complete",
1644
  "execution_budget": 50,
1645
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1646
+ "training_epochs": 10,
1647
+ "training_seed": 42,
1648
+ "model_variant": "PLDM stable-worldmodel recipe"
1649
  },
1650
  {
1651
  "model": "pldm",
1652
  "dataset": "pusht",
1653
  "offset": 25,
1654
  "seed": 45,
1655
+ "samples_completed": 200,
1656
  "samples_required": 200,
1657
+ "successes": 154,
1658
+ "success_percent": 77.0,
1659
+ "status": "complete",
1660
  "execution_budget": 50,
1661
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1662
+ "training_epochs": 10,
1663
+ "training_seed": 42,
1664
+ "model_variant": "PLDM stable-worldmodel recipe"
1665
  },
1666
  {
1667
  "model": "pldm",
1668
  "dataset": "pusht",
1669
  "offset": 25,
1670
  "seed": 46,
1671
+ "samples_completed": 200,
1672
  "samples_required": 200,
1673
+ "successes": 152,
1674
+ "success_percent": 76.0,
1675
+ "status": "complete",
1676
  "execution_budget": 50,
1677
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1678
+ "training_epochs": 10,
1679
+ "training_seed": 42,
1680
+ "model_variant": "PLDM stable-worldmodel recipe"
1681
  },
1682
  {
1683
  "model": "pldm",
1684
  "dataset": "pusht",
1685
  "offset": 50,
1686
  "seed": 42,
1687
+ "samples_completed": 200,
1688
  "samples_required": 200,
1689
+ "successes": 56,
1690
+ "success_percent": 28.0,
1691
+ "status": "complete",
1692
  "execution_budget": 50,
1693
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1694
+ "training_epochs": 10,
1695
+ "training_seed": 42,
1696
+ "model_variant": "PLDM stable-worldmodel recipe"
1697
  },
1698
  {
1699
  "model": "pldm",
1700
  "dataset": "pusht",
1701
  "offset": 50,
1702
  "seed": 43,
1703
+ "samples_completed": 200,
1704
  "samples_required": 200,
1705
+ "successes": 41,
1706
+ "success_percent": 20.5,
1707
+ "status": "complete",
1708
  "execution_budget": 50,
1709
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1710
+ "training_epochs": 10,
1711
+ "training_seed": 42,
1712
+ "model_variant": "PLDM stable-worldmodel recipe"
1713
  },
1714
  {
1715
  "model": "pldm",
1716
  "dataset": "pusht",
1717
  "offset": 50,
1718
  "seed": 44,
1719
+ "samples_completed": 200,
1720
  "samples_required": 200,
1721
+ "successes": 61,
1722
+ "success_percent": 30.5,
1723
+ "status": "complete",
1724
  "execution_budget": 50,
1725
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1726
+ "training_epochs": 10,
1727
+ "training_seed": 42,
1728
+ "model_variant": "PLDM stable-worldmodel recipe"
1729
  },
1730
  {
1731
  "model": "pldm",
1732
  "dataset": "pusht",
1733
  "offset": 50,
1734
  "seed": 45,
1735
+ "samples_completed": 200,
1736
  "samples_required": 200,
1737
+ "successes": 48,
1738
+ "success_percent": 24.0,
1739
+ "status": "complete",
1740
  "execution_budget": 50,
1741
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1742
+ "training_epochs": 10,
1743
+ "training_seed": 42,
1744
+ "model_variant": "PLDM stable-worldmodel recipe"
1745
  },
1746
  {
1747
  "model": "pldm",
1748
  "dataset": "pusht",
1749
  "offset": 50,
1750
  "seed": 46,
1751
+ "samples_completed": 200,
1752
  "samples_required": 200,
1753
+ "successes": 59,
1754
+ "success_percent": 29.5,
1755
+ "status": "complete",
1756
  "execution_budget": 50,
1757
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1758
+ "training_epochs": 10,
1759
+ "training_seed": 42,
1760
+ "model_variant": "PLDM stable-worldmodel recipe"
1761
  },
1762
  {
1763
  "model": "pldm",
1764
  "dataset": "pusht",
1765
  "offset": 75,
1766
  "seed": 42,
1767
+ "samples_completed": 200,
1768
  "samples_required": 200,
1769
+ "successes": 27,
1770
+ "success_percent": 13.5,
1771
+ "status": "complete",
1772
  "execution_budget": 50,
1773
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1774
+ "training_epochs": 10,
1775
+ "training_seed": 42,
1776
+ "model_variant": "PLDM stable-worldmodel recipe"
1777
  },
1778
  {
1779
  "model": "pldm",
1780
  "dataset": "pusht",
1781
  "offset": 75,
1782
  "seed": 43,
1783
+ "samples_completed": 200,
1784
  "samples_required": 200,
1785
+ "successes": 19,
1786
+ "success_percent": 9.5,
1787
+ "status": "complete",
1788
  "execution_budget": 50,
1789
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1790
+ "training_epochs": 10,
1791
+ "training_seed": 42,
1792
+ "model_variant": "PLDM stable-worldmodel recipe"
1793
  },
1794
  {
1795
  "model": "pldm",
1796
  "dataset": "pusht",
1797
  "offset": 75,
1798
  "seed": 44,
1799
+ "samples_completed": 200,
1800
  "samples_required": 200,
1801
+ "successes": 28,
1802
+ "success_percent": 14.0,
1803
+ "status": "complete",
1804
  "execution_budget": 50,
1805
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1806
+ "training_epochs": 10,
1807
+ "training_seed": 42,
1808
+ "model_variant": "PLDM stable-worldmodel recipe"
1809
  },
1810
  {
1811
  "model": "pldm",
1812
  "dataset": "pusht",
1813
  "offset": 75,
1814
  "seed": 45,
1815
+ "samples_completed": 200,
1816
  "samples_required": 200,
1817
+ "successes": 22,
1818
+ "success_percent": 11.0,
1819
+ "status": "complete",
1820
  "execution_budget": 50,
1821
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1822
+ "training_epochs": 10,
1823
+ "training_seed": 42,
1824
+ "model_variant": "PLDM stable-worldmodel recipe"
1825
  },
1826
  {
1827
  "model": "pldm",
1828
  "dataset": "pusht",
1829
  "offset": 75,
1830
  "seed": 46,
1831
+ "samples_completed": 200,
1832
  "samples_required": 200,
1833
+ "successes": 27,
1834
+ "success_percent": 13.5,
1835
+ "status": "complete",
1836
  "execution_budget": 50,
1837
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1838
+ "training_epochs": 10,
1839
+ "training_seed": 42,
1840
+ "model_variant": "PLDM stable-worldmodel recipe"
1841
  },
1842
  {
1843
  "model": "pldm",
1844
  "dataset": "pusht",
1845
  "offset": 100,
1846
  "seed": 42,
1847
+ "samples_completed": 200,
1848
  "samples_required": 200,
1849
+ "successes": 13,
1850
+ "success_percent": 6.5,
1851
+ "status": "complete",
1852
  "execution_budget": 50,
1853
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1854
+ "training_epochs": 10,
1855
+ "training_seed": 42,
1856
+ "model_variant": "PLDM stable-worldmodel recipe"
1857
  },
1858
  {
1859
  "model": "pldm",
1860
  "dataset": "pusht",
1861
  "offset": 100,
1862
  "seed": 43,
1863
+ "samples_completed": 200,
1864
  "samples_required": 200,
1865
+ "successes": 13,
1866
+ "success_percent": 6.5,
1867
+ "status": "complete",
1868
  "execution_budget": 50,
1869
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1870
+ "training_epochs": 10,
1871
+ "training_seed": 42,
1872
+ "model_variant": "PLDM stable-worldmodel recipe"
1873
  },
1874
  {
1875
  "model": "pldm",
1876
  "dataset": "pusht",
1877
  "offset": 100,
1878
  "seed": 44,
1879
+ "samples_completed": 200,
1880
  "samples_required": 200,
1881
+ "successes": 9,
1882
+ "success_percent": 4.5,
1883
+ "status": "complete",
1884
  "execution_budget": 50,
1885
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1886
+ "training_epochs": 10,
1887
+ "training_seed": 42,
1888
+ "model_variant": "PLDM stable-worldmodel recipe"
1889
  },
1890
  {
1891
  "model": "pldm",
1892
  "dataset": "pusht",
1893
  "offset": 100,
1894
  "seed": 45,
1895
+ "samples_completed": 200,
1896
  "samples_required": 200,
1897
+ "successes": 12,
1898
+ "success_percent": 6.0,
1899
+ "status": "complete",
1900
  "execution_budget": 50,
1901
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1902
+ "training_epochs": 10,
1903
+ "training_seed": 42,
1904
+ "model_variant": "PLDM stable-worldmodel recipe"
1905
  },
1906
  {
1907
  "model": "pldm",
1908
  "dataset": "pusht",
1909
  "offset": 100,
1910
  "seed": 46,
1911
+ "samples_completed": 200,
1912
  "samples_required": 200,
1913
+ "successes": 15,
1914
+ "success_percent": 7.5,
1915
+ "status": "complete",
1916
  "execution_budget": 50,
1917
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1918
+ "training_epochs": 10,
1919
+ "training_seed": 42,
1920
+ "model_variant": "PLDM stable-worldmodel recipe"
1921
  },
1922
  {
1923
  "model": "fast-lewm",
results/planning_success.xlsx CHANGED
Binary files a/results/planning_success.xlsx and b/results/planning_success.xlsx differ
 
results/planning_summary.csv CHANGED
@@ -19,10 +19,10 @@ pldm,cube,25,55.0,52.0,56.0,56.5,66.0,5,57.1,5.272570530585627,57.10 ± 5.27
19
  pldm,cube,50,49.5,39.0,41.5,44.5,50.5,5,45.0,4.9749371855331,45.00 ± 4.97
