Hugging Face's logo

Datasets:
Bench2Dex
/
Leaderboard

Languages:
English
ArXiv:
Tags:
robotics
benchmark
evaluation
Dataset card Files
xet
 Community
4

Add Mora V0 evaluation logs with revised model introduction

#4
by martinwang1123 - opened
base: refs/heads/main
←
from: refs/pr/4
Discussion Files changed
+102431
-0
This view is limited to 50 files because it contains too many changes. See the raw diff here.
Files changed (50) hide show
  1. mora_v0/README.md +62 -0
  2. mora_v0/leaderboard_mora_v0.json +2577 -0
  3. mora_v0/per_task_eval/03_wine_glass_plate_balance/Equi/native_shards/000/per_episode.jsonl +0 -0
  4. mora_v0/per_task_eval/03_wine_glass_plate_balance/Equi/native_shards/000/summary.json +211 -0
  5. mora_v0/per_task_eval/03_wine_glass_plate_balance/Equi/native_shards/001/per_episode.jsonl +0 -0
  6. mora_v0/per_task_eval/03_wine_glass_plate_balance/Equi/native_shards/001/summary.json +218 -0
  7. mora_v0/per_task_eval/03_wine_glass_plate_balance/Equi/per_episode.jsonl +0 -0
  8. mora_v0/per_task_eval/03_wine_glass_plate_balance/Equi/summary.json +222 -0
  9. mora_v0/per_task_eval/03_wine_glass_plate_balance/Full/native_shards/000/per_episode.jsonl +0 -0
  10. mora_v0/per_task_eval/03_wine_glass_plate_balance/Full/native_shards/000/summary.json +216 -0
  11. mora_v0/per_task_eval/03_wine_glass_plate_balance/Full/native_shards/001/per_episode.jsonl +0 -0
  12. mora_v0/per_task_eval/03_wine_glass_plate_balance/Full/native_shards/001/summary.json +219 -0
  13. mora_v0/per_task_eval/03_wine_glass_plate_balance/Full/per_episode.jsonl +0 -0
  14. mora_v0/per_task_eval/03_wine_glass_plate_balance/Full/summary.json +227 -0
  15. mora_v0/per_task_eval/03_wine_glass_plate_balance/Inv/native_shards/000/per_episode.jsonl +0 -0
  16. mora_v0/per_task_eval/03_wine_glass_plate_balance/Inv/native_shards/000/summary.json +209 -0
  17. mora_v0/per_task_eval/03_wine_glass_plate_balance/Inv/native_shards/001/per_episode.jsonl +0 -0
  18. mora_v0/per_task_eval/03_wine_glass_plate_balance/Inv/native_shards/001/summary.json +212 -0
  19. mora_v0/per_task_eval/03_wine_glass_plate_balance/Inv/per_episode.jsonl +0 -0
  20. mora_v0/per_task_eval/03_wine_glass_plate_balance/Inv/summary.json +215 -0
  21. mora_v0/per_task_eval/03_wine_glass_plate_balance/None/native_shards/000/per_episode.jsonl +0 -0
  22. mora_v0/per_task_eval/03_wine_glass_plate_balance/None/native_shards/000/summary.json +206 -0
  23. mora_v0/per_task_eval/03_wine_glass_plate_balance/None/native_shards/001/per_episode.jsonl +0 -0
  24. mora_v0/per_task_eval/03_wine_glass_plate_balance/None/native_shards/001/summary.json +219 -0
  25. mora_v0/per_task_eval/03_wine_glass_plate_balance/None/per_episode.jsonl +0 -0
  26. mora_v0/per_task_eval/03_wine_glass_plate_balance/None/summary.json +224 -0
  27. mora_v0/per_task_eval/06_fruit_bowl_loading/Equi/native_shards/000/per_episode.jsonl +0 -0
  28. mora_v0/per_task_eval/06_fruit_bowl_loading/Equi/native_shards/000/summary.json +268 -0
  29. mora_v0/per_task_eval/06_fruit_bowl_loading/Equi/native_shards/001/per_episode.jsonl +0 -0
  30. mora_v0/per_task_eval/06_fruit_bowl_loading/Equi/native_shards/001/summary.json +268 -0
  31. mora_v0/per_task_eval/06_fruit_bowl_loading/Equi/per_episode.jsonl +0 -0
  32. mora_v0/per_task_eval/06_fruit_bowl_loading/Equi/summary.json +270 -0
  33. mora_v0/per_task_eval/06_fruit_bowl_loading/Full/native_shards/000/per_episode.jsonl +0 -0
  34. mora_v0/per_task_eval/06_fruit_bowl_loading/Full/native_shards/000/summary.json +264 -0
  35. mora_v0/per_task_eval/06_fruit_bowl_loading/Full/native_shards/001/per_episode.jsonl +0 -0
  36. mora_v0/per_task_eval/06_fruit_bowl_loading/Full/native_shards/001/summary.json +259 -0
  37. mora_v0/per_task_eval/06_fruit_bowl_loading/Full/per_episode.jsonl +0 -0
  38. mora_v0/per_task_eval/06_fruit_bowl_loading/Full/summary.json +265 -0
  39. mora_v0/per_task_eval/06_fruit_bowl_loading/Inv/native_shards/000/per_episode.jsonl +0 -0
  40. mora_v0/per_task_eval/06_fruit_bowl_loading/Inv/native_shards/000/summary.json +268 -0
  41. mora_v0/per_task_eval/06_fruit_bowl_loading/Inv/native_shards/001/per_episode.jsonl +0 -0
  42. mora_v0/per_task_eval/06_fruit_bowl_loading/Inv/native_shards/001/summary.json +255 -0
  43. mora_v0/per_task_eval/06_fruit_bowl_loading/Inv/per_episode.jsonl +0 -0
  44. mora_v0/per_task_eval/06_fruit_bowl_loading/Inv/summary.json +268 -0
  45. mora_v0/per_task_eval/06_fruit_bowl_loading/None/native_shards/000/per_episode.jsonl +4 -0
  46. mora_v0/per_task_eval/06_fruit_bowl_loading/None/native_shards/000/summary.json +241 -0
  47. mora_v0/per_task_eval/06_fruit_bowl_loading/None/native_shards/001/per_episode.jsonl +0 -0
  48. mora_v0/per_task_eval/06_fruit_bowl_loading/None/native_shards/001/summary.json +265 -0
  49. mora_v0/per_task_eval/06_fruit_bowl_loading/None/native_shards/002/per_episode.jsonl +0 -0
  50. mora_v0/per_task_eval/06_fruit_bowl_loading/None/native_shards/002/summary.json +261 -0
mora_v0/README.md ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Mora V0
2
+
3
+ Mora V0 is a vision-language-action (VLA) foundation model developed by 墨奇智能 (Morphi Robot). We perform task-specific post-training on the Mora V0 base model using Bench2Dex robot demonstrations. We evaluate the resulting policies across all 26 official Bench2Dex task and robot settings.
4
+
5
+ ## Results
6
+
7
+ Four-channel task-macro SR is 33.2308% (1,728 stable successes in 5,200 episodes). LSCR is 52.3083%. Efficiency is 0.973682 over 25 tasks with at least one success. Task 27 is included at 0/200 and its efficiency is undefined. Task 08 uses the official Schunk embodiment.
8
+
9
+ ## Selection and protocol
10
+
11
+ The reported entry uses one retrospectively selected Mora V0 checkpoint per task. Selection used recorded success rates on the evaluation scenes. The same checkpoint is used for all four channels within a task. Results from other model variants are excluded. The results JSON identifies each selected checkpoint by its SHA-256 digest. Each task has 50 episodes per channel (None, Equi., Inv., Full), numbered 1–50. Seeds, scene perturbations, embodiment and physics-step budgets are in the accompanying results JSON and logs. Each run uses reach-and-stop with a 0.5 s dwell and executes 20 action steps per policy query.
12
+
13
+ ## Evidence
14
+
15
+ Each channel includes the 50 finalized native episode records in episode order. The original worker files are preserved under `native_shards/` after sanitizing private paths. For split runs, the channel-level summary was generated from the accepted episode records using the Bench2Dex results writer. When a shard contains retried indices, `accepted_indices.json` records which rows belong to the finalized run. All selected successes and failures enter the reported metrics. The simulation metric schema is `core-v1.8-policy-step-query-split`. The aggregation code snapshot is `33c2d2eb178c7a62385ed113c3380e5933c81037`.
16
+
17
+ ## Evaluation budget per episode
18
+
19
+ The same cap applies to all four channels within each task. One policy action spans three physics steps. The caps below are simulator execution limits; training updates are counted separately.