20
  pldm,cube,75,54.5,58.5,57.0,54.5,51.5,5,55.2,2.6832815729997477,55.20 ± 2.68
21
  pldm,cube,100,57.5,58.0,51.5,54.0,53.0,5,54.8,2.8416544476765644,54.80 ± 2.84
22
- pldm,pusht,25,,,,,,0,,,
23
- pldm,pusht,50,,,,,,0,,,
24
- pldm,pusht,75,,,,,,0,,,
25
- pldm,pusht,100,,,,,,0,,,
26
  fast-lewm,tworoom,25,88.0,90.0,86.5,88.5,91.0,5,88.8,1.753567791675018,88.80 ± 1.75
27
  fast-lewm,tworoom,50,64.0,58.0,59.0,63.5,66.0,5,62.1,3.4351128074635335,62.10 ± 3.44
28
  fast-lewm,tworoom,75,44.5,33.5,35.5,34.0,38.0,5,37.1,4.491658936295141,37.10 ± 4.49
 
19
  pldm,cube,50,49.5,39.0,41.5,44.5,50.5,5,45.0,4.9749371855331,45.00 ± 4.97
20
  pldm,cube,75,54.5,58.5,57.0,54.5,51.5,5,55.2,2.6832815729997477,55.20 ± 2.68
21
  pldm,cube,100,57.5,58.0,51.5,54.0,53.0,5,54.8,2.8416544476765644,54.80 ± 2.84
22
+ pldm,pusht,25,76.5,75.0,72.0,77.0,76.0,5,75.3,1.9874606914351791,75.30 ± 1.99
23
+ pldm,pusht,50,28.0,20.5,30.5,24.0,29.5,5,26.5,4.168333000133266,26.50 ± 4.17
24
+ pldm,pusht,75,13.5,9.5,14.0,11.0,13.5,5,12.3,1.9557607215607946,12.30 ± 1.96
25
+ pldm,pusht,100,6.5,6.5,4.5,6.0,7.5,5,6.2,1.0954451150103321,6.20 ± 1.10
26
  fast-lewm,tworoom,25,88.0,90.0,86.5,88.5,91.0,5,88.8,1.753567791675018,88.80 ± 1.75
27
  fast-lewm,tworoom,50,64.0,58.0,59.0,63.5,66.0,5,62.1,3.4351128074635335,62.10 ± 3.44
28
  fast-lewm,tworoom,75,44.5,33.5,35.5,34.0,38.0,5,37.1,4.491658936295141,37.10 ± 4.49
results/planning_summary.json CHANGED
@@ -283,57 +283,57 @@
283
  "model": "pldm",
284
  "dataset": "pusht",
285
  "offset": 25,
286
- "seed_42": null,
287
- "seed_43": null,
288
- "seed_44": null,
289
- "seed_45": null,
290
- "seed_46": null,
291
- "seeds_completed": 0,
292
- "mean_percent": null,
293
- "std_percent": null,
294
- "mean_plus_std": null
295
  },
296
  {
297
  "model": "pldm",
298
  "dataset": "pusht",
299
  "offset": 50,
300
- "seed_42": null,
301
- "seed_43": null,
302
- "seed_44": null,
303
- "seed_45": null,
304
- "seed_46": null,
305
- "seeds_completed": 0,
306
- "mean_percent": null,
307
- "std_percent": null,
308
- "mean_plus_std": null
309
  },
310
  {
311
  "model": "pldm",
312
  "dataset": "pusht",
313
  "offset": 75,
314
- "seed_42": null,
315
- "seed_43": null,
316
- "seed_44": null,
317
- "seed_45": null,
318
- "seed_46": null,
319
- "seeds_completed": 0,
320
- "mean_percent": null,
321
- "std_percent": null,
322
- "mean_plus_std": null
323
  },
324
  {
325
  "model": "pldm",
326
  "dataset": "pusht",
327
  "offset": 100,
328
- "seed_42": null,
329
- "seed_43": null,
330
- "seed_44": null,
331
- "seed_45": null,
332
- "seed_46": null,
333
- "seeds_completed": 0,
334
- "mean_percent": null,
335
- "std_percent": null,
336
- "mean_plus_std": null
337
  },
338
  {
339
  "model": "fast-lewm",
 
283
  "model": "pldm",
284
  "dataset": "pusht",
285
  "offset": 25,
286
+ "seed_42": 76.5,
287
+ "seed_43": 75.0,
288
+ "seed_44": 72.0,
289
+ "seed_45": 77.0,
290
+ "seed_46": 76.0,
291
+ "seeds_completed": 5,
292
+ "mean_percent": 75.3,
293
+ "std_percent": 1.9874606914351791,
294
+ "mean_plus_std": "75.30 \u00b1 1.99"
295
  },
296
  {
297
  "model": "pldm",
298
  "dataset": "pusht",
299
  "offset": 50,
300
+ "seed_42": 28.0,
301
+ "seed_43": 20.5,
302
+ "seed_44": 30.5,
303
+ "seed_45": 24.0,
304
+ "seed_46": 29.5,
305
+ "seeds_completed": 5,
306
+ "mean_percent": 26.5,
307
+ "std_percent": 4.168333000133266,
308
+ "mean_plus_std": "26.50 \u00b1 4.17"
309
  },
310
  {
311
  "model": "pldm",
312
  "dataset": "pusht",
313
  "offset": 75,
314
+ "seed_42": 13.5,
315
+ "seed_43": 9.5,
316
+ "seed_44": 14.0,
317
+ "seed_45": 11.0,
318
+ "seed_46": 13.5,
319
+ "seeds_completed": 5,
320
+ "mean_percent": 12.3,
321
+ "std_percent": 1.9557607215607946,
322
+ "mean_plus_std": "12.30 \u00b1 1.96"
323
  },
324
  {
325
  "model": "pldm",
326
  "dataset": "pusht",
327
  "offset": 100,
328
+ "seed_42": 6.5,
329
+ "seed_43": 6.5,
330
+ "seed_44": 4.5,
331
+ "seed_45": 6.0,
332
+ "seed_46": 7.5,
333
+ "seeds_completed": 5,
334
+ "mean_percent": 6.2,
335
+ "std_percent": 1.0954451150103321,
336
+ "mean_plus_std": "6.20 \u00b1 1.10"
337
  },
338
  {
339
  "model": "fast-lewm",
results/pldm_pusht_100.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 100,
5
+ "seed": 42,
6
+ "successes": [
7
+ false,
8
+ false,
9
+ false,
10
+ false,
11
+ false,
12
+ true,
13
+ false,
14
+ false,
15
+ false,
16
+ false,
17
+ true,
18
+ false,
19
+ false,
20
+ false,
21
+ false,
22
+ false,
23
+ false,
24
+ false,
25
+ false,
26
+ false,
27
+ false,
28
+ false,
29
+ false,
30
+ false,
31
+ false,
32
+ false,
33
+ false,
34
+ false,
35
+ false,
36
+ false,
37
+ false,
38
+ false,
39
+ false,
40
+ false,
41
+ false,
42
+ false,
43
+ false,
44
+ true,
45
+ false,
46
+ false,
47
+ false,
48
+ false,
49
+ false,
50
+ false,
51
+ false,
52
+ false,
53
+ false,
54
+ false,
55
+ false,
56
+ false,
57
+ false,
58
+ false,
59
+ false,
60
+ false,
61
+ false,
62
+ false,
63
+ false,
64
+ false,
65
+ false,
66
+ false,
67
+ true,
68
+ false,
69
+ false,
70
+ false,
71
+ false,
72
+ false,
73
+ false,
74
+ false,
75
+ false,
76
+ false,
77
+ false,
78
+ false,
79
+ false,
80
+ false,
81
+ false,
82
+ false,
83
+ false,
84
+ false,
85
+ false,
86
+ false,
87
+ false,
88
+ false,
89
+ false,
90
+ false,
91
+ false,
92
+ false,
93
+ false,
94
+ false,
95
+ false,
96
+ false,
97
+ false,
98
+ false,
99
+ false,
100
+ false,
101
+ false,
102
+ false,
103
+ true,
104
+ false,
105
+ false,
106
+ false,
107
+ false,
108
+ false,
109
+ false,
110
+ false,
111
+ false,
112
+ false,
113
+ false,
114
+ false,
115
+ false,
116
+ false,
117
+ false,
118
+ false,
119
+ false,
120
+ false,
121
+ false,
122
+ false,
123
+ false,
124
+ false,
125
+ false,
126
+ false,
127
+ true,
128
+ false,
129
+ true,
130
+ false,
131
+ false,
132
+ false,
133
+ false,
134
+ false,
135
+ false,
136
+ false,
137
+ false,
138
+ false,
139
+ false,
140
+ false,
141
+ false,
142
+ false,
143
+ false,
144
+ false,
145
+ false,
146
+ false,
147
+ false,
148
+ false,
149
+ false,
150
+ true,
151
+ false,
152
+ false,
153
+ false,
154
+ false,
155
+ false,
156
+ false,
157
+ true,
158
+ false,
159
+ false,
160
+ false,
161
+ false,
162
+ false,
163
+ false,
164
+ false,
165
+ false,
166
+ false,
167
+ false,
168
+ false,
169
+ false,
170
+ false,
171
+ false,
172
+ false,
173
+ false,
174
+ false,
175
+ false,
176
+ false,
177
+ false,
178
+ false,
179
+ false,
180
+ true,
181
+ false,
182
+ true,
183
+ false,
184
+ false,
185
+ false,
186
+ false,
187
+ false,
188
+ false,
189
+ false,
190
+ false,
191
+ false,
192
+ false,
193
+ false,
194
+ true,
195
+ false,
196
+ true,
197
+ false,
198
+ false,
199
+ false,
200
+ false,
201
+ false,
202
+ false,
203
+ false,
204
+ false,
205
+ false,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 161.83232998847961,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_100_seed43.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 100,
5
+ "seed": 43,
6
+ "successes": [
7
+ false,
8
+ false,
9
+ false,
10
+ false,
11
+ false,
12
+ false,
13
+ false,
14
+ false,
15
+ false,
16
+ false,
17
+ false,
18
+ false,
19
+ false,
20
+ false,
21
+ false,
22
+ false,
23
+ false,
24
+ false,
25
+ false,
26
+ false,
27
+ false,
28
+ false,
29
+ false,
30
+ false,
31
+ false,
32
+ false,
33
+ false,
34
+ false,
35
+ false,
36
+ false,
37
+ false,
38
+ false,
39
+ false,
40
+ false,
41
+ false,
42
+ false,
43
+ false,
44
+ false,
45
+ false,
46
+ false,
47
+ false,
48
+ false,
49
+ false,
50
+ false,
51
+ false,
52
+ false,
53
+ false,
54
+ true,
55
+ false,
56
+ false,
57
+ false,
58
+ false,
59
+ false,
60
+ false,
61
+ false,
62
+ false,
63
+ true,
64
+ false,
65
+ false,
66
+ false,
67
+ false,
68
+ false,
69
+ true,
70
+ false,
71
+ false,
72
+ false,
73
+ false,
74
+ false,
75
+ false,
76
+ false,
77
+ false,
78
+ false,
79
+ false,
80
+ false,
81
+ false,
82
+ false,
83
+ false,
84
+ false,
85
+ false,
86
+ false,
87
+ false,
88
+ false,
89
+ false,
90
+ false,
91
+ false,
92
+ false,
93
+ false,
94
+ false,
95
+ false,
96
+ false,
97
+ false,
98
+ false,
99
+ false,
100
+ false,
101
+ false,
102
+ false,
103
+ false,
104
+ false,
105
+ false,
106
+ false,
107
+ false,
108
+ false,
109
+ false,
110
+ false,
111
+ false,
112
+ false,
113
+ false,
114
+ false,
115
+ false,
116
+ false,
117
+ false,
118
+ true,
119
+ false,
120
+ true,
121
+ true,
122
+ false,
123
+ false,
124
+ false,
125
+ false,
126
+ true,
127
+ false,
128
+ true,
129
+ false,
130
+ false,
131
+ false,
132
+ true,
133
+ false,
134
+ false,
135
+ false,
136
+ false,
137
+ false,
138
+ false,
139
+ false,
140
+ false,
141
+ false,
142
+ false,
143
+ false,
144
+ false,
145
+ false,
146
+ false,
147
+ false,
148
+ false,
149
+ false,
150
+ false,
151
+ false,
152
+ false,
153
+ false,
154
+ false,
155
+ false,
156
+ false,
157
+ false,
158
+ false,
159
+ false,
160
+ false,
161
+ false,
162
+ false,
163
+ false,
164
+ false,
165
+ false,
166
+ false,
167
+ true,
168
+ false,
169
+ false,
170
+ false,
171
+ false,
172
+ false,
173
+ false,
174
+ false,
175
+ false,
176
+ false,
177
+ false,
178
+ false,
179
+ true,
180
+ false,
181
+ false,
182
+ false,
183
+ false,
184
+ false,
185
+ true,
186
+ false,
187
+ false,
188
+ false,
189
+ false,
190
+ false,
191
+ false,
192
+ false,
193
+ false,
194
+ false,
195
+ false,
196
+ false,
197
+ false,
198
+ false,
199
+ false,
200
+ false,
201
+ false,
202
+ false,
203
+ true,
204
+ false,
205
+ false,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 161.93198895454407,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_100_seed44.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 100,
5
+ "seed": 44,
6
+ "successes": [
7
+ false,
8
+ false,
9
+ false,
10
+ false,
11
+ false,
12
+ false,
13
+ false,
14
+ false,
15
+ false,
16
+ false,
17
+ false,
18
+ false,
19
+ false,
20
+ false,
21
+ false,
22
+ false,
23
+ false,
24
+ false,
25
+ false,
26
+ false,
27
+ false,
28
+ false,
29
+ false,
30
+ true,
31
+ false,
32
+ false,
33
+ false,
34
+ false,
35
+ false,
36
+ false,
37
+ false,
38
+ false,
39
+ false,
40
+ false,
41
+ false,
42
+ false,
43
+ false,
44
+ false,
45
+ false,
46
+ false,
47
+ false,
48
+ false,
49
+ false,
50
+ false,
51
+ false,
52
+ false,
53
+ false,
54
+ false,
55
+ false,
56
+ false,
57
+ false,
58
+ false,
59
+ false,
60
+ false,
61
+ false,
62
+ false,
63
+ false,
64
+ false,
65
+ false,
66
+ false,
67
+ false,
68
+ false,
69
+ true,
70
+ false,
71
+ false,
72
+ false,
73
+ false,
74
+ false,
75
+ false,
76
+ false,
77
+ false,
78
+ false,
79
+ false,
80
+ false,
81
+ false,
82
+ false,
83
+ false,
84
+ false,
85
+ false,
86
+ false,
87
+ false,
88
+ false,
89
+ true,
90
+ false,
91
+ false,
92
+ false,
93
+ false,
94
+ false,
95
+ false,
96
+ false,
97
+ false,
98
+ false,
99
+ false,
100
+ false,
101
+ false,
102
+ false,
103
+ false,
104
+ false,
105
+ false,
106
+ false,
107
+ false,
108
+ true,
109
+ false,
110
+ false,
111
+ false,
112
+ false,
113
+ false,
114
+ false,
115
+ false,
116
+ false,
117
+ false,
118
+ false,
119
+ false,
120
+ false,
121
+ false,
122
+ false,
123
+ false,
124
+ false,
125
+ false,
126
+ false,
127
+ true,
128
+ false,
129
+ false,
130
+ false,
131
+ false,
132
+ false,
133
+ false,
134
+ false,
135
+ false,
136
+ false,
137
+ false,
138
+ false,
139
+ false,
140
+ false,
141
+ false,
142
+ false,
143
+ false,
144
+ false,
145
+ false,
146
+ false,
147
+ false,
148
+ false,
149
+ false,
150
+ false,
151
+ false,
152
+ false,
153
+ false,
154
+ false,
155
+ false,
156
+ false,
157
+ false,
158
+ false,
159
+ false,
160
+ false,
161
+ false,
162
+ false,
163
+ false,
164
+ false,
165
+ false,
166
+ false,
167
+ false,
168
+ false,
169
+ false,
170
+ false,
171
+ false,
172
+ false,
173
+ false,
174
+ false,
175
+ false,
176
+ false,
177
+ false,
178
+ false,
179
+ false,
180
+ false,
181
+ false,
182
+ false,
183
+ false,
184
+ false,
185
+ false,
186
+ true,
187
+ true,
188
+ false,
189
+ false,
190
+ true,
191
+ false,
192
+ true,
193
+ false,
194
+ false,
195
+ false,
196
+ false,
197
+ false,
198
+ false,
199
+ false,
200
+ false,
201
+ false,
202
+ false,
203
+ false,
204
+ false,
205
+ false,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 163.49373078346252,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_100_seed45.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 100,
5
+ "seed": 45,
6
+ "successes": [
7
+ false,
8
+ false,
9
+ false,
10
+ false,
11
+ false,
12
+ false,
13
+ false,
14
+ false,
15
+ false,
16
+ false,
17
+ false,
18
+ false,
19
+ false,
20
+ false,
21
+ false,
22
+ false,
23
+ false,
24
+ false,
25
+ false,
26
+ false,
27
+ false,
28
+ false,
29
+ false,
30
+ false,
31
+ false,
32
+ false,
33
+ true,
34
+ false,
35
+ false,
36
+ false,
37
+ false,
38
+ false,
39
+ false,
40
+ false,
41
+ false,
42
+ false,
43
+ false,
44
+ false,
45
+ false,
46
+ false,
47
+ false,
48
+ false,
49
+ false,
50
+ false,
51
+ false,
52
+ false,
53
+ false,
54
+ false,
55
+ false,
56
+ true,
57
+ false,
58
+ false,
59
+ false,
60
+ false,
61
+ false,
62
+ false,
63
+ false,
64
+ false,
65
+ false,
66
+ false,
67
+ false,
68
+ false,
69
+ false,
70
+ false,
71
+ false,
72
+ false,
73
+ false,
74
+ false,
75
+ false,
76
+ false,
77
+ false,
78
+ false,
79
+ false,
80
+ false,
81
+ false,
82
+ false,
83
+ false,
84
+ false,
85
+ false,
86
+ false,
87
+ false,
88
+ false,
89
+ false,
90
+ false,
91
+ false,
92
+ false,
93
+ false,
94
+ false,
95
+ false,
96
+ false,
97
+ false,
98
+ false,
99
+ false,
100
+ false,
101
+ false,
102
+ false,
103
+ false,
104
+ false,
105
+ false,
106
+ false,
107
+ false,
108
+ true,
109
+ false,
110
+ false,
111
+ false,
112
+ false,
113
+ false,
114
+ true,
115
+ false,
116
+ false,
117
+ false,
118
+ false,
119
+ false,
120
+ false,
121
+ false,
122
+ false,
123
+ false,
124
+ true,
125
+ true,
126
+ false,
127
+ false,
128
+ false,
129
+ false,
130
+ false,
131
+ false,
132
+ false,
133
+ false,
134
+ false,
135
+ false,
136
+ false,
137
+ false,
138
+ false,
139
+ false,
140
+ false,
141
+ false,
142
+ false,
143
+ false,
144
+ false,
145
+ false,
146
+ false,
147
+ false,
148
+ false,
149
+ false,
150
+ false,
151
+ false,
152
+ true,
153
+ false,
154
+ false,
155
+ false,
156
+ false,
157
+ false,
158
+ false,
159
+ false,
160
+ false,
161
+ false,
162
+ false,
163
+ false,
164
+ false,
165
+ false,
166
+ false,
167
+ false,
168
+ false,
169
+ false,
170
+ false,
171
+ false,
172
+ false,
173
+ false,
174
+ false,
175
+ false,
176
+ false,
177
+ false,
178
+ true,
179
+ false,
180
+ false,
181
+ false,
182
+ true,
183
+ false,
184
+ false,
185
+ false,
186
+ false,
187
+ false,
188
+ false,
189
+ false,
190
+ false,
191
+ false,
192
+ false,
193
+ false,
194
+ false,
195
+ false,
196
+ true,
197
+ true,
198
+ false,
199
+ false,
200
+ false,
201
+ true,
202
+ false,
203
+ false,
204
+ false,
205
+ false,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 160.739990234375,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_100_seed46.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 100,
5
+ "seed": 46,
6
+ "successes": [
7
+ false,
8
+ false,
9
+ false,
10
+ false,
11
+ false,
12
+ true,
13
+ false,
14
+ false,
15
+ false,
16
+ false,
17
+ false,
18
+ true,
19
+ false,
20
+ false,
21
+ false,
22
+ false,
23
+ false,
24
+ false,
25
+ false,
26
+ false,
27
+ false,
28
+ false,
29
+ false,
30
+ false,
31
+ false,
32
+ false,
33
+ false,
34
+ false,
35
+ false,
36
+ false,
37
+ false,
38
+ false,
39
+ false,
40
+ false,
41
+ false,
42
+ false,
43
+ false,
44
+ false,
45
+ false,
46
+ false,
47
+ false,
48
+ false,
49
+ false,
50
+ false,
51
+ false,
52
+ false,
53
+ false,
54
+ false,
55
+ false,
56
+ false,
57
+ false,
58
+ false,
59
+ false,
60
+ false,
61
+ false,
62
+ false,
63
+ false,
64
+ false,
65
+ false,
66
+ false,
67
+ false,
68
+ false,
69
+ false,
70
+ false,
71
+ false,
72
+ false,
73
+ false,
74
+ true,
75
+ false,
76
+ false,
77
+ false,
78
+ false,
79
+ false,
80
+ false,
81
+ false,
82
+ false,
83
+ false,
84
+ false,
85
+ false,
86
+ false,
87
+ false,
88
+ false,
89
+ false,
90
+ false,
91
+ false,
92
+ false,
93
+ false,
94
+ false,
95
+ false,
96
+ false,
97
+ false,
98
+ false,
99
+ false,
100
+ false,
101
+ false,
102
+ false,
103
+ false,
104
+ false,
105
+ false,
106
+ false,
107
+ false,
108
+ false,
109
+ false,
110
+ false,
111
+ false,
112
+ false,
113
+ false,
114
+ false,
115
+ false,
116
+ false,
117
+ false,
118
+ true,
119
+ false,
120
+ false,
121
+ false,
122
+ false,
123
+ false,
124
+ false,
125
+ false,
126
+ false,
127
+ false,
128
+ false,
129
+ true,
130
+ false,
131
+ false,
132
+ false,
133
+ false,
134
+ false,
135
+ false,
136
+ false,
137
+ false,
138
+ false,
139
+ true,
140
+ false,
141
+ false,
142
+ false,
143
+ false,
144
+ false,
145
+ false,
146
+ false,
147
+ false,
148
+ false,
149
+ false,
150
+ false,
151
+ true,
152
+ false,
153
+ false,
154
+ false,
155
+ false,
156
+ false,
157
+ false,
158