20
+
21
+ | Task | Policy actions / episode (20 Hz) | Physics steps / episode (60 Hz) |
22
+ |---|---:|---:|
23
+ | 03 · Wine Glass Plate Balance | 1,082 | 3,246 |
24
+ | 06 · Fruit Bowl Loading | 640 | 1,920 |
25
+ | 07 · Citrus Plate Loading | 602 | 1,806 |
26
+ | 08 · Frypan Stand & Pour | 1,144 | 3,432 |
27
+ | 09 · Cleaner Box Loading | 642 | 1,926 |
28
+ | 12 · Screwdriver Box & Hammer | 1,149 | 3,447 |
29
+ | 21 · Condiment Box Loading | 827 | 2,481 |
30
+ | 22 · Tool Box Loading | 681 | 2,043 |
31
+ | 24 · Stationery Category Sorting | 1,542 | 4,626 |
32
+ | 26 · Canned Food Tray Arrangement | 871 | 2,613 |
33
+ | 27 · Ball Box Loading | 1,190 | 3,570 |
34
+ | 32 · Baking Tray Prep | 852 | 2,556 |
35
+ | 34 · Fridge Wine Interhand Pour | 1,080 | 3,240 |
36
+ | 42 · Trash Disposal | 668 | 2,004 |
37
+ | 43 · Fridge Fruit Shelf Sorting | 1,375 | 4,125 |
38
+ | 44 · Microwave Bowl Loading | 1,061 | 3,183 |
39
+ | 51 · Toilet Lid Cleaner Pour | 1,023 | 3,069 |
40
+ | 60 · Breadbasket Fast-Food Loading | 995 | 2,985 |
41
+ | 61 · Medicine Shoebox Pack | 1,034 | 3,102 |
42
+ | 62 · Shoebox Accessory Pack | 1,077 | 3,231 |
43
+ | 64 · Sports Ball Cup Sort | 665 | 1,995 |
44
+ | 67 · Faucet Cup Water Fill | 964 | 2,892 |
45
+ | 73 · Jigsaw Puzzle Assembly | 1,213 | 3,639 |
46
+ | 76 · Soup Serving | 986 | 2,958 |
47
+ | 79 · Bimanual Piano Melody | 681 | 2,043 |
48
+ | 80 · Gaming Desk Setup | 593 | 1,779 |
49
+
50
+ ## Recovery disclosure
51
+
52
+ Seven episode indices in tasks 34 and 60 initially produced an angle-decoding ValueError. The archived evaluator retried those indices with the same seed. The finalized record contains three successes and four failures. We report the finalized result and retain the seven initial error records in native_shards for review. Counting the initial errors as failures gives 1,725/5,200 (33.1731% SR). We request maintainer review of this recovery convention.
53
+
54
+ | Task | Channel | Episode index | Seed | Finalized result |
55
+ |---|---|---:|---:|---|
56
+ | 34 | Equi | 8 | 103400007 | Failure |
57
+ | 34 | Inv | 35 | 103400034 | Success |
58
+ | 34 | None | 2 | 103400001 | Failure |
59
+ | 60 | Equi | 15 | 106000014 | Success |
60
+ | 60 | Equi | 25 | 106000024 | Failure |
61
+ | 60 | Equi | 47 | 106000046 | Failure |
62
+ | 60 | Inv | 22 | 106000021 | Success |
mora_v0/leaderboard_mora_v0.json ADDED
@@ -0,0 +1,2577 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model": {
3
+ "id": "mora_v0",
4
+ "label": "Mora V0",
5
+ "variantLabel": "",
6
+ "family": "Mora",
7
+ "modelVersion": "Mora V0",
8
+ "checkpoint": null,
9
+ "inputs": "RGB, proprioception, task instruction",
10
+ "protocolId": "standard",
11
+ "displayOrder": 110,
12
+ "trainingSetting": "EgoDex mid-training followed by task-specific robot post-training",
13
+ "resultVersion": "2026-11-09",
14
+ "updatedAt": "2026-11-09",
15
+ "evaluationSetting": "One selected Mora V0 checkpoint per official embodiment/task; all four channels",
16
+ "sourceLabel": "Evaluation logs",
17
+ "sourceUrl": null,
18
+ "resultDetailsUrl": null,
19
+ "contributor": "Morphi Robot",
20
+ "contributorUrl": "https://www.morphi.com/",
21
+ "type": "VLA",
22
+ "modelUrl": "https://www.morphi.com/",
23
+ "logoPath": "leaderboard/mora_v0/logo.png",
24
+ "logoAlt": "Morphi Robot",
25
+ "weightsUrl": null,
26
+ "codeUrl": null
27
+ },
28
+ "overall": {
29
+ "taskCount": 26,
30
+ "scope": "task-macro",
31
+ "sr": 33.23076923076923,
32
+ "lscr": 52.30833333333333,
33
+ "efficiency": 0.9736822617743787,
34
+ "efficiencyStatus": "reported",
35
+ "efficiencyTaskCount": 25,
36
+ "channelSr": {
37
+ "None": 38.84615384615385,
38
+ "Equi.": 29.53846153846154,
39
+ "Inv.": 32.46153846153846,
40
+ "Full": 32.07692307692308
41
+ },
42
+ "channelLscr": {
43
+ "None": 56.276923076923076,
44
+ "Equi.": 49.19102564102564,
45
+ "Inv.": 51.9025641025641,
46
+ "Full": 51.86282051282051
47
+ },
48
+ "channelTaskCounts": {
49
+ "None": 26,
50
+ "Equi.": 26,
51
+ "Inv.": 26,
52
+ "Full": 26
53
+ },
54
+ "status": "reported"
55
+ },
56
+ "per_task": {
57
+ "12": {
58
+ "channels": {
59
+ "None": {
60
+ "status": "reported",
61
+ "episodes": 50,
62
+ "successes": 26,
63
+ "lscr": 76.5,
64
+ "source": {
65
+ "runId": "mora_v0_12_a7e364578da1",
66
+ "perEpisodeFile": "per_task_eval/12_screwdriver_box_and_hammer/None/per_episode.jsonl",
67
+ "summaryFile": "per_task_eval/12_screwdriver_box_and_hammer/None/summary.json"
68
+ }
69
+ },
70
+ "Equi.": {
71
+ "status": "reported",
72
+ "episodes": 50,
73
+ "successes": 23,
74
+ "lscr": 63.5,
75
+ "source": {
76
+ "runId": "mora_v0_12_a7e364578da1",
77
+ "perEpisodeFile": "per_task_eval/12_screwdriver_box_and_hammer/Equi/per_episode.jsonl",
78
+ "summaryFile": "per_task_eval/12_screwdriver_box_and_hammer/Equi/summary.json"
79
+ }
80
+ },
81
+ "Inv.": {
82
+ "status": "reported",
83
+ "episodes": 50,
84
+ "successes": 11,
85
+ "lscr": 55.50000000000001,
86
+ "source": {
87
+ "runId": "mora_v0_12_a7e364578da1",
88
+ "perEpisodeFile": "per_task_eval/12_screwdriver_box_and_hammer/Inv/per_episode.jsonl",
89
+ "summaryFile": "per_task_eval/12_screwdriver_box_and_hammer/Inv/summary.json"
90
+ }
91
+ },
92
+ "Full": {
93
+ "status": "reported",
94
+ "episodes": 50,
95
+ "successes": 21,
96
+ "lscr": 57.49999999999999,
97
+ "source": {
98
+ "runId": "mora_v0_12_a7e364578da1",
99
+ "perEpisodeFile": "per_task_eval/12_screwdriver_box_and_hammer/Full/per_episode.jsonl",
100
+ "summaryFile": "per_task_eval/12_screwdriver_box_and_hammer/Full/summary.json"
101
+ }
102
+ }
103
+ },
104
+ "mean": {
105
+ "status": "reported",
106
+ "sr": 40.5,
107
+ "srScope": "four-channel-mean",
108
+ "srEpisodes": 200,
109
+ "lscr": 63.25,
110
+ "lscrScope": "four-channel-mean",
111
+ "lscrEpisodes": 200,
112
+ "efficiency": 1.0374250745769327,
113
+ "efficiencyStatus": "reported"
114
+ }
115
+ },
116
+ "21": {
117
+ "channels": {
118
+ "None": {
119
+ "status": "reported",
120
+ "episodes": 50,
121
+ "successes": 17,
122
+ "lscr": 70.0,
123
+ "source": {
124
+ "runId": "mora_v0_21_b1d200c9f34f",
125
+ "perEpisodeFile": "per_task_eval/21_condiment_box_loading/None/per_episode.jsonl",
126
+ "summaryFile": "per_task_eval/21_condiment_box_loading/None/summary.json"
127
+ }
128
+ },
129
+ "Equi.": {
130
+ "status": "reported",
131
+ "episodes": 50,
132
+ "successes": 13,
133
+ "lscr": 59.33333333333333,
134
+ "source": {
135
+ "runId": "mora_v0_21_b1d200c9f34f",