+ false,
159
+ false,
160
+ true,
161
+ true,
162
+ false,
163
+ false,
164
+ false,
165
+ false,
166
+ false,
167
+ false,
168
+ true,
169
+ false,
170
+ false,
171
+ false,
172
+ false,
173
+ false,
174
+ false,
175
+ false,
176
+ false,
177
+ false,
178
+ false,
179
+ false,
180
+ false,
181
+ false,
182
+ false,
183
+ false,
184
+ false,
185
+ true,
186
+ false,
187
+ false,
188
+ false,
189
+ false,
190
+ true,
191
+ false,
192
+ false,
193
+ true,
194
+ false,
195
+ false,
196
+ false,
197
+ false,
198
+ false,
199
+ false,
200
+ false,
201
+ true,
202
+ true,
203
+ false,
204
+ false,
205
+ false,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 161.83216452598572,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_25.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 25,
5
+ "seed": 42,
6
+ "successes": [
7
+ true,
8
+ true,
9
+ true,
10
+ true,
11
+ true,
12
+ true,
13
+ true,
14
+ false,
15
+ true,
16
+ true,
17
+ false,
18
+ true,
19
+ true,
20
+ true,
21
+ true,
22
+ false,
23
+ false,
24
+ false,
25
+ false,
26
+ true,
27
+ true,
28
+ true,
29
+ true,
30
+ true,
31
+ true,
32
+ true,
33
+ true,
34
+ true,
35
+ true,
36
+ true,
37
+ true,
38
+ true,
39
+ true,
40
+ true,
41
+ true,
42
+ false,
43
+ false,
44
+ false,
45
+ true,
46
+ true,
47
+ false,
48
+ true,
49
+ true,
50
+ false,
51
+ false,
52
+ false,
53
+ true,
54
+ true,
55
+ true,
56
+ false,
57
+ true,
58
+ true,
59
+ true,
60
+ true,
61
+ false,
62
+ true,
63
+ true,
64
+ false,
65
+ true,
66
+ true,
67
+ false,
68
+ true,
69
+ true,
70
+ true,
71
+ true,
72
+ true,
73
+ false,
74
+ false,
75
+ true,
76
+ true,
77
+ false,
78
+ true,
79
+ true,
80
+ true,
81
+ true,
82
+ true,
83
+ true,
84
+ true,
85
+ true,
86
+ true,
87
+ true,
88
+ true,
89
+ true,
90
+ true,
91
+ true,
92
+ true,
93
+ true,
94
+ true,
95
+ false,
96
+ false,
97
+ true,
98
+ true,
99
+ true,
100
+ true,
101
+ true,
102
+ true,
103
+ true,
104
+ true,
105
+ false,
106
+ false,
107
+ true,
108
+ true,
109
+ false,
110
+ true,
111
+ true,
112
+ true,
113
+ true,
114
+ false,
115
+ true,
116
+ true,
117
+ true,
118
+ false,
119
+ true,
120
+ true,
121
+ true,
122
+ true,
123
+ true,
124
+ true,
125
+ true,
126
+ true,
127
+ true,
128
+ true,
129
+ true,
130
+ true,
131
+ true,
132
+ true,
133
+ true,
134
+ true,
135
+ true,
136
+ true,
137
+ true,
138
+ true,
139
+ true,
140
+ false,
141
+ true,
142
+ true,
143
+ true,
144
+ true,
145
+ true,
146
+ false,
147
+ true,
148
+ true,
149
+ false,
150
+ true,
151
+ false,
152
+ true,
153
+ false,
154
+ false,
155
+ true,
156
+ true,
157
+ false,
158
+ true,
159
+ false,
160
+ true,
161
+ true,
162
+ true,
163
+ true,
164
+ true,
165
+ true,
166
+ true,
167
+ false,
168
+ true,
169
+ true,
170
+ true,
171
+ true,
172
+ true,
173
+ true,
174
+ true,
175
+ true,
176
+ false,
177
+ false,
178
+ false,
179
+ true,
180
+ true,
181
+ true,
182
+ true,
183
+ false,
184
+ true,
185
+ true,
186
+ true,
187
+ false,
188
+ false,
189
+ true,
190
+ true,
191
+ false,
192
+ true,
193
+ true,
194
+ false,
195
+ true,
196
+ true,
197
+ true,
198
+ true,
199
+ true,
200
+ false,
201
+ true,
202
+ false,
203
+ false,
204
+ true,
205
+ true,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 106.69858884811401,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_25_seed43.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 25,
5
+ "seed": 43,
6
+ "successes": [
7
+ true,
8
+ true,
9
+ true,
10
+ true,
11
+ true,
12
+ true,
13
+ true,
14
+ true,
15
+ true,
16
+ true,
17
+ true,
18
+ true,
19
+ true,
20
+ false,
21
+ true,
22
+ false,
23
+ true,
24
+ true,
25
+ false,
26
+ false,
27
+ false,
28
+ true,
29
+ true,
30
+ false,
31
+ false,
32
+ true,
33
+ false,
34
+ true,
35
+ true,
36
+ false,
37
+ false,
38
+ true,
39
+ true,
40
+ false,
41
+ false,
42
+ true,
43
+ true,
44
+ true,
45
+ true,
46
+ true,
47
+ true,
48
+ false,
49
+ true,
50
+ true,
51
+ true,
52
+ true,
53
+ true,
54
+ true,
55
+ true,
56
+ true,
57
+ true,
58
+ true,
59
+ true,
60
+ true,
61
+ false,
62
+ true,
63
+ false,
64
+ false,
65
+ true,
66
+ true,
67
+ true,
68
+ false,
69
+ true,
70
+ true,
71
+ false,
72
+ true,
73
+ true,
74
+ false,
75
+ true,
76
+ true,
77
+ false,
78
+ true,
79
+ false,
80
+ true,
81
+ true,
82
+ true,
83
+ true,
84
+ false,
85
+ true,
86
+ true,
87
+ false,
88
+ true,
89
+ true,
90
+ true,
91
+ false,
92
+ true,
93
+ true,
94
+ true,
95
+ false,
96
+ true,
97
+ true,
98
+ false,
99
+ true,
100
+ false,
101
+ true,
102
+ true,
103
+ true,
104
+ true,
105
+ true,
106
+ true,
107
+ true,
108
+ true,
109
+ true,
110
+ true,
111
+ false,
112
+ false,
113
+ true,
114
+ true,
115
+ false,
116
+ false,
117
+ true,
118
+ true,
119
+ false,
120
+ true,
121
+ true,
122
+ true,
123
+ true,
124
+ true,
125
+ true,
126
+ true,
127
+ true,
128
+ false,
129
+ true,
130
+ true,
131
+ false,
132
+ true,
133
+ true,
134
+ false,
135
+ true,
136
+ true,
137
+ true,
138
+ true,
139
+ true,
140
+ true,
141
+ true,
142
+ true,
143
+ false,
144
+ true,
145
+ true,
146
+ true,
147
+ true,
148
+ true,
149
+ false,
150
+ false,
151
+ false,
152
+ true,
153
+ true,
154
+ true,
155
+ true,
156
+ false,
157
+ true,
158
+ false,
159
+ true,
160
+ false,
161
+ true,
162
+ true,
163
+ true,
164
+ true,
165
+ true,
166
+ true,
167
+ false,
168
+ true,
169
+ false,
170
+ false,
171
+ true,
172
+ true,
173
+ true,
174
+ true,
175
+ true,
176
+ true,
177
+ true,
178
+ true,
179
+ true,
180
+ true,
181
+ true,
182
+ true,
183
+ true,
184
+ false,
185
+ true,
186
+ true,
187
+ true,
188
+ true,
189
+ false,
190
+ true,
191
+ true,
192
+ false,
193
+ false,
194
+ true,
195
+ false,
196
+ true,
197
+ true,
198
+ true,
199
+ true,
200
+ true,
201
+ true,
202
+ true,
203
+ true,
204
+ true,
205
+ true,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 107.19233298301697,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_25_seed44.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 25,
5
+ "seed": 44,
6
+ "successes": [
7
+ true,
8
+ true,
9
+ true,
10
+ true,
11
+ false,
12
+ true,
13
+ false,
14
+ true,
15
+ true,
16
+ true,
17
+ false,
18
+ false,
19
+ true,
20
+ true,
21
+ false,
22
+ true,
23
+ false,
24
+ true,
25
+ true,
26
+ false,
27
+ false,
28
+ true,
29
+ true,
30
+ true,
31
+ true,
32
+ true,
33
+ true,
34
+ true,
35
+ true,
36
+ true,
37
+ false,
38
+ true,
39
+ true,
40
+ true,
41
+ true,
42
+ true,
43
+ true,
44
+ true,
45
+ false,
46
+ true,
47
+ true,
48
+ true,
49
+ false,
50
+ false,
51
+ true,
52
+ true,
53
+ false,
54
+ true,
55
+ true,
56
+ true,
57
+ true,
58
+ true,
59
+ true,
60
+ true,
61
+ false,
62
+ true,
63
+ true,
64
+ false,
65
+ true,
66
+ false,
67
+ false,
68
+ false,
69
+ false,
70
+ true,
71
+ true,
72
+ true,
73
+ true,
74
+ false,
75
+ false,
76
+ true,
77
+ true,
78
+ false,
79
+ false,
80
+ true,
81
+ true,
82
+ false,
83
+ true,
84
+ true,
85
+ true,
86
+ true,
87
+ true,
88
+ false,
89
+ false,
90
+ true,
91
+ true,
92
+ false,
93
+ false,
94
+ true,
95
+ false,
96
+ true,
97
+ true,
98
+ true,
99
+ true,
100
+ true,
101
+ true,
102
+ true,
103
+ true,
104
+ true,
105
+ false,
106
+ false,
107
+ true,
108
+ false,
109
+ true,
110
+ false,
111
+ false,
112
+ true,
113
+ true,
114
+ false,
115
+ true,
116
+ false,
117
+ false,
118
+ true,
119
+ true,
120
+ true,
121
+ false,
122
+ true,
123
+ true,
124
+ true,
125
+ true,
126
+ true,
127
+ true,
128
+ false,
129
+ false,
130
+ false,
131
+ true,
132
+ true,
133
+ false,
134
+ true,
135
+ true,
136
+ true,
137
+ true,
138
+ false,
139
+ false,
140
+ false,
141
+ true,
142
+ true,
143
+ true,
144
+ false,
145
+ false,
146
+ true,
147
+ true,
148
+ false,
149
+ true,
150
+ true,
151
+ true,
152
+ true,
153
+ true,
154
+ false,
155
+ true,
156
+ true,
157
+ false,
158
+ true,
159
+ true,
160
+ true,
161
+ true,
162
+ true,
163
+ true,
164
+ false,
165
+ true,
166
+ false,
167
+ true,
168
+ true,
169
+ true,
170
+ true,
171
+ true,
172
+ true,
173
+ true,
174
+ true,
175
+ true,
176
+ true,
177
+ true,
178
+ true,
179
+ false,
180
+ true,
181
+ true,
182
+ true,
183
+ true,
184
+ true,
185
+ true,
186
+ true,
187
+ true,
188
+ true,
189
+ true,
190
+ false,
191
+ true,
192
+ true,
193
+ true,
194
+ true,
195
+ false,
196
+ true,
197
+ true,
198
+ true,
199
+ true,
200
+ true,
201
+ true,
202
+ true,
203
+ true,
204
+ true,
205
+ true,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 108.53423619270325,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_25_seed45.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 25,