136
+ "perEpisodeFile": "per_task_eval/21_condiment_box_loading/Equi/per_episode.jsonl",
137
+ "summaryFile": "per_task_eval/21_condiment_box_loading/Equi/summary.json"
138
+ }
139
+ },
140
+ "Inv.": {
141
+ "status": "reported",
142
+ "episodes": 50,
143
+ "successes": 19,
144
+ "lscr": 68.0,
145
+ "source": {
146
+ "runId": "mora_v0_21_b1d200c9f34f",
147
+ "perEpisodeFile": "per_task_eval/21_condiment_box_loading/Inv/per_episode.jsonl",
148
+ "summaryFile": "per_task_eval/21_condiment_box_loading/Inv/summary.json"
149
+ }
150
+ },
151
+ "Full": {
152
+ "status": "reported",
153
+ "episodes": 50,
154
+ "successes": 13,
155
+ "lscr": 65.33333333333333,
156
+ "source": {
157
+ "runId": "mora_v0_21_b1d200c9f34f",
158
+ "perEpisodeFile": "per_task_eval/21_condiment_box_loading/Full/per_episode.jsonl",
159
+ "summaryFile": "per_task_eval/21_condiment_box_loading/Full/summary.json"
160
+ }
161
+ }
162
+ },
163
+ "mean": {
164
+ "status": "reported",
165
+ "sr": 31.0,
166
+ "srScope": "four-channel-mean",
167
+ "srEpisodes": 200,
168
+ "lscr": 65.66666666666666,
169
+ "lscrScope": "four-channel-mean",
170
+ "lscrEpisodes": 200,
171
+ "efficiency": 0.9049577295751778,
172
+ "efficiencyStatus": "reported"
173
+ }
174
+ },
175
+ "22": {
176
+ "channels": {
177
+ "None": {
178
+ "status": "reported",
179
+ "episodes": 50,
180
+ "successes": 38,
181
+ "lscr": 90.0,
182
+ "source": {
183
+ "runId": "mora_v0_22_7ad5f3956e12",
184
+ "perEpisodeFile": "per_task_eval/22_tool_box_loading/None/per_episode.jsonl",
185
+ "summaryFile": "per_task_eval/22_tool_box_loading/None/summary.json"
186
+ }
187
+ },
188
+ "Equi.": {
189
+ "status": "reported",
190
+ "episodes": 50,
191
+ "successes": 31,
192
+ "lscr": 81.0,
193
+ "source": {
194
+ "runId": "mora_v0_22_7ad5f3956e12",
195
+ "perEpisodeFile": "per_task_eval/22_tool_box_loading/Equi/per_episode.jsonl",
196
+ "summaryFile": "per_task_eval/22_tool_box_loading/Equi/summary.json"
197
+ }
198
+ },
199
+ "Inv.": {
200
+ "status": "reported",
201
+ "episodes": 50,
202
+ "successes": 37,
203
+ "lscr": 88.0,
204
+ "source": {
205
+ "runId": "mora_v0_22_7ad5f3956e12",
206
+ "perEpisodeFile": "per_task_eval/22_tool_box_loading/Inv/per_episode.jsonl",
207
+ "summaryFile": "per_task_eval/22_tool_box_loading/Inv/summary.json"
208
+ }
209
+ },
210
+ "Full": {
211
+ "status": "reported",
212
+ "episodes": 50,
213
+ "successes": 28,
214
+ "lscr": 79.0,
215
+ "source": {
216
+ "runId": "mora_v0_22_7ad5f3956e12",
217
+ "perEpisodeFile": "per_task_eval/22_tool_box_loading/Full/per_episode.jsonl",
218
+ "summaryFile": "per_task_eval/22_tool_box_loading/Full/summary.json"
219
+ }
220
+ }
221
+ },
222
+ "mean": {
223
+ "status": "reported",
224
+ "sr": 67.0,
225
+ "srScope": "four-channel-mean",
226
+ "srEpisodes": 200,
227
+ "lscr": 84.5,
228
+ "lscrScope": "four-channel-mean",
229
+ "lscrEpisodes": 200,
230
+ "efficiency": 0.9425646167442774,
231
+ "efficiencyStatus": "reported"
232
+ }
233
+ },
234
+ "24": {
235
+ "channels": {
236
+ "None": {
237
+ "status": "reported",
238
+ "episodes": 50,
239
+ "successes": 3,
240
+ "lscr": 31.0,
241
+ "source": {
242
+ "runId": "mora_v0_24_923879d5d0cf",
243
+ "perEpisodeFile": "per_task_eval/24_stationery_category_sorting/None/per_episode.jsonl",
244
+ "summaryFile": "per_task_eval/24_stationery_category_sorting/None/summary.json"
245
+ }
246
+ },
247
+ "Equi.": {
248
+ "status": "reported",
249
+ "episodes": 50,
250
+ "successes": 1,
251
+ "lscr": 35.0,
252
+ "source": {
253
+ "runId": "mora_v0_24_923879d5d0cf",
254
+ "perEpisodeFile": "per_task_eval/24_stationery_category_sorting/Equi/per_episode.jsonl",
255
+ "summaryFile": "per_task_eval/24_stationery_category_sorting/Equi/summary.json"
256
+ }
257
+ },
258
+ "Inv.": {
259
+ "status": "reported",
260
+ "episodes": 50,
261
+ "successes": 3,
262
+ "lscr": 32.5,
263
+ "source": {
264
+ "runId": "mora_v0_24_923879d5d0cf",
265
+ "perEpisodeFile": "per_task_eval/24_stationery_category_sorting/Inv/per_episode.jsonl",
266
+ "summaryFile": "per_task_eval/24_stationery_category_sorting/Inv/summary.json"
267
+ }
268
+ },
269
+ "Full": {
270
+ "status": "reported",
271
+ "episodes": 50,
272
+ "successes": 4,
273
+ "lscr": 30.5,
274
+ "source": {
275
+ "runId": "mora_v0_24_923879d5d0cf",
276
+ "perEpisodeFile": "per_task_eval/24_stationery_category_sorting/Full/per_episode.jsonl",
277
+ "summaryFile": "per_task_eval/24_stationery_category_sorting/Full/summary.json"
278
+ }
279
+ }
280
+ },
281
+ "mean": {
282
+ "status": "reported",
283
+ "sr": 5.5,
284
+ "srScope": "four-channel-mean",
285
+ "srEpisodes": 200,
286
+ "lscr": 32.25,
287
+ "lscrScope": "four-channel-mean",
288
+ "lscrEpisodes": 200,
289
+ "efficiency": 1.128777792570896,
290
+ "efficiencyStatus": "reported"
291
+ }
292
+ },
293
+ "26": {
294
+ "channels": {
295
+ "None": {
296
+ "status": "reported",
297
+ "episodes": 50,
298
+ "successes": 20,
299
+ "lscr": 70.0,
300
+ "source": {
301
+ "runId": "mora_v0_26_b7f8e9b7a64f",
302
+ "perEpisodeFile": "per_task_eval/26_canned_food_tray_line_arrangement/None/per_episode.jsonl",
303
+ "summaryFile": "per_task_eval/26_canned_food_tray_line_arrangement/None/summary.json"
304
+ }
305
+ },
306
+ "Equi.": {
307
+ "status": "reported",
308
+ "episodes": 50,
309
+ "successes": 9,
310
+ "lscr": 52.5,
311
+ "source": {
312
+ "runId": "mora_v0_26_b7f8e9b7a64f",
313
+ "perEpisodeFile": "per_task_eval/26_canned_food_tray_line_arrangement/Equi/per_episode.jsonl",
314
+ "summaryFile": "per_task_eval/26_canned_food_tray_line_arrangement/Equi/summary.json"
315
+ }
316
+ },
317
+ "Inv.": {
318
+ "status": "reported",
319
+ "episodes": 50,
320
+ "successes": 12,
321
+ "lscr": 54.0,
322
+ "source": {
323
+ "runId": "mora_v0_26_b7f8e9b7a64f",
324
+ "perEpisodeFile": "per_task_eval/26_canned_food_tray_line_arrangement/Inv/per_episode.jsonl",
325
+ "summaryFile": "per_task_eval/26_canned_food_tray_line_arrangement/Inv/summary.json"
326
+ }
327
+ },
328
+ "Full": {
329
+ "status": "reported",
330
+ "episodes": 50,
331
+ "successes": 17,
332
+ "lscr": 60.5,
333
+ "source": {
334
+ "runId": "mora_v0_26_b7f8e9b7a64f",
335
+ "perEpisodeFile": "per_task_eval/26_canned_food_tray_line_arrangement/Full/per_episode.jsonl",
336
+ "summaryFile": "per_task_eval/26_canned_food_tray_line_arrangement/Full/summary.json"
337
+ }
338
+ }
339
+ },
340
+ "mean": {
341
+ "status": "reported",
342
+ "sr": 29.0,
343
+ "srScope": "four-channel-mean",
344
+ "srEpisodes": 200,
345
+ "lscr": 59.25,
346
+ "lscrScope": "four-channel-mean",
347
+ "lscrEpisodes": 200,
348
+ "efficiency": 0.9530798981228186,
349
+ "efficiencyStatus": "reported"
350
+ }
351
+ },
352
+ "27": {
353
+ "channels": {
354
+ "None": {
355
+ "status": "reported",
356
+ "episodes": 50,
357
+ "successes": 0,
358
+ "lscr": 17.0,
359
+ "source": {
360
+ "runId": "mora_v0_27_0e0aafb7f407",
361
+ "perEpisodeFile": "per_task_eval/27_ball_box_loading/None/per_episode.jsonl",
362
+ "summaryFile": "per_task_eval/27_ball_box_loading/None/summary.json"
363
+ }
364
+ },
365
+ "Equi.": {
366
+ "status": "reported",
367
+ "episodes": 50,
368
+ "successes": 0,
369
+ "lscr": 15.5,
370
+ "source": {
371
+ "runId": "mora_v0_27_0e0aafb7f407",
372
+ "perEpisodeFile": "per_task_eval/27_ball_box_loading/Equi/per_episode.jsonl",
373
+ "summaryFile": "per_task_eval/27_ball_box_loading/Equi/summary.json"
374
+ }
375
+ },
376
+ "Inv.": {
377
+ "status": "reported",
378
+ "episodes": 50,
379
+ "successes": 0,
380
+ "lscr": 23.5,
381
+ "source": {
382
+ "runId": "mora_v0_27_0e0aafb7f407",
383
+ "perEpisodeFile": "per_task_eval/27_ball_box_loading/Inv/per_episode.jsonl",
384
+ "summaryFile": "per_task_eval/27_ball_box_loading/Inv/summary.json"
385
+ }
386
+ },
387
+ "Full": {
388
+ "status": "reported",
389
+ "episodes": 50,
390
+ "successes": 0,
391
+ "lscr": 26.0,
392
+ "source": {
393
+ "runId": "mora_v0_27_0e0aafb7f407",
394
+ "perEpisodeFile": "per_task_eval/27_ball_box_loading/Full/per_episode.jsonl",
395
+ "summaryFile": "per_task_eval/27_ball_box_loading/Full/summary.json"
396
+ }
397
+ }
398
+ },
399
+ "mean": {
400
+ "status": "reported",
401
+ "sr": 0.0,
402
+ "srScope": "four-channel-mean",
403
+ "srEpisodes": 200,
404
+ "lscr": 20.5,
405
+ "lscrScope": "four-channel-mean",
406
+ "lscrEpisodes": 200,
407
+ "efficiency": null,
408
+ "efficiencyStatus": "no-success"
409
+ }
410
+ },
411
+ "32": {
412
+ "channels": {
413
+ "None": {
414
+ "status": "reported",
415
+ "episodes": 50,
416
+ "successes": 15,
417
+ "lscr": 75.0,
418
+ "source": {
419
+ "runId": "mora_v0_32_a13ca410c29e",
420
+ "perEpisodeFile": "per_task_eval/32_baking_tray_prep_with_tools/None/per_episode.jsonl",
421
+ "summaryFile": "per_task_eval/32_baking_tray_prep_with_tools/None/summary.json"
422
+ }
423
+ },
424
+ "Equi.": {
425
+ "status": "reported",
426
+ "episodes": 50,
427
+ "successes": 15,
428
+ "lscr": 66.5,
429
+ "source": {
430
+ "runId": "mora_v0_32_a13ca410c29e",
431
+ "perEpisodeFile": "per_task_eval/32_baking_tray_prep_with_tools/Equi/per_episode.jsonl",
432
+ "summaryFile": "per_task_eval/32_baking_tray_prep_with_tools/Equi/summary.json"
433
+ }
434
+ },
435
+ "Inv.": {
436
+ "status": "reported",
437
+ "episodes": 50,
438
+ "successes": 10,
439
+ "lscr": 69.0,
440
+ "source": {
441
+ "runId": "mora_v0_32_a13ca410c29e",
442
+ "perEpisodeFile": "per_task_eval/32_baking_tray_prep_with_tools/Inv/per_episode.jsonl",
443
+ "summaryFile": "per_task_eval/32_baking_tray_prep_with_tools/Inv/summary.json"
444
+ }
445
+ },
446
+ "Full": {
447
+ "status": "reported",
448
+ "episodes": 50,
449
+ "successes": 3,
450
+ "lscr": 55.00000000000001,
451
+ "source": {
452
+ "runId": "mora_v0_32_a13ca410c29e",
453
+ "perEpisodeFile": "per_task_eval/32_baking_tray_prep_with_tools/Full/per_episode.jsonl",
454
+ "summaryFile": "per_task_eval/32_baking_tray_prep_with_tools/Full/summary.json"
455
+ }
456
+ }
457
+ },
458
+ "mean": {
459
+ "status": "reported",
460
+ "sr": 21.5,
461
+ "srScope": "four-channel-mean",
462
+ "srEpisodes": 200,
463
+ "lscr": 66.375,
464
+ "lscrScope": "four-channel-mean",
465
+ "lscrEpisodes": 200,
466
+ "efficiency": 0.9769207166584399,
467
+ "efficiencyStatus": "reported"
468
+ }
469
+ },
470
+ "34": {
471
+ "channels": {
472
+ "None": {
473
+ "status": "reported",
474
+ "episodes": 50,
475
+ "successes": 9,
476
+ "lscr": 55.2,
477
+ "source": {
478
+ "runId": "mora_v0_34_f4d6423dad86",
479
+ "perEpisodeFile": "per_task_eval/34_fridge_wine_interhand_pour/None/per_episode.jsonl",
480
+ "summaryFile": "per_task_eval/34_fridge_wine_interhand_pour/None/summary.json"
481
+ }
482
+ },
483
+ "Equi.": {
484
+ "status": "reported",
485
+ "episodes": 50,
486
+ "successes": 9,
487
+ "lscr": 52.800000000000004,
488
+ "source": {
489
+ "runId": "mora_v0_34_f4d6423dad86",
490
+ "perEpisodeFile": "per_task_eval/34_fridge_wine_interhand_pour/Equi/per_episode.jsonl",
491
+ "summaryFile": "per_task_eval/34_fridge_wine_interhand_pour/Equi/summary.json"
492
+ }
493
+ },
494
+ "Inv.": {
495
+ "status": "reported",
496
+ "episodes": 50,
497
+ "successes": 6,
498
+ "lscr": 48.8,
499
+ "source": {
500
+ "runId": "mora_v0_34_f4d6423dad86",
501
+ "perEpisodeFile": "per_task_eval/34_fridge_wine_interhand_pour/Inv/per_episode.jsonl",
502
+ "summaryFile": "per_task_eval/34_fridge_wine_interhand_pour/Inv/summary.json"
503
+ }
504
+ },
505
+ "Full": {
506
+ "status": "reported",
507
+ "episodes": 50,
508
+ "successes": 5,
509
+ "lscr": 45.6,
510
+ "source": {
511
+ "runId": "mora_v0_34_f4d6423dad86",
512
+ "perEpisodeFile": "per_task_eval/34_fridge_wine_interhand_pour/Full/per_episode.jsonl",
513
+ "summaryFile": "per_task_eval/34_fridge_wine_interhand_pour/Full/summary.json"
514
+ }
515
+ }
516
+ },
517
+ "mean": {
518
+ "status": "reported",
519
+ "sr": 14.5,
520
+ "srScope": "four-channel-mean",
521
+ "srEpisodes": 200,
522
+ "lscr": 50.6,
523
+ "lscrScope": "four-channel-mean",
524
+ "lscrEpisodes": 200,
525
+ "efficiency": 1.0490880771742954,
526
+ "efficiencyStatus": "reported"
527
+ }
528
+ },
529
+ "42": {
530
+ "channels": {
531
+ "None": {
532
+ "status": "reported",
533
+ "episodes": 50,
534
+ "successes": 16,
535
+ "lscr": 58.666666666666664,
536
+ "source": {
537
+ "runId": "mora_v0_42_f7fb5f6fc3fb",
538
+ "perEpisodeFile": "per_task_eval/42_trash_disposal/None/per_episode.jsonl",
539
+ "summaryFile": "per_task_eval/42_trash_disposal/None/summary.json"
540
+ }
541
+ },
542
+ "Equi.": {
543
+ "status": "reported",
544
+ "episodes": 50,
545
+ "successes": 12,
546
+ "lscr": 46.0,
547
+ "source": {
548
+ "runId": "mora_v0_42_f7fb5f6fc3fb",
549
+ "perEpisodeFile": "per_task_eval/42_trash_disposal/Equi/per_episode.jsonl",
550
+ "summaryFile": "per_task_eval/42_trash_disposal/Equi/summary.json"
551
+ }
552
+ },
553
+ "Inv.": {
554
+ "status": "reported",
555
+ "episodes": 50,
556
+ "successes": 19,
557
+ "lscr": 55.99999999999999,
558