5
+ "seed": 45,
6
+ "successes": [
7
+ true,
8
+ true,
9
+ true,
10
+ true,
11
+ true,
12
+ true,
13
+ true,
14
+ false,
15
+ false,
16
+ true,
17
+ true,
18
+ true,
19
+ false,
20
+ true,
21
+ false,
22
+ true,
23
+ false,
24
+ false,
25
+ false,
26
+ true,
27
+ true,
28
+ true,
29
+ true,
30
+ true,
31
+ true,
32
+ true,
33
+ false,
34
+ true,
35
+ false,
36
+ true,
37
+ true,
38
+ true,
39
+ true,
40
+ true,
41
+ true,
42
+ true,
43
+ true,
44
+ true,
45
+ false,
46
+ true,
47
+ true,
48
+ false,
49
+ true,
50
+ true,
51
+ true,
52
+ false,
53
+ true,
54
+ false,
55
+ true,
56
+ true,
57
+ true,
58
+ true,
59
+ true,
60
+ true,
61
+ true,
62
+ true,
63
+ false,
64
+ true,
65
+ true,
66
+ false,
67
+ true,
68
+ true,
69
+ true,
70
+ true,
71
+ false,
72
+ true,
73
+ false,
74
+ true,
75
+ true,
76
+ false,
77
+ true,
78
+ true,
79
+ false,
80
+ true,
81
+ false,
82
+ false,
83
+ true,
84
+ true,
85
+ true,
86
+ true,
87
+ true,
88
+ false,
89
+ true,
90
+ false,
91
+ true,
92
+ true,
93
+ true,
94
+ true,
95
+ false,
96
+ true,
97
+ true,
98
+ true,
99
+ true,
100
+ false,
101
+ true,
102
+ true,
103
+ false,
104
+ true,
105
+ true,
106
+ true,
107
+ true,
108
+ true,
109
+ true,
110
+ true,
111
+ true,
112
+ true,
113
+ false,
114
+ true,
115
+ true,
116
+ false,
117
+ true,
118
+ true,
119
+ true,
120
+ false,
121
+ true,
122
+ true,
123
+ true,
124
+ true,
125
+ true,
126
+ true,
127
+ true,
128
+ true,
129
+ false,
130
+ true,
131
+ true,
132
+ true,
133
+ true,
134
+ true,
135
+ true,
136
+ true,
137
+ true,
138
+ false,
139
+ true,
140
+ false,
141
+ false,
142
+ false,
143
+ true,
144
+ true,
145
+ false,
146
+ true,
147
+ true,
148
+ true,
149
+ true,
150
+ true,
151
+ true,
152
+ true,
153
+ true,
154
+ true,
155
+ false,
156
+ true,
157
+ true,
158
+ false,
159
+ true,
160
+ true,
161
+ false,
162
+ true,
163
+ true,
164
+ true,
165
+ false,
166
+ true,
167
+ true,
168
+ true,
169
+ true,
170
+ true,
171
+ false,
172
+ true,
173
+ true,
174
+ true,
175
+ true,
176
+ true,
177
+ false,
178
+ true,
179
+ true,
180
+ true,
181
+ true,
182
+ true,
183
+ true,
184
+ true,
185
+ true,
186
+ false,
187
+ true,
188
+ true,
189
+ false,
190
+ true,
191
+ true,
192
+ true,
193
+ true,
194
+ true,
195
+ true,
196
+ false,
197
+ true,
198
+ true,
199
+ true,
200
+ true,
201
+ true,
202
+ true,
203
+ false,
204
+ true,
205
+ false,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 105.54125499725342,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_25_seed46.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 25,
5
+ "seed": 46,
6
+ "successes": [
7
+ true,
8
+ true,
9
+ true,
10
+ true,
11
+ true,
12
+ true,
13
+ true,
14
+ true,
15
+ true,
16
+ false,
17
+ true,
18
+ false,
19
+ true,
20
+ true,
21
+ true,
22
+ true,
23
+ true,
24
+ true,
25
+ true,
26
+ true,
27
+ true,
28
+ true,
29
+ false,
30
+ true,
31
+ false,
32
+ false,
33
+ true,
34
+ true,
35
+ false,
36
+ true,
37
+ true,
38
+ true,
39
+ false,
40
+ false,
41
+ true,
42
+ true,
43
+ true,
44
+ true,
45
+ true,
46
+ true,
47
+ true,
48
+ true,
49
+ true,
50
+ true,
51
+ false,
52
+ true,
53
+ true,
54
+ true,
55
+ true,
56
+ true,
57
+ false,
58
+ true,
59
+ true,
60
+ false,
61
+ false,
62
+ true,
63
+ true,
64
+ false,
65
+ true,
66
+ false,
67
+ true,
68
+ true,
69
+ true,
70
+ false,
71
+ true,
72
+ true,
73
+ false,
74
+ true,
75
+ false,
76
+ false,
77
+ true,
78
+ true,
79
+ false,
80
+ true,
81
+ true,
82
+ true,
83
+ false,
84
+ true,
85
+ false,
86
+ false,
87
+ false,
88
+ false,
89
+ true,
90
+ false,
91
+ false,
92
+ true,
93
+ false,
94
+ false,
95
+ false,
96
+ true,
97
+ true,
98
+ true,
99
+ true,
100
+ true,
101
+ true,
102
+ true,
103
+ true,
104
+ false,
105
+ true,
106
+ true,
107
+ true,
108
+ true,
109
+ true,
110
+ true,
111
+ true,
112
+ false,
113
+ false,
114
+ false,
115
+ true,
116
+ true,
117
+ true,
118
+ true,
119
+ false,
120
+ true,
121
+ true,
122
+ true,
123
+ true,
124
+ false,
125
+ true,
126
+ true,
127
+ true,
128
+ true,
129
+ true,
130
+ true,
131
+ true,
132
+ true,
133
+ true,
134
+ true,
135
+ true,
136
+ true,
137
+ false,
138
+ true,
139
+ true,
140
+ true,
141
+ true,
142
+ true,
143
+ true,
144
+ true,
145
+ true,
146
+ true,
147
+ true,
148
+ true,
149
+ true,
150
+ true,
151
+ true,
152
+ true,
153
+ true,
154
+ false,
155
+ true,
156
+ true,
157
+ true,
158
+ true,
159
+ true,
160
+ true,
161
+ false,
162
+ false,
163
+ false,
164
+ true,
165
+ true,
166
+ true,
167
+ true,
168
+ true,
169
+ true,
170
+ true,
171
+ true,
172
+ true,
173
+ true,
174
+ true,
175
+ true,
176
+ true,
177
+ true,
178
+ true,
179
+ true,
180
+ true,
181
+ true,
182
+ false,
183
+ true,
184
+ true,
185
+ true,
186
+ true,
187
+ true,
188
+ true,
189
+ false,
190
+ false,
191
+ true,
192
+ true,
193
+ false,
194
+ true,
195
+ false,
196
+ false,
197
+ true,
198
+ true,
199
+ true,
200
+ true,
201
+ true,
202
+ true,
203
+ false,
204
+ false,
205
+ true,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 104.94565558433533,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_50.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 50,
5
+ "seed": 42,
6
+ "successes": [
7
+ true,
8
+ false,
9
+ false,
10
+ false,
11
+ false,
12
+ false,
13
+ false,
14
+ false,
15
+ true,
16
+ true,
17
+ false,
18
+ false,
19
+ true,
20
+ false,
21
+ false,
22
+ false,
23
+ false,
24
+ false,
25
+ false,
26
+ false,
27
+ true,
28
+ true,
29
+ true,
30
+ true,
31
+ false,
32
+ true,
33
+ true,
34
+ false,
35
+ false,
36
+ false,
37
+ false,
38
+ false,
39
+ true,
40
+ true,
41
+ false,
42
+ false,
43
+ true,
44
+ false,
45
+ false,
46
+ false,
47
+ false,
48
+ false,
49
+ false,
50
+ false,
51
+ false,
52
+ false,
53
+ false,
54
+ false,
55
+ false,
56
+ false,
57
+ false,
58
+ false,
59
+ true,
60
+ false,
61
+ true,
62
+ false,
63
+ false,
64
+ false,
65
+ false,
66
+ false,
67
+ false,
68
+ false,
69
+ false,
70
+ true,
71
+ false,
72
+ false,
73
+ false,
74
+ false,
75
+ false,
76
+ false,
77
+ false,
78
+ false,
79
+ false,
80
+ false,
81
+ false,
82
+ false,
83
+ false,
84
+ false,
85
+ false,
86
+ true,
87
+ false,
88
+ false,
89
+ false,
90
+ false,
91
+ false,
92
+ false,
93
+ false,
94
+ false,
95
+ false,
96
+ false,
97
+ false,
98
+ true,
99
+ false,
100
+ true,
101
+ true,
102
+ false,
103
+ false,
104
+ false,
105
+ false,
106
+ true,
107
+ false,
108
+ true,
109
+ false,
110
+ false,
111
+ true,
112
+ false,
113
+ true,
114
+ true,
115
+ false,
116
+ false,
117
+ false,
118
+ true,
119
+ false,
120
+ true,
121
+ true,
122
+ true,
123
+ false,
124
+ true,
125
+ false,
126
+ false,
127
+ false,
128
+ false,
129
+ false,
130
+ false,
131
+ false,
132
+ true,
133
+ true,
134
+ false,
135
+ true,
136
+ false,
137
+ false,
138
+ false,
139
+ false,
140
+ false,
141
+ false,
142
+ false,
143
+ false,
144
+ false,
145
+ false,
146
+ false,
147
+ false,
148
+ true,
149
+ false,
150
+ true,
151
+ false,
152
+ true,
153
+ true,
154
+ false,
155
+ true,
156
+ false,
157
+ false,
158
+ false,
159
+ false,
160
+ false,
161
+ false,
162
+ true,
163
+ false,
164
+ true,
165
+ false,
166
+ false,
167
+ false,
168
+ false,
169
+ false,
170
+ true,
171
+ true,
172
+ false,
173
+ false,
174
+ false,
175
+ false,
176
+ true,
177
+ true,
178
+ false,
179
+ true,
180
+ false,
181
+ true,
182
+ false,
183
+ true,
184
+ true,
185
+ false,
186
+ true,
187
+ true,
188
+ true,
189
+ false,
190
+ false,
191
+ false,
192
+ false,
193
+ false,
194
+ true,
195
+ false,
196
+ false,
197
+ true,
198
+ false,
199
+ false,
200
+ false,
201
+ true,
202
+ false,
203
+ true,
204
+ false,
205
+ true,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 148.25762224197388,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_50_seed43.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 50,
5
+ "seed": 43,
6
+ "successes": [
7
+ false,
8
+ false,
9
+ false,
10
+ false,
11
+ false,
12
+ false,
13
+ false,
14
+ false,
15
+ false,
16
+ true,
17
+ false,
18
+ true,
19
+ true,
20
+ true,
21
+ true,
22
+ false,
23
+ false,
24
+ false,
25
+ false,
26
+ false,
27
+ false,
28
+ false,
29
+ false,
30
+ false,
31
+ false,
32
+ false,
33
+ true,
34
+ false,
35
+ true,
36
+ true,
37
+ false,
38
+ false,
39
+ false,
40
+ false,
41
+ true,
42
+ false,
43
+ true,
44
+ false,
45
+ false,
46
+ false,
47
+ false,
48
+ false,
49
+ true,
50
+ false,
51
+ true,
52
+ false,
53
+ false,
54
+ false,
55
+ false,
56
+ false,
57
+ false,
58