+ "source": {
559
+ "runId": "mora_v0_42_f7fb5f6fc3fb",
560
+ "perEpisodeFile": "per_task_eval/42_trash_disposal/Inv/per_episode.jsonl",
561
+ "summaryFile": "per_task_eval/42_trash_disposal/Inv/summary.json"
562
+ }
563
+ },
564
+ "Full": {
565
+ "status": "reported",
566
+ "episodes": 50,
567
+ "successes": 17,
568
+ "lscr": 55.99999999999999,
569
+ "source": {
570
+ "runId": "mora_v0_42_f7fb5f6fc3fb",
571
+ "perEpisodeFile": "per_task_eval/42_trash_disposal/Full/per_episode.jsonl",
572
+ "summaryFile": "per_task_eval/42_trash_disposal/Full/summary.json"
573
+ }
574
+ }
575
+ },
576
+ "mean": {
577
+ "status": "reported",
578
+ "sr": 32.0,
579
+ "srScope": "four-channel-mean",
580
+ "srEpisodes": 200,
581
+ "lscr": 54.166666666666664,
582
+ "lscrScope": "four-channel-mean",
583
+ "lscrEpisodes": 200,
584
+ "efficiency": 0.9627311009525525,
585
+ "efficiencyStatus": "reported"
586
+ }
587
+ },
588
+ "43": {
589
+ "channels": {
590
+ "None": {
591
+ "status": "reported",
592
+ "episodes": 50,
593
+ "successes": 3,
594
+ "lscr": 27.500000000000004,
595
+ "source": {
596
+ "runId": "mora_v0_43_6aa35c9de9ab",
597
+ "perEpisodeFile": "per_task_eval/43_fridge_fruit_shelf_sorting/None/per_episode.jsonl",
598
+ "summaryFile": "per_task_eval/43_fridge_fruit_shelf_sorting/None/summary.json"
599
+ }
600
+ },
601
+ "Equi.": {
602
+ "status": "reported",
603
+ "episodes": 50,
604
+ "successes": 1,
605
+ "lscr": 23.5,
606
+ "source": {
607
+ "runId": "mora_v0_43_6aa35c9de9ab",
608
+ "perEpisodeFile": "per_task_eval/43_fridge_fruit_shelf_sorting/Equi/per_episode.jsonl",
609
+ "summaryFile": "per_task_eval/43_fridge_fruit_shelf_sorting/Equi/summary.json"
610
+ }
611
+ },
612
+ "Inv.": {
613
+ "status": "reported",
614
+ "episodes": 50,
615
+ "successes": 3,
616
+ "lscr": 35.0,
617
+ "source": {
618
+ "runId": "mora_v0_43_6aa35c9de9ab",
619
+ "perEpisodeFile": "per_task_eval/43_fridge_fruit_shelf_sorting/Inv/per_episode.jsonl",
620
+ "summaryFile": "per_task_eval/43_fridge_fruit_shelf_sorting/Inv/summary.json"
621
+ }
622
+ },
623
+ "Full": {
624
+ "status": "reported",
625
+ "episodes": 50,
626
+ "successes": 2,
627
+ "lscr": 30.5,
628
+ "source": {
629
+ "runId": "mora_v0_43_6aa35c9de9ab",
630
+ "perEpisodeFile": "per_task_eval/43_fridge_fruit_shelf_sorting/Full/per_episode.jsonl",
631
+ "summaryFile": "per_task_eval/43_fridge_fruit_shelf_sorting/Full/summary.json"
632
+ }
633
+ }
634
+ },
635
+ "mean": {
636
+ "status": "reported",
637
+ "sr": 4.5,
638
+ "srScope": "four-channel-mean",
639
+ "srEpisodes": 200,
640
+ "lscr": 29.125,
641
+ "lscrScope": "four-channel-mean",
642
+ "lscrEpisodes": 200,
643
+ "efficiency": 0.9532429517793869,
644
+ "efficiencyStatus": "reported"
645
+ }
646
+ },
647
+ "44": {
648
+ "channels": {
649
+ "None": {
650
+ "status": "reported",
651
+ "episodes": 50,
652
+ "successes": 33,
653
+ "lscr": 73.5,
654
+ "source": {
655
+ "runId": "mora_v0_44_606362923423",
656
+ "perEpisodeFile": "per_task_eval/44_microwave_bowl_loading/None/per_episode.jsonl",
657
+ "summaryFile": "per_task_eval/44_microwave_bowl_loading/None/summary.json"
658
+ }
659
+ },
660
+ "Equi.": {
661
+ "status": "reported",
662
+ "episodes": 50,
663
+ "successes": 29,
664
+ "lscr": 70.0,
665
+ "source": {
666
+ "runId": "mora_v0_44_606362923423",
667
+ "perEpisodeFile": "per_task_eval/44_microwave_bowl_loading/Equi/per_episode.jsonl",
668
+ "summaryFile": "per_task_eval/44_microwave_bowl_loading/Equi/summary.json"
669
+ }
670
+ },
671
+ "Inv.": {
672
+ "status": "reported",
673
+ "episodes": 50,
674
+ "successes": 25,
675
+ "lscr": 62.5,
676
+ "source": {
677
+ "runId": "mora_v0_44_606362923423",
678
+ "perEpisodeFile": "per_task_eval/44_microwave_bowl_loading/Inv/per_episode.jsonl",
679
+ "summaryFile": "per_task_eval/44_microwave_bowl_loading/Inv/summary.json"
680
+ }
681
+ },
682
+ "Full": {
683
+ "status": "reported",
684
+ "episodes": 50,
685
+ "successes": 29,
686
+ "lscr": 72.0,
687
+ "source": {
688
+ "runId": "mora_v0_44_606362923423",
689
+ "perEpisodeFile": "per_task_eval/44_microwave_bowl_loading/Full/per_episode.jsonl",
690
+ "summaryFile": "per_task_eval/44_microwave_bowl_loading/Full/summary.json"
691
+ }
692
+ }
693
+ },
694
+ "mean": {
695
+ "status": "reported",
696
+ "sr": 58.0,
697
+ "srScope": "four-channel-mean",
698
+ "srEpisodes": 200,
699
+ "lscr": 69.5,
700
+ "lscrScope": "four-channel-mean",
701
+ "lscrEpisodes": 200,
702
+ "efficiency": 1.0178552230042095,
703
+ "efficiencyStatus": "reported"
704
+ }
705
+ },
706
+ "51": {
707
+ "channels": {
708
+ "None": {
709
+ "status": "reported",
710
+ "episodes": 50,
711
+ "successes": 18,
712
+ "lscr": 43.333333333333336,
713
+ "source": {
714
+ "runId": "mora_v0_51_930eadc35cc0",
715
+ "perEpisodeFile": "per_task_eval/51_toilet_lid_cleaner_pour/None/per_episode.jsonl",
716
+ "summaryFile": "per_task_eval/51_toilet_lid_cleaner_pour/None/summary.json"
717
+ }
718
+ },
719
+ "Equi.": {
720
+ "status": "reported",
721
+ "episodes": 50,
722
+ "successes": 25,
723
+ "lscr": 59.333333333333336,
724
+ "source": {
725
+ "runId": "mora_v0_51_930eadc35cc0",
726
+ "perEpisodeFile": "per_task_eval/51_toilet_lid_cleaner_pour/Equi/per_episode.jsonl",
727
+ "summaryFile": "per_task_eval/51_toilet_lid_cleaner_pour/Equi/summary.json"
728
+ }
729
+ },
730
+ "Inv.": {
731
+ "status": "reported",
732
+ "episodes": 50,
733
+ "successes": 25,
734
+ "lscr": 58.666666666666664,
735
+ "source": {
736
+ "runId": "mora_v0_51_930eadc35cc0",
737
+ "perEpisodeFile": "per_task_eval/51_toilet_lid_cleaner_pour/Inv/per_episode.jsonl",
738
+ "summaryFile": "per_task_eval/51_toilet_lid_cleaner_pour/Inv/summary.json"
739
+ }
740
+ },
741
+ "Full": {
742
+ "status": "reported",
743
+ "episodes": 50,
744
+ "successes": 23,
745
+ "lscr": 52.0,
746
+ "source": {
747
+ "runId": "mora_v0_51_930eadc35cc0",
748
+ "perEpisodeFile": "per_task_eval/51_toilet_lid_cleaner_pour/Full/per_episode.jsonl",
749
+ "summaryFile": "per_task_eval/51_toilet_lid_cleaner_pour/Full/summary.json"
750
+ }
751
+ }
752
+ },
753
+ "mean": {
754
+ "status": "reported",
755
+ "sr": 45.5,
756
+ "srScope": "four-channel-mean",
757
+ "srEpisodes": 200,
758
+ "lscr": 53.333333333333336,
759
+ "lscrScope": "four-channel-mean",
760
+ "lscrEpisodes": 200,
761
+ "efficiency": 1.0515418502202643,
762
+ "efficiencyStatus": "reported"
763
+ }
764
+ },
765
+ "60": {
766
+ "channels": {
767
+ "None": {
768
+ "status": "reported",
769
+ "episodes": 50,
770
+ "successes": 12,
771
+ "lscr": 32.5,
772
+ "source": {
773
+ "runId": "mora_v0_60_648585ddfe8f",
774
+ "perEpisodeFile": "per_task_eval/60_breadbasket_fast_food_loading/None/per_episode.jsonl",
775
+ "summaryFile": "per_task_eval/60_breadbasket_fast_food_loading/None/summary.json"
776
+ }
777
+ },
778
+ "Equi.": {
779
+ "status": "reported",
780
+ "episodes": 50,
781
+ "successes": 12,
782
+ "lscr": 36.5,
783
+ "source": {
784
+ "runId": "mora_v0_60_648585ddfe8f",
785
+ "perEpisodeFile": "per_task_eval/60_breadbasket_fast_food_loading/Equi/per_episode.jsonl",
786
+ "summaryFile": "per_task_eval/60_breadbasket_fast_food_loading/Equi/summary.json"
787
+ }
788
+ },
789
+ "Inv.": {
790
+ "status": "reported",
791
+ "episodes": 50,
792
+ "successes": 12,
793
+ "lscr": 28.999999999999996,
794
+ "source": {
795
+ "runId": "mora_v0_60_648585ddfe8f",
796
+ "perEpisodeFile": "per_task_eval/60_breadbasket_fast_food_loading/Inv/per_episode.jsonl",
797
+ "summaryFile": "per_task_eval/60_breadbasket_fast_food_loading/Inv/summary.json"
798
+ }
799
+ },
800
+ "Full": {
801
+ "status": "reported",
802
+ "episodes": 50,
803
+ "successes": 17,
804
+ "lscr": 49.0,
805
+ "source": {
806
+ "runId": "mora_v0_60_648585ddfe8f",
807
+ "perEpisodeFile": "per_task_eval/60_breadbasket_fast_food_loading/Full/per_episode.jsonl",
808
+ "summaryFile": "per_task_eval/60_breadbasket_fast_food_loading/Full/summary.json"
809
+ }
810
+ }
811
+ },
812
+ "mean": {
813
+ "status": "reported",
814
+ "sr": 26.5,
815
+ "srScope": "four-channel-mean",
816
+ "srEpisodes": 200,
817
+ "lscr": 36.75,
818
+ "lscrScope": "four-channel-mean",
819
+ "lscrEpisodes": 200,
820
+ "efficiency": 0.9389847417317645,
821
+ "efficiencyStatus": "reported"
822
+ }
823
+ },
824
+ "61": {
825
+ "channels": {
826
+ "None": {
827
+ "status": "reported",
828
+ "episodes": 50,
829
+ "successes": 23,
830
+ "lscr": 68.0,
831
+ "source": {
832
+ "runId": "mora_v0_61_bb96fa88d33c",
833
+ "perEpisodeFile": "per_task_eval/61_medicine_shoebox_pack/None/per_episode.jsonl",
834
+ "summaryFile": "per_task_eval/61_medicine_shoebox_pack/None/summary.json"
835
+ }
836
+ },
837
+ "Equi.": {
838
+ "status": "reported",
839
+ "episodes": 50,
840
+ "successes": 11,
841
+ "lscr": 45.33333333333333,
842
+ "source": {
843
+ "runId": "mora_v0_61_bb96fa88d33c",
844
+ "perEpisodeFile": "per_task_eval/61_medicine_shoebox_pack/Equi/per_episode.jsonl",
845
+ "summaryFile": "per_task_eval/61_medicine_shoebox_pack/Equi/summary.json"
846
+ }
847
+ },
848
+ "Inv.": {
849
+ "status": "reported",
850
+ "episodes": 50,
851
+ "successes": 24,
852
+ "lscr": 66.66666666666666,
853
+ "source": {
854
+ "runId": "mora_v0_61_bb96fa88d33c",
855
+ "perEpisodeFile": "per_task_eval/61_medicine_shoebox_pack/Inv/per_episode.jsonl",
856
+ "summaryFile": "per_task_eval/61_medicine_shoebox_pack/Inv/summary.json"
857
+ }
858
+ },
859
+ "Full": {
860
+ "status": "reported",
861
+ "episodes": 50,
862
+ "successes": 20,
863
+ "lscr": 59.33333333333333,
864
+ "source": {
865
+ "runId": "mora_v0_61_bb96fa88d33c",
866
+ "perEpisodeFile": "per_task_eval/61_medicine_shoebox_pack/Full/per_episode.jsonl",
867
+ "summaryFile": "per_task_eval/61_medicine_shoebox_pack/Full/summary.json"
868
+ }
869
+ }
870
+ },
871
+ "mean": {
872
+ "status": "reported",
873
+ "sr": 39.0,
874
+ "srScope": "four-channel-mean",
875
+ "srEpisodes": 200,
876
+ "lscr": 59.83333333333333,
877
+ "lscrScope": "four-channel-mean",
878
+ "lscrEpisodes": 200,
879
+ "efficiency": 0.9715745768125199,
880
+ "efficiencyStatus": "reported"
881
+ }
882
+ },
883
+ "62": {
884
+ "channels": {
885
+ "None": {
886
+ "status": "reported",
887
+ "episodes": 50,
888
+ "successes": 33,
889
+ "lscr": 82.0,
890
+ "source": {
891
+ "runId": "mora_v0_62_67bb9aa3551f",
892
+ "perEpisodeFile": "per_task_eval/62_shoebox_accessory_pack/None/per_episode.jsonl",
893
+ "summaryFile": "per_task_eval/62_shoebox_accessory_pack/None/summary.json"
894
+ }
895
+ },
896
+ "Equi.": {
897
+ "status": "reported",
898
+ "episodes": 50,
899
+ "successes": 27,
900
+ "lscr": 64.0,
901
+ "source": {
902
+ "runId": "mora_v0_62_67bb9aa3551f",
903
+ "perEpisodeFile": "per_task_eval/62_shoebox_accessory_pack/Equi/per_episode.jsonl",
904
+ "summaryFile": "per_task_eval/62_shoebox_accessory_pack/Equi/summary.json"
905
+ }
906
+ },
907
+ "Inv.": {
908
+ "status": "reported",
909
+ "episodes": 50,
910
+ "successes": 23,
911
+ "lscr": 61.33333333333333,
912
+ "source": {
913
+ "runId": "mora_v0_62_67bb9aa3551f",
914
+ "perEpisodeFile": "per_task_eval/62_shoebox_accessory_pack/Inv/per_episode.jsonl",
915
+ "summaryFile": "per_task_eval/62_shoebox_accessory_pack/Inv/summary.json"
916
+ }
917
+ },
918
+ "Full": {
919
+ "status": "reported",
920
+ "episodes": 50,
921
+ "successes": 28,
922
+ "lscr": 76.0,
923
+ "source": {
924
+ "runId": "mora_v0_62_67bb9aa3551f",
925
+ "perEpisodeFile": "per_task_eval/62_shoebox_accessory_pack/Full/per_episode.jsonl",
926
+ "summaryFile": "per_task_eval/62_shoebox_accessory_pack/Full/summary.json"
927
+ }
928
+ }
929
+ },
930
+ "mean": {
931
+ "status": "reported",
932
+ "sr": 55.5,
933
+ "srScope": "four-channel-mean",
934
+ "srEpisodes": 200,
935
+ "lscr": 70.83333333333333,
936
+ "lscrScope": "four-channel-mean",
937
+ "lscrEpisodes": 200,
938
+ "efficiency": 0.955844550429961,
939
+ "efficiencyStatus": "reported"
940
+ }
941
+ },
942
+ "64": {
943
+ "channels": {
944
+ "None": {
945
+ "status": "reported",
946
+ "episodes": 50,
947
+ "successes": 16,
948
+ "lscr": 50.0,
949
+ "source": {
950
+ "runId": "mora_v0_64_dcaf852fdbc3",
951
+ "perEpisodeFile": "per_task_eval/64_sports_ball_cup_sort/None/per_episode.jsonl",
952
+ "summaryFile": "per_task_eval/64_sports_ball_cup_sort/None/summary.json"
953
+ }
954
+ },
955
+ "Equi.": {
956
+ "status": "reported",
957
+ "episodes": 50,
958
+ "successes": 7,
959
+ "lscr": 36.0,
960
+ "source": {
961
+ "runId": "mora_v0_64_dcaf852fdbc3",
962
+ "perEpisodeFile": "per_task_eval/64_sports_ball_cup_sort/Equi/per_episode.jsonl",
963
+ "summaryFile": "per_task_eval/64_sports_ball_cup_sort/Equi/summary.json"
964
+ }
965
+ },
966
+ "Inv.": {
967
+ "status": "reported",
968
+ "episodes": 50,
969
+ "successes": 8,
970
+ "lscr": 35.0,
971
+ "source": {
972
+ "runId": "mora_v0_64_dcaf852fdbc3",
973
+ "perEpisodeFile": "per_task_eval/64_sports_ball_cup_sort/Inv/per_episode.jsonl",
974
+ "summaryFile": "per_task_eval/64_sports_ball_cup_sort/Inv/summary.json"