+ false,
59
+ false,
60
+ false,
61
+ true,
62
+ false,
63
+ false,
64
+ false,
65
+ false,
66
+ false,
67
+ false,
68
+ false,
69
+ true,
70
+ false,
71
+ true,
72
+ true,
73
+ false,
74
+ false,
75
+ false,
76
+ false,
77
+ false,
78
+ false,
79
+ false,
80
+ false,
81
+ false,
82
+ false,
83
+ false,
84
+ false,
85
+ false,
86
+ true,
87
+ false,
88
+ false,
89
+ false,
90
+ false,
91
+ false,
92
+ false,
93
+ false,
94
+ false,
95
+ false,
96
+ false,
97
+ true,
98
+ true,
99
+ false,
100
+ true,
101
+ false,
102
+ false,
103
+ false,
104
+ false,
105
+ false,
106
+ false,
107
+ false,
108
+ false,
109
+ false,
110
+ false,
111
+ true,
112
+ false,
113
+ false,
114
+ false,
115
+ false,
116
+ false,
117
+ false,
118
+ false,
119
+ false,
120
+ false,
121
+ true,
122
+ true,
123
+ false,
124
+ false,
125
+ false,
126
+ false,
127
+ true,
128
+ false,
129
+ true,
130
+ false,
131
+ true,
132
+ false,
133
+ true,
134
+ true,
135
+ false,
136
+ true,
137
+ false,
138
+ false,
139
+ false,
140
+ false,
141
+ false,
142
+ false,
143
+ true,
144
+ false,
145
+ false,
146
+ false,
147
+ false,
148
+ false,
149
+ false,
150
+ false,
151
+ true,
152
+ false,
153
+ false,
154
+ false,
155
+ true,
156
+ true,
157
+ false,
158
+ false,
159
+ false,
160
+ false,
161
+ false,
162
+ true,
163
+ false,
164
+ false,
165
+ false,
166
+ false,
167
+ false,
168
+ true,
169
+ false,
170
+ false,
171
+ false,
172
+ false,
173
+ false,
174
+ false,
175
+ true,
176
+ false,
177
+ true,
178
+ false,
179
+ false,
180
+ true,
181
+ false,
182
+ false,
183
+ false,
184
+ true,
185
+ false,
186
+ true,
187
+ false,
188
+ false,
189
+ false,
190
+ false,
191
+ false,
192
+ false,
193
+ false,
194
+ false,
195
+ false,
196
+ false,
197
+ false,
198
+ false,
199
+ false,
200
+ false,
201
+ false,
202
+ false,
203
+ false,
204
+ true,
205
+ false,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 149.55380821228027,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_50_seed44.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 50,
5
+ "seed": 44,
6
+ "successes": [
7
+ true,
8
+ false,
9
+ false,
10
+ false,
11
+ false,
12
+ true,
13
+ true,
14
+ false,
15
+ false,
16
+ false,
17
+ false,
18
+ false,
19
+ true,
20
+ false,
21
+ false,
22
+ true,
23
+ true,
24
+ false,
25
+ false,
26
+ false,
27
+ false,
28
+ true,
29
+ false,
30
+ true,
31
+ false,
32
+ true,
33
+ false,
34
+ true,
35
+ false,
36
+ true,
37
+ false,
38
+ false,
39
+ false,
40
+ false,
41
+ false,
42
+ false,
43
+ true,
44
+ true,
45
+ false,
46
+ false,
47
+ false,
48
+ false,
49
+ false,
50
+ false,
51
+ false,
52
+ false,
53
+ false,
54
+ false,
55
+ false,
56
+ false,
57
+ false,
58
+ false,
59
+ false,
60
+ false,
61
+ true,
62
+ false,
63
+ true,
64
+ false,
65
+ false,
66
+ false,
67
+ true,
68
+ false,
69
+ false,
70
+ true,
71
+ true,
72
+ false,
73
+ true,
74
+ false,
75
+ true,
76
+ true,
77
+ false,
78
+ false,
79
+ false,
80
+ false,
81
+ false,
82
+ true,
83
+ false,
84
+ true,
85
+ false,
86
+ false,
87
+ false,
88
+ true,
89
+ false,
90
+ false,
91
+ false,
92
+ false,
93
+ false,
94
+ true,
95
+ false,
96
+ false,
97
+ false,
98
+ false,
99
+ true,
100
+ true,
101
+ false,
102
+ false,
103
+ false,
104
+ true,
105
+ false,
106
+ false,
107
+ false,
108
+ true,
109
+ true,
110
+ false,
111
+ false,
112
+ false,
113
+ true,
114
+ true,
115
+ false,
116
+ false,
117
+ true,
118
+ true,
119
+ false,
120
+ true,
121
+ false,
122
+ false,
123
+ false,
124
+ true,
125
+ false,
126
+ true,
127
+ true,
128
+ false,
129
+ false,
130
+ false,
131
+ true,
132
+ true,
133
+ true,
134
+ false,
135
+ false,
136
+ true,
137
+ true,
138
+ false,
139
+ false,
140
+ false,
141
+ false,
142
+ false,
143
+ false,
144
+ false,
145
+ false,
146
+ false,
147
+ true,
148
+ true,
149
+ false,
150
+ false,
151
+ false,
152
+ true,
153
+ false,
154
+ true,
155
+ false,
156
+ false,
157
+ true,
158
+ false,
159
+ false,
160
+ false,
161
+ true,
162
+ false,
163
+ false,
164
+ false,
165
+ false,
166
+ false,
167
+ false,
168
+ false,
169
+ false,
170
+ true,
171
+ false,
172
+ false,
173
+ false,
174
+ false,
175
+ true,
176
+ false,
177
+ true,
178
+ true,
179
+ false,
180
+ false,
181
+ true,
182
+ false,
183
+ false,
184
+ false,
185
+ false,
186
+ false,
187
+ false,
188
+ false,
189
+ false,
190
+ true,
191
+ false,
192
+ false,
193
+ false,
194
+ false,
195
+ false,
196
+ true,
197
+ true,
198
+ true,
199
+ false,
200
+ false,
201
+ true,
202
+ false,
203
+ true,
204
+ false,
205
+ false,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 145.85111236572266,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_50_seed45.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 50,
5
+ "seed": 45,
6
+ "successes": [
7
+ false,
8
+ false,
9
+ false,
10
+ false,
11
+ true,
12
+ false,
13
+ false,
14
+ false,
15
+ true,
16
+ true,
17
+ false,
18
+ false,
19
+ false,
20
+ true,
21
+ false,
22
+ false,
23
+ false,
24
+ false,
25
+ true,
26
+ false,
27
+ true,
28
+ true,
29
+ false,
30
+ true,
31
+ true,
32
+ true,
33
+ false,
34
+ false,
35
+ true,
36
+ false,
37
+ true,
38
+ false,
39
+ false,
40
+ false,
41
+ false,
42
+ false,
43
+ false,
44
+ false,
45
+ false,
46
+ false,
47
+ false,
48
+ false,
49
+ true,
50
+ false,
51
+ false,
52
+ false,
53
+ true,
54
+ false,
55
+ false,
56
+ false,
57
+ false,
58
+ false,
59
+ false,
60
+ false,
61
+ false,
62
+ false,
63
+ false,
64
+ false,
65
+ false,
66
+ false,
67
+ true,
68
+ true,
69
+ false,
70
+ true,
71
+ true,
72
+ true,
73
+ false,
74
+ false,
75
+ false,
76
+ false,
77
+ false,
78
+ true,
79
+ false,
80
+ false,
81
+ false,
82
+ false,
83
+ false,
84
+ false,
85
+ false,
86
+ false,
87
+ false,
88
+ false,
89
+ false,
90
+ false,
91
+ false,
92
+ false,
93
+ true,
94
+ false,
95
+ false,
96
+ false,
97
+ false,
98
+ false,
99
+ false,
100
+ true,
101
+ false,
102
+ false,
103
+ false,
104
+ false,
105
+ false,
106
+ false,
107
+ false,
108
+ false,
109
+ false,
110
+ true,
111
+ false,
112
+ false,
113
+ true,
114
+ true,
115
+ true,
116
+ true,
117
+ false,
118
+ false,
119
+ false,
120
+ false,
121
+ false,
122
+ false,
123
+ false,
124
+ false,
125
+ true,
126
+ true,
127
+ false,
128
+ false,
129
+ false,
130
+ true,
131
+ false,
132
+ true,
133
+ false,
134
+ true,
135
+ false,
136
+ false,
137
+ true,
138
+ false,
139
+ false,
140
+ false,
141
+ true,
142
+ false,
143
+ false,
144
+ false,
145
+ false,
146
+ false,
147
+ false,
148
+ true,
149
+ false,
150
+ false,
151
+ false,
152
+ false,
153
+ false,
154
+ true,
155
+ false,
156
+ false,
157
+ true,
158
+ false,
159
+ true,
160
+ false,
161
+ false,
162
+ false,
163
+ false,
164
+ false,
165
+ false,
166
+ false,
167
+ false,
168
+ true,
169
+ true,
170
+ true,
171
+ true,
172
+ true,
173
+ false,
174
+ false,
175
+ false,
176
+ false,
177
+ false,
178
+ false,
179
+ false,
180
+ false,
181
+ false,
182
+ false,
183
+ false,
184
+ false,
185
+ false,
186
+ false,
187
+ false,
188
+ false,
189
+ false,
190
+ false,
191
+ false,
192
+ false,
193
+ false,
194
+ true,
195
+ false,
196
+ true,
197
+ true,
198
+ false,
199
+ false,
200
+ false,
201
+ false,
202
+ false,
203
+ true,
204
+ false,
205
+ false,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 149.69330883026123,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_50_seed46.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 50,
5
+ "seed": 46,
6
+ "successes": [
7
+ true,
8
+ false,
9
+ false,
10
+ true,
11
+ false,
12
+ false,
13
+ false,
14
+ false,
15
+ true,
16
+ true,
17
+ false,
18
+ false,
19
+ false,
20
+ true,
21
+ false,
22
+ false,
23
+ false,
24
+ true,
25
+ false,
26
+ false,
27
+ true,
28
+ false,
29
+ false,
30
+ false,
31
+ false,
32
+ true,
33
+ false,
34
+ false,
35
+ true,
36
+ false,
37
+ false,
38
+ true,
39
+ false,
40
+ false,
41
+ false,
42
+ false,
43
+ false,
44
+ false,
45
+ false,
46
+ false,
47
+ false,
48
+ true,
49
+ false,
50
+ true,
51
+ false,
52
+ false,
53
+ false,
54
+ false,
55
+ false,
56
+ false,
57
+ false,
58
+ true,
59
+ false,
60
+ false,
61
+ false,
62
+ false,
63
+ false,
64
+ true,
65
+ false,
66
+ true,
67
+ false,
68
+ false,
69
+ false,
70
+ true,
71
+ true,
72
+ true,
73
+ false,
74
+ false,
75
+ false,
76
+ false,
77
+ false,
78
+ false,
79
+ false,
80
+ false,
81
+ false,
82
+ false,
83
+ true,
84
+ false,
85
+ false,
86
+ true,
87
+ false,
88
+ false,
89
+ true,
90
+ true,
91
+ false,
92
+ false,
93
+ false,
94
+ false,
95
+ false,
96
+ false,
97