975
+ }
976
+ },
977
+ "Full": {
978
+ "status": "reported",
979
+ "episodes": 50,
980
+ "successes": 8,
981
+ "lscr": 26.0,
982
+ "source": {
983
+ "runId": "mora_v0_64_dcaf852fdbc3",
984
+ "perEpisodeFile": "per_task_eval/64_sports_ball_cup_sort/Full/per_episode.jsonl",
985
+ "summaryFile": "per_task_eval/64_sports_ball_cup_sort/Full/summary.json"
986
+ }
987
+ }
988
+ },
989
+ "mean": {
990
+ "status": "reported",
991
+ "sr": 19.5,
992
+ "srScope": "four-channel-mean",
993
+ "srEpisodes": 200,
994
+ "lscr": 36.75,
995
+ "lscrScope": "four-channel-mean",
996
+ "lscrEpisodes": 200,
997
+ "efficiency": 0.7815053828292271,
998
+ "efficiencyStatus": "reported"
999
+ }
1000
+ },
1001
+ "67": {
1002
+ "channels": {
1003
+ "None": {
1004
+ "status": "reported",
1005
+ "episodes": 50,
1006
+ "successes": 14,
1007
+ "lscr": 45.5,
1008
+ "source": {
1009
+ "runId": "mora_v0_67_ff2e261104ed",
1010
+ "perEpisodeFile": "per_task_eval/67_faucet_cup_water_fill/None/per_episode.jsonl",
1011
+ "summaryFile": "per_task_eval/67_faucet_cup_water_fill/None/summary.json"
1012
+ }
1013
+ },
1014
+ "Equi.": {
1015
+ "status": "reported",
1016
+ "episodes": 50,
1017
+ "successes": 15,
1018
+ "lscr": 47.5,
1019
+ "source": {
1020
+ "runId": "mora_v0_67_ff2e261104ed",
1021
+ "perEpisodeFile": "per_task_eval/67_faucet_cup_water_fill/Equi/per_episode.jsonl",
1022
+ "summaryFile": "per_task_eval/67_faucet_cup_water_fill/Equi/summary.json"
1023
+ }
1024
+ },
1025
+ "Inv.": {
1026
+ "status": "reported",
1027
+ "episodes": 50,
1028
+ "successes": 18,
1029
+ "lscr": 51.0,
1030
+ "source": {
1031
+ "runId": "mora_v0_67_ff2e261104ed",
1032
+ "perEpisodeFile": "per_task_eval/67_faucet_cup_water_fill/Inv/per_episode.jsonl",
1033
+ "summaryFile": "per_task_eval/67_faucet_cup_water_fill/Inv/summary.json"
1034
+ }
1035
+ },
1036
+ "Full": {
1037
+ "status": "reported",
1038
+ "episodes": 50,
1039
+ "successes": 8,
1040
+ "lscr": 39.5,
1041
+ "source": {
1042
+ "runId": "mora_v0_67_ff2e261104ed",
1043
+ "perEpisodeFile": "per_task_eval/67_faucet_cup_water_fill/Full/per_episode.jsonl",
1044
+ "summaryFile": "per_task_eval/67_faucet_cup_water_fill/Full/summary.json"
1045
+ }
1046
+ }
1047
+ },
1048
+ "mean": {
1049
+ "status": "reported",
1050
+ "sr": 27.5,
1051
+ "srScope": "four-channel-mean",
1052
+ "srEpisodes": 200,
1053
+ "lscr": 45.875,
1054
+ "lscrScope": "four-channel-mean",
1055
+ "lscrEpisodes": 200,
1056
+ "efficiency": 0.9459495624403429,
1057
+ "efficiencyStatus": "reported"
1058
+ }
1059
+ },
1060
+ "73": {
1061
+ "channels": {
1062
+ "None": {
1063
+ "status": "reported",
1064
+ "episodes": 50,
1065
+ "successes": 1,
1066
+ "lscr": 52.0,
1067
+ "source": {
1068
+ "runId": "mora_v0_73_5525c14a4fae",
1069
+ "perEpisodeFile": "per_task_eval/73_jigsaw_puzzle_assembly/None/per_episode.jsonl",
1070
+ "summaryFile": "per_task_eval/73_jigsaw_puzzle_assembly/None/summary.json"
1071
+ }
1072
+ },
1073
+ "Equi.": {
1074
+ "status": "reported",
1075
+ "episodes": 50,
1076
+ "successes": 0,
1077
+ "lscr": 25.5,
1078
+ "source": {
1079
+ "runId": "mora_v0_73_5525c14a4fae",
1080
+ "perEpisodeFile": "per_task_eval/73_jigsaw_puzzle_assembly/Equi/per_episode.jsonl",
1081
+ "summaryFile": "per_task_eval/73_jigsaw_puzzle_assembly/Equi/summary.json"
1082
+ }
1083
+ },
1084
+ "Inv.": {
1085
+ "status": "reported",
1086
+ "episodes": 50,
1087
+ "successes": 0,
1088
+ "lscr": 34.5,
1089
+ "source": {
1090
+ "runId": "mora_v0_73_5525c14a4fae",
1091
+ "perEpisodeFile": "per_task_eval/73_jigsaw_puzzle_assembly/Inv/per_episode.jsonl",
1092
+ "summaryFile": "per_task_eval/73_jigsaw_puzzle_assembly/Inv/summary.json"
1093
+ }
1094
+ },
1095
+ "Full": {
1096
+ "status": "reported",
1097
+ "episodes": 50,
1098
+ "successes": 0,
1099
+ "lscr": 30.0,
1100
+ "source": {
1101
+ "runId": "mora_v0_73_5525c14a4fae",
1102
+ "perEpisodeFile": "per_task_eval/73_jigsaw_puzzle_assembly/Full/per_episode.jsonl",
1103
+ "summaryFile": "per_task_eval/73_jigsaw_puzzle_assembly/Full/summary.json"
1104
+ }
1105
+ }
1106
+ },
1107
+ "mean": {
1108
+ "status": "reported",
1109
+ "sr": 0.5,
1110
+ "srScope": "four-channel-mean",
1111
+ "srEpisodes": 200,
1112
+ "lscr": 35.5,
1113
+ "lscrScope": "four-channel-mean",
1114
+ "lscrEpisodes": 200,
1115
+ "efficiency": 0.9113448534936138,
1116
+ "efficiencyStatus": "reported"
1117
+ }
1118
+ },
1119
+ "76": {
1120
+ "channels": {
1121
+ "None": {
1122
+ "status": "reported",
1123
+ "episodes": 50,
1124
+ "successes": 1,
1125
+ "lscr": 21.0,
1126
+ "source": {
1127
+ "runId": "mora_v0_76_adab38bae3cc",
1128
+ "perEpisodeFile": "per_task_eval/76_soup_serving/None/per_episode.jsonl",
1129
+ "summaryFile": "per_task_eval/76_soup_serving/None/summary.json"
1130
+ }
1131
+ },
1132
+ "Equi.": {
1133
+ "status": "reported",
1134
+ "episodes": 50,
1135
+ "successes": 0,
1136
+ "lscr": 24.5,
1137
+ "source": {
1138
+ "runId": "mora_v0_76_adab38bae3cc",
1139
+ "perEpisodeFile": "per_task_eval/76_soup_serving/Equi/per_episode.jsonl",
1140
+ "summaryFile": "per_task_eval/76_soup_serving/Equi/summary.json"
1141
+ }
1142
+ },
1143
+ "Inv.": {
1144
+ "status": "reported",
1145
+ "episodes": 50,
1146
+ "successes": 7,
1147
+ "lscr": 28.499999999999996,
1148
+ "source": {
1149
+ "runId": "mora_v0_76_adab38bae3cc",
1150
+ "perEpisodeFile": "per_task_eval/76_soup_serving/Inv/per_episode.jsonl",
1151
+ "summaryFile": "per_task_eval/76_soup_serving/Inv/summary.json"
1152
+ }
1153
+ },
1154
+ "Full": {
1155
+ "status": "reported",
1156
+ "episodes": 50,
1157
+ "successes": 1,
1158
+ "lscr": 18.0,
1159
+ "source": {
1160
+ "runId": "mora_v0_76_adab38bae3cc",
1161
+ "perEpisodeFile": "per_task_eval/76_soup_serving/Full/per_episode.jsonl",
1162
+ "summaryFile": "per_task_eval/76_soup_serving/Full/summary.json"
1163
+ }
1164
+ }
1165
+ },
1166
+ "mean": {
1167
+ "status": "reported",
1168
+ "sr": 4.5,
1169
+ "srScope": "four-channel-mean",
1170
+ "srEpisodes": 200,
1171
+ "lscr": 23.0,
1172
+ "lscrScope": "four-channel-mean",
1173
+ "lscrEpisodes": 200,
1174
+ "efficiency": 0.9461058691827923,
1175
+ "efficiencyStatus": "reported"
1176
+ }
1177
+ },
1178
+ "79": {
1179
+ "channels": {
1180
+ "None": {
1181
+ "status": "reported",
1182
+ "episodes": 50,
1183
+ "successes": 3,
1184
+ "lscr": 16.0,
1185
+ "source": {
1186
+ "runId": "mora_v0_79_3ca6b6de9ea7",
1187
+ "perEpisodeFile": "per_task_eval/79_bimanual_piano_melody/None/per_episode.jsonl",
1188