+ true,
98
+ true,
99
+ false,
100
+ false,
101
+ false,
102
+ false,
103
+ false,
104
+ false,
105
+ false,
106
+ false,
107
+ false,
108
+ false,
109
+ true,
110
+ false,
111
+ false,
112
+ true,
113
+ false,
114
+ false,
115
+ false,
116
+ true,
117
+ false,
118
+ false,
119
+ true,
120
+ true,
121
+ false,
122
+ true,
123
+ false,
124
+ false,
125
+ false,
126
+ false,
127
+ true,
128
+ false,
129
+ false,
130
+ false,
131
+ false,
132
+ false,
133
+ true,
134
+ true,
135
+ true,
136
+ true,
137
+ true,
138
+ false,
139
+ false,
140
+ false,
141
+ false,
142
+ true,
143
+ true,
144
+ true,
145
+ false,
146
+ true,
147
+ false,
148
+ false,
149
+ false,
150
+ true,
151
+ false,
152
+ false,
153
+ false,
154
+ false,
155
+ false,
156
+ true,
157
+ true,
158
+ true,
159
+ true,
160
+ false,
161
+ false,
162
+ false,
163
+ false,
164
+ false,
165
+ true,
166
+ false,
167
+ true,
168
+ true,
169
+ false,
170
+ false,
171
+ false,
172
+ false,
173
+ false,
174
+ false,
175
+ false,
176
+ false,
177
+ false,
178
+ true,
179
+ true,
180
+ true,
181
+ false,
182
+ false,
183
+ false,
184
+ true,
185
+ true,
186
+ true,
187
+ false,
188
+ false,
189
+ false,
190
+ true,
191
+ true,
192
+ false,
193
+ false,
194
+ false,
195
+ false,
196
+ false,
197
+ false,
198
+ false,
199
+ true,
200
+ false,
201
+ false,
202
+ false,
203
+ true,
204
+ false,
205
+ true,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 143.82871055603027,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_75.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 75,
5
+ "seed": 42,
6
+ "successes": [
7
+ false,
8
+ false,
9
+ true,
10
+ false,
11
+ false,
12
+ false,
13
+ false,
14
+ true,
15
+ false,
16
+ false,
17
+ false,
18
+ false,
19
+ false,
20
+ false,
21
+ false,
22
+ false,
23
+ false,
24
+ false,
25
+ false,
26
+ false,
27
+ true,
28
+ true,
29
+ false,
30
+ false,
31
+ false,
32
+ false,
33
+ false,
34
+ false,
35
+ false,
36
+ true,
37
+ false,
38
+ false,
39
+ false,
40
+ false,
41
+ false,
42
+ false,
43
+ true,
44
+ false,
45
+ false,
46
+ false,
47
+ false,
48
+ false,
49
+ true,
50
+ true,
51
+ false,
52
+ false,
53
+ false,
54
+ false,
55
+ false,
56
+ false,
57
+ false,
58
+ false,
59
+ false,
60
+ true,
61
+ false,
62
+ false,
63
+ true,
64
+ false,
65
+ false,
66
+ false,
67
+ false,
68
+ true,
69
+ false,
70
+ false,
71
+ false,
72
+ false,
73
+ false,
74
+ false,
75
+ false,
76
+ true,
77
+ false,
78
+ false,
79
+ false,
80
+ false,
81
+ false,
82
+ false,
83
+ false,
84
+ false,
85
+ false,
86
+ false,
87
+ false,
88
+ false,
89
+ false,
90
+ false,
91
+ false,
92
+ false,
93
+ false,
94
+ false,
95
+ false,
96
+ false,
97
+ false,
98
+ false,
99
+ false,
100
+ false,
101
+ false,
102
+ false,
103
+ false,
104
+ false,
105
+ false,
106
+ false,
107
+ false,
108
+ false,
109
+ false,
110
+ false,
111
+ true,
112
+ true,
113
+ true,
114
+ false,
115
+ false,
116
+ false,
117
+ false,
118
+ false,
119
+ false,
120
+ true,
121
+ false,
122
+ false,
123
+ false,
124
+ false,
125
+ false,
126
+ true,
127
+ false,
128
+ false,
129
+ false,
130
+ false,
131
+ true,
132
+ false,
133
+ false,
134
+ false,
135
+ false,
136
+ false,
137
+ true,
138
+ false,
139
+ false,
140
+ false,
141
+ false,
142
+ true,
143
+ false,
144
+ false,
145
+ false,
146
+ false,
147
+ false,
148
+ false,
149
+ true,
150
+ false,
151
+ false,
152
+ false,
153
+ false,
154
+ true,
155
+ false,
156
+ false,
157
+ true,
158
+ false,
159
+ false,
160
+ false,
161
+ false,
162
+ false,
163
+ false,
164
+ false,
165
+ false,
166
+ false,
167
+ false,
168
+ false,
169
+ false,
170
+ false,
171
+ false,
172
+ false,
173
+ false,
174
+ true,
175
+ false,
176
+ false,
177
+ false,
178
+ false,
179
+ false,
180
+ false,
181
+ true,
182
+ false,
183
+ false,
184
+ false,
185
+ false,
186
+ true,
187
+ false,
188
+ false,
189
+ false,
190
+ false,
191
+ false,
192
+ false,
193
+ false,
194
+ false,
195
+ false,
196
+ false,
197
+ false,
198
+ false,
199
+ false,
200
+ false,
201
+ false,
202
+ false,
203
+ false,
204
+ false,
205
+ true,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 157.96101546287537,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_75_seed43.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 75,
5
+ "seed": 43,
6
+ "successes": [
7
+ false,
8
+ false,
9
+ false,
10
+ false,
11
+ false,
12
+ false,
13
+ false,
14
+ false,
15
+ false,
16
+ false,
17
+ false,
18
+ false,
19
+ false,
20
+ false,
21
+ false,
22
+ false,
23
+ false,
24
+ false,
25
+ false,
26
+ false,
27
+ false,
28
+ false,
29
+ false,
30
+ false,
31
+ false,
32
+ false,
33
+ false,
34
+ true,
35
+ false,
36
+ false,
37
+ false,
38
+ false,
39
+ false,
40
+ false,
41
+ false,
42
+ false,
43
+ false,
44
+ false,
45
+ false,
46
+ false,
47
+ false,
48
+ false,
49
+ false,
50
+ false,
51
+ false,
52
+ false,
53
+ true,
54
+ true,
55
+ false,
56
+ false,
57
+ false,
58
+ false,
59
+ false,
60
+ false,
61
+ true,
62
+ false,
63
+ false,
64
+ false,
65
+ false,
66
+ true,
67
+ false,
68
+ false,
69
+ true,
70
+ false,
71
+ false,
72
+ false,
73
+ false,
74
+ false,
75
+ false,
76
+ false,
77
+ false,
78
+ false,
79
+ false,
80
+ true,
81
+ false,
82
+ false,
83
+ false,
84
+ false,
85
+ false,
86
+ false,
87
+ false,
88
+ false,
89
+ true,
90
+ false,
91
+ false,
92
+ false,
93
+ false,
94
+ false,
95
+ false,
96
+ false,
97
+ false,
98
+ false,
99
+ false,
100
+ true,
101
+ false,
102
+ false,
103
+ false,
104
+ false,
105
+ false,
106
+ false,
107
+ false,
108
+ false,
109
+ true,
110
+ false,
111
+ false,
112
+ false,
113
+ false,
114
+ false,
115
+ false,
116
+ false,
117
+ false,
118
+ false,
119
+ false,
120
+ false,
121
+ false,
122
+ false,
123
+ false,
124
+ false,
125
+ false,
126
+ true,
127
+ false,
128
+ false,
129
+ true,
130
+ false,
131
+ false,
132
+ true,
133
+ false,
134
+ false,
135
+ false,
136
+ false,
137
+ false,
138
+ false,
139
+ false,
140
+ false,
141
+ false,
142
+ false,
143
+ false,
144
+ false,
145
+ true,
146
+ false,
147
+ false,
148
+ false,
149
+ false,
150
+ false,
151
+ false,
152
+ false,
153
+ false,
154
+ false,
155
+ false,
156
+ true,
157
+ false,
158
+ false,
159
+ false,
160
+ false,
161
+ false,
162
+ false,
163
+ false,
164
+ false,
165
+ false,
166
+ false,
167
+ true,
168
+ false,
169
+ false,
170
+ false,
171
+ false,
172
+ false,
173
+ false,
174
+ false,
175
+ true,
176
+ false,
177
+ false,
178
+ false,
179
+ false,
180
+ false,
181
+ false,
182
+ false,
183
+ true,
184
+ false,
185
+ false,
186
+ false,
187
+ false,
188
+ false,
189
+ false,
190
+ false,
191
+ false,
192
+ false,
193
+ true,
194
+ false,
195
+ false,
196
+ false,
197
+ false,
198
+ false,
199
+ false,
200
+ false,
201
+ false,
202
+ false,
203
+ false,
204
+ false,
205
+ false,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 161.97326135635376,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_75_seed44.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 75,
5
+ "seed": 44,
6
+ "successes": [
7
+ false,
8
+ false,
9
+ true,
10
+ false,
11
+ false,
12
+ false,
13
+ false,
14
+ false,
15
+ false,
16
+ true,
17
+ false,
18
+ false,
19
+ false,
20
+ false,
21
+ true,
22
+ true,
23
+ true,
24
+ false,
25
+ false,
26
+ false,
27
+ false,
28
+ false,
29
+ false,
30
+ false,
31
+ false,
32
+ true,
33
+ false,
34
+ false,
35
+ false,
36
+ true,
37
+ false,
38
+ false,
39
+ false,
40
+ false,
41
+ false,
42
+ false,
43
+ false,
44
+ false,
45
+ false,
46
+ true,
47
+ false,
48
+ false,
49
+ false,
50
+ false,
51
+ false,
52
+ false,
53
+ false,
54
+ false,
55
+ false,
56
+ false,
57
+ false,
58
+ false,
59
+ false,
60
+ false,
61
+ false,
62
+ false,
63
+ true,
64
+ false,
65
+ false,
66
+ false,
67
+ false,
68
+ false,
69
+ true,
70
+ false,
71
+ false,
72
+ true,
73
+ false,
74
+ false,
75
+ false,
76
+ false,
77
+ true,
78
+ false,
79
+ false,
80
+ false,
81
+ true,
82
+ false,
83
+ false,
84
+ false,
85
+ false,
86
+ false,
87
+ false,
88
+ false,
89
+ false,
90
+ false,
91
+ false,
92
+ false,
93
+ false,
94
+ false,
95
+ false,
96
+ false,
97
+ true,
98
+ false,
99
+ false,
100
+ false,
101
+ false,
102
+ false,
103
+ false,
104
+ false,
105
+ false,
106
+ false,
107
+ true,
108
+ true,
109
+ true,
110
+ true,
111
+ true,
112
+ false,
113
+ true,
114
+ false,
115
+ false,
116
+ false,
117
+ false,
118
+ false,
119
+ false,
120
+ false,
121
+ false,
122
+ false,
123
+ false,
124
+ false,
125
+ false,
126
+ false,
127
+ true,
128
+ false,
129
+ false,
130
+ true,
131
+ false,
132
+ false,
133
+ false,
134
+ false,
135
+ false,
136
+ false,
137
+ false,
138
+ false,
139
+ false,
140
+ false,
141
+ false,
142
+ false,
143
+ false,
144
+ false,
145
+ false,
146
+ false,
147
+ false,
148
+ false,
149
+ false,
150
+ false,
151
+ true,
152
+ false,
153
+ false,
154
+ false,
155
+ false,
156
+ false,
157
+ false,
158
+ false,
159
+ false,
160
+ false,
161
+ false,
162
+ false,
163
+ false,
164
+ false,
165
+ false,
166
+ false,
167
+ false,
168
+ true,
169
+ false,
170
+ false,
171
+ false,
172
+ false,
173
+ false,
174
+ true,
175
+ false,
176
+ false,
177
+ true,
178
+ false,
179
+ false,
180
+ false,
181
+ false,
182
+ false,
183
+ false,
184
+ false,
185
+ false,
186
+ false,
187
+ false,
188
+ false,
189
+ false,
190
+ false,
191
+ false,
192
+ false,
193
+ false,
194
+ false,
195
+ true,
196
+ false,
197
+ false,
198
+ true,
199
+ false,
200
+ false,
201
+ false,
202
+ false,
203
+ false,
204
+ false,
205
+ false,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 155.4924898147583,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_75_seed45.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 75,
5
+ "seed": 45,
6
+ "successes": [
7
+ false,
8
+ false,
9
+ false,
10
+ false,
11
+ false,
12
+ false,
13
+ true,
14
+ false,
15
+ false,
16
+ false,
17
+ false,
18
+ false,
19
+ false,
20
+ false,
21
+ false,
22
+ false,
23
+ false,
24
+ false,
25
+ false,
26
+ false,
27
+ false,
28
+ false,
29
+ false,
30
+ false,
31
+ false,
32
+ false,
33
+ false,
34
+ true,
35
+ false,
36
+ false,
37
+ false,
38
+ false,
39
+ false,
40
+ false,
41
+ false,
42
+ false,
43
+ false,
44
+ false,
45
+ true,
46
+ false,
47
+ false,
48
+ false,
49
+ false,
50
+ false,
51
+ false,
52
+ false,
53
+ false,
54
+ false,
55
+ false,
56
+ false,
57
+ false,
58
+ false,
59
+ false,
60
+ false,
61
+ false,
62
+ false,
63
+ false,
64
+ false,
65
+ false,
66
+ false,
67
+ false,
68
+ false,
69
+ false,
70
+ false,
71
+ false,
72
+ false,
73
+ true,
74
+ false,
75
+ false,
76
+ false,
77
+ false,
78
+ false,
79
+ false,
80
+ false,
81
+ false,
82
+ false,
83
+ true,
84
+ false,
85
+ false,
86
+ false,
87
+ false,
88
+ true,
89
+ true,
90
+ false,
91
+ false,
92
+ true,
93
+ false,
94
+ false,
95
+ false,
96
+ false,
97
+ false,
98
+ false,
99
+ true,
100
+ false,
101
+ false,
102
+ false,
103
+ false,
104
+ false,
105
+ false,
106
+ false,
107
+ false,
108
+ false,
109
+ false,
110
+ false,
111
+ false,
112
+ false,
113
+ false,
114
+ false,
115
+ false,
116
+ false,
117
+ false,
118
+ false,
119
+ false,
120
+ false,
121
+ true,
122
+ false,
123
+ false,
124
+ false,
125
+ false,
126
+ false,
127
+ false,
128
+ false,
129
+ false,
130
+ false,
131
+ false,
132
+ false,
133
+ false,
134
+ false,
135
+ false,
136
+ false,
137
+ true,
138
+ false,
139
+ false,
140
+ false,
141
+ false,
142
+ false,
143
+ false,
144
+ false,
145
+ false,
146
+ true,
147
+ false,
148
+ false,
149
+ false,
150
+ false,
151
+ true,
152
+ false,
153
+ false,
154
+ false,
155
+ false,
156
+ true,
157
+ false,
158
+ false,
159
+ false,
160
+ true,
161
+ false,
162
+ false,
163
+ false,
164
+ false,
165
+ false,
166
+ false,
167
+ false,
168
+ false,
169
+ false,
170
+ false,
171
+ false,
172
+ false,
173
+ false,
174
+ false,
175
+ false,
176
+ false,
177
+ false,
178
+ false,
179
+ false,
180
+ false,
181
+ false,
182
+ false,
183
+ false,
184
+ true,
185
+ false,
186
+ false,
187
+ false,
188
+ true,
189
+ false,
190
+ false,
191
+ false,
192
+ true,
193
+ false,
194
+ true,
195
+ false,
196
+ false,
197
+ false,
198
+ true,
199
+ false,
200
+ true,
201
+ false,
202
+ false,
203
+ false,
204
+ false,
205
+ false,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 158.13068628311157,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }
results/pldm_pusht_75_seed46.json ADDED
@@ -0,0 +1,306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "pusht",
4
+ "offset": 75,
5
+ "seed": 46,
6
+ "successes": [
7
+ false,
8
+ false,
9
+ false,
10
+ false,
11
+ false,
12
+ false,
13
+ false,
14
+ false,
15
+ false,
16
+ false,
17
+ false,
18
+ false,
19
+ true,
20
+ true,
21
+ false,
22
+ false,
23
+ false,
24
+ false,
25
+ false,
26
+ false,
27
+ false,
28
+ false,
29
+ false,
30
+ false,
31
+ false,
32
+ false,
33
+ false,
34
+ false,
35
+ false,
36
+ false,
37
+ false,
38
+ false,
39
+ false,
40
+ false,
41
+ false,
42
+ false,
43
+ false,
44
+ false,
45
+ true,
46
+ false,
47
+ false,
48
+ false,
49
+ false,
50
+ false,
51
+ false,
52
+ false,
53
+ false,
54
+ false,
55
+ false,
56
+ false,
57
+ false,
58
+ true,
59
+ false,
60
+ false,
61
+ false,
62
+ true,
63
+ false,
64
+ false,
65
+ false,
66
+ false,
67
+ true,
68
+ true,
69
+ false,
70
+ false,
71
+ false,
72
+ false,
73
+ false,
74
+ false,
75
+ false,
76
+ false,
77
+ false,
78
+ false,
79
+ false,
80
+ false,
81
+ true,
82
+ false,
83
+ false,
84
+ false,
85
+ true,
86
+ false,
87
+ false,
88
+ false,
89
+ false,
90
+ false,
91
+ false,
92
+ false,
93
+ false,
94
+ false,
95
+ false,
96
+ false,
97
+ false,
98
+ false,
99
+ false,
100
+ false,
101
+ false,
102
+ false,
103
+ false,
104
+ false,
105
+ false,
106
+ false,
107
+ true,
108
+ false,
109
+ false,
110
+ false,
111
+ true,
112
+ false,
113
+ true,
114
+ true,
115
+ false,
116
+ false,
117
+ false,
118
+ false,
119
+ false,
120
+ false,
121
+ false,
122
+ false,
123
+ true,
124
+ false,
125
+ true,
126
+ false,
127
+ false,
128
+ false,
129
+ false,
130
+ false,
131
+ false,
132
+ false,
133
+ false,
134
+ true,
135
+ false,
136
+ false,
137
+ true,
138
+ false,
139
+ false,
140
+ false,
141
+ false,
142
+ false,
143
+ false,
144
+ false,
145
+ false,
146
+ true,
147
+ false,
148
+ false,
149
+ false,
150
+ true,
151
+ false,
152
+ false,
153
+ false,
154
+ false,
155
+ false,
156
+ false,
157
+ false,
158
+ false,
159
+ false,
160
+ true,
161
+ true,
162
+ false,
163
+ false,
164
+ false,
165
+ false,
166
+ true,
167
+ false,
168
+ false,
169
+ false,
170
+ false,
171
+ false,
172
+ false,
173
+ false,
174
+ false,
175
+ false,
176
+ false,
177
+ false,
178
+ false,
179
+ true,
180
+ true,
181
+ false,
182
+ false,
183
+ false,
184
+ false,
185
+ false,
186
+ false,
187
+ false,
188
+ false,
189
+ false,
190
+ false,
191
+ false,
192
+ false,
193
+ true,
194
+ false,
195
+ false,
196
+ true,
197
+ false,
198
+ false,
199
+ false,
200
+ false,
201
+ false,
202
+ false,
203
+ false,
204
+ false,
205
+ true,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "b6180e03b93def3f278d6819b36c3224919750b4a25dfed6f0b8a709f32367d6",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 156.72871041297913,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "pusht",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1783548,
224
+ "total_clips": 1981721,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 13933,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ -0.007812564379916172,
234
+ 0.006860687229453032
235
+ ],
236
+ "action_std": [
237
+ 0.20846744284501714,
238
+ 0.20674862637362224
239
+ ],
240
+ "optimizer": "AdamW",
241
+ "lr": 5e-05,
242
+ "weight_decay": 0.001,
243
+ "loss_weights": {
244
+ "std_loss": 18.0,
245
+ "std_t_loss": 0.7,
246
+ "cov_loss": 12.0,
247
+ "cov_t_loss": 0.0,
248
+ "temp_align_loss": 0.2
249
+ },
250
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
251
+ "epochs_completed": 10,
252
+ "history": [
253
+ {
254
+ "epoch": 1,
255
+ "loss": 1.4006494909069316,
256
+ "seconds": 3629.956325531006
257
+ },
258
+ {
259
+ "epoch": 2,
260
+ "loss": 0.7311255787088499,
261
+ "seconds": 3613.1148853302
262
+ },
263
+ {
264
+ "epoch": 3,
265
+ "loss": 0.6521564047300472,
266
+ "seconds": 3687.2167246341705
267
+ },
268
+ {
269
+ "epoch": 4,
270
+ "loss": 0.6134618710506453,
271
+ "seconds": 3652.27525472641
272
+ },
273
+ {
274
+ "epoch": 5,
275
+ "loss": 0.5886265315927727,
276
+ "seconds": 3657.845335960388
277
+ },
278
+ {
279
+ "epoch": 6,
280
+ "loss": 0.5718021439580233,
281
+ "seconds": 3673.083461523056
282
+ },
283
+ {
284
+ "epoch": 7,
285
+ "loss": 0.5604614451167178,
286
+ "seconds": 3962.2561428546906
287
+ },
288
+ {
289
+ "epoch": 8,
290
+ "loss": 0.5531895026433051,
291
+ "seconds": 3918.543078184128
292
+ },
293
+ {
294
+ "epoch": 9,
295
+ "loss": 0.5492187889900706,
296
+ "seconds": 3667.5334765911102
297
+ },
298
+ {
299
+ "epoch": 10,
300
+ "loss": 0.5478048717214997,
301
+ "seconds": 3660.553477525711
302
+ }
303
+ ]
304
+ },
305
+ "smoke": false
306
+ }