+ "summaryFile": "per_task_eval/79_bimanual_piano_melody/None/summary.json"
1189
+ }
1190
+ },
1191
+ "Equi.": {
1192
+ "status": "reported",
1193
+ "episodes": 50,
1194
+ "successes": 7,
1195
+ "lscr": 20.0,
1196
+ "source": {
1197
+ "runId": "mora_v0_79_3ca6b6de9ea7",
1198
+ "perEpisodeFile": "per_task_eval/79_bimanual_piano_melody/Equi/per_episode.jsonl",
1199
+ "summaryFile": "per_task_eval/79_bimanual_piano_melody/Equi/summary.json"
1200
+ }
1201
+ },
1202
+ "Inv.": {
1203
+ "status": "reported",
1204
+ "episodes": 50,
1205
+ "successes": 3,
1206
+ "lscr": 12.0,
1207
+ "source": {
1208
+ "runId": "mora_v0_79_3ca6b6de9ea7",
1209
+ "perEpisodeFile": "per_task_eval/79_bimanual_piano_melody/Inv/per_episode.jsonl",
1210
+ "summaryFile": "per_task_eval/79_bimanual_piano_melody/Inv/summary.json"
1211
+ }
1212
+ },
1213
+ "Full": {
1214
+ "status": "reported",
1215
+ "episodes": 50,
1216
+ "successes": 14,
1217
+ "lscr": 34.0,
1218
+ "source": {
1219
+ "runId": "mora_v0_79_3ca6b6de9ea7",
1220
+ "perEpisodeFile": "per_task_eval/79_bimanual_piano_melody/Full/per_episode.jsonl",
1221
+ "summaryFile": "per_task_eval/79_bimanual_piano_melody/Full/summary.json"
1222
+ }
1223
+ }
1224
+ },
1225
+ "mean": {
1226
+ "status": "reported",
1227
+ "sr": 13.5,
1228
+ "srScope": "four-channel-mean",
1229
+ "srEpisodes": 200,
1230
+ "lscr": 20.5,
1231
+ "lscrScope": "four-channel-mean",
1232
+ "lscrEpisodes": 200,
1233
+ "efficiency": 0.9060536625027719,
1234
+ "efficiencyStatus": "reported"
1235
+ }
1236
+ },
1237
+ "80": {
1238
+ "channels": {
1239
+ "None": {
1240
+ "status": "reported",
1241
+ "episodes": 50,
1242
+ "successes": 41,
1243
+ "lscr": 87.33333333333333,
1244
+ "source": {
1245
+ "runId": "mora_v0_80_f1166cc6864b",
1246
+ "perEpisodeFile": "per_task_eval/80_gaming_desk_setup/None/per_episode.jsonl",
1247
+ "summaryFile": "per_task_eval/80_gaming_desk_setup/None/summary.json"
1248
+ }
1249
+ },
1250
+ "Equi.": {
1251
+ "status": "reported",
1252
+ "episodes": 50,
1253
+ "successes": 31,
1254
+ "lscr": 70.0,
1255
+ "source": {
1256
+ "runId": "mora_v0_80_f1166cc6864b",
1257
+ "perEpisodeFile": "per_task_eval/80_gaming_desk_setup/Equi/per_episode.jsonl",
1258
+ "summaryFile": "per_task_eval/80_gaming_desk_setup/Equi/summary.json"
1259
+ }
1260
+ },
1261
+ "Inv.": {
1262
+ "status": "reported",
1263
+ "episodes": 50,
1264
+ "successes": 25,
1265
+ "lscr": 60.66666666666667,
1266
+ "source": {
1267
+ "runId": "mora_v0_80_f1166cc6864b",
1268
+ "perEpisodeFile": "per_task_eval/80_gaming_desk_setup/Inv/per_episode.jsonl",
1269
+ "summaryFile": "per_task_eval/80_gaming_desk_setup/Inv/summary.json"
1270
+ }
1271
+ },
1272
+ "Full": {
1273
+ "status": "reported",
1274
+ "episodes": 50,
1275
+ "successes": 31,
1276
+ "lscr": 76.0,
1277
+ "source": {
1278
+ "runId": "mora_v0_80_f1166cc6864b",
1279
+ "perEpisodeFile": "per_task_eval/80_gaming_desk_setup/Full/per_episode.jsonl",
1280
+ "summaryFile": "per_task_eval/80_gaming_desk_setup/Full/summary.json"
1281
+ }
1282
+ }
1283
+ },
1284
+ "mean": {
1285
+ "status": "reported",
1286
+ "sr": 64.0,
1287
+ "srScope": "four-channel-mean",
1288
+ "srEpisodes": 200,
1289
+ "lscr": 73.5,
1290
+ "lscrScope": "four-channel-mean",
1291
+ "lscrEpisodes": 200,
1292
+ "efficiency": 1.0431624989683914,
1293
+ "efficiencyStatus": "reported"
1294
+ }
1295
+ },
1296
+ "03": {
1297
+ "channels": {
1298
+ "None": {
1299
+ "status": "reported",
1300
+ "episodes": 50,
1301
+ "successes": 18,
1302
+ "lscr": 72.66666666666667,
1303
+ "source": {
1304
+ "runId": "mora_v0_03_ec7d439e3746",
1305
+ "perEpisodeFile": "per_task_eval/03_wine_glass_plate_balance/None/per_episode.jsonl",
1306
+ "summaryFile": "per_task_eval/03_wine_glass_plate_balance/None/summary.json"
1307
+ }
1308
+ },
1309
+ "Equi.": {
1310
+ "status": "reported",
1311
+ "episodes": 50,
1312
+ "successes": 13,
1313
+ "lscr": 74.66666666666667,
1314
+ "source": {
1315
+ "runId": "mora_v0_03_ec7d439e3746",
1316
+ "perEpisodeFile": "per_task_eval/03_wine_glass_plate_balance/Equi/per_episode.jsonl",
1317
+ "summaryFile": "per_task_eval/03_wine_glass_plate_balance/Equi/summary.json"
1318
+ }
1319
+ },
1320
+ "Inv.": {
1321
+ "status": "reported",
1322
+ "episodes": 50,
1323
+ "successes": 13,
1324
+ "lscr": 73.33333333333333,
1325
+ "source": {
1326
+ "runId": "mora_v0_03_ec7d439e3746",
1327
+ "perEpisodeFile": "per_task_eval/03_wine_glass_plate_balance/Inv/per_episode.jsonl",
1328
+ "summaryFile": "per_task_eval/03_wine_glass_plate_balance/Inv/summary.json"
1329
+ }
1330
+ },
1331
+ "Full": {
1332
+ "status": "reported",
1333
+ "episodes": 50,
1334
+ "successes": 12,
1335
+ "lscr": 80.66666666666666,
1336
+ "source": {
1337
+ "runId": "mora_v0_03_ec7d439e3746",
1338
+ "perEpisodeFile": "per_task_eval/03_wine_glass_plate_balance/Full/per_episode.jsonl",
1339
+ "summaryFile": "per_task_eval/03_wine_glass_plate_balance/Full/summary.json"
1340
+ }
1341
+ }
1342
+ },
1343
+ "mean": {
1344
+ "status": "reported",
1345
+ "sr": 28.0,
1346
+ "srScope": "four-channel-mean",
1347
+ "srEpisodes": 200,
1348
+ "lscr": 75.33333333333333,
1349
+ "lscrScope": "four-channel-mean",
1350
+ "lscrEpisodes": 200,
1351
+ "efficiency": 0.987934520315659,
1352
+ "efficiencyStatus": "reported"
1353
+ }
1354
+ },
1355
+ "06": {
1356
+ "channels": {
1357
+ "None": {
1358
+ "status": "reported",
1359
+ "episodes": 50,
1360
+ "successes": 36,
1361
+ "lscr": 88.0,
1362
+ "source": {
1363
+ "runId": "mora_v0_06_04465b5911a4",
1364
+ "perEpisodeFile": "per_task_eval/06_fruit_bowl_loading/None/per_episode.jsonl",
1365
+ "summaryFile": "per_task_eval/06_fruit_bowl_loading/None/summary.json"
1366
+ }
1367
+ },
1368
+ "Equi.": {
1369
+ "status": "reported",
1370
+ "episodes": 50,
1371
+ "successes": 17,
1372
+ "lscr": 72.5,
1373
+ "source": {
1374
+ "runId": "mora_v0_06_04465b5911a4",
1375
+ "perEpisodeFile": "per_task_eval/06_fruit_bowl_loading/Equi/per_episode.jsonl",
1376
+ "summaryFile": "per_task_eval/06_fruit_bowl_loading/Equi/summary.json"
1377
+ }
1378
+ },
1379
+ "Inv.": {
1380
+ "status": "reported",
1381
+ "episodes": 50,
1382
+ "successes": 31,
1383
+ "lscr": 83.5,
1384
+ "source": {
1385
+ "runId": "mora_v0_06_04465b5911a4",
1386
+ "perEpisodeFile": "per_task_eval/06_fruit_bowl_loading/Inv/per_episode.jsonl",
1387
+ "summaryFile": "per_task_eval/06_fruit_bowl_loading/Inv/summary.json"
1388
+ }
1389
+ },
1390
+ "Full": {
1391
+ "status": "reported",
1392
+ "episodes": 50,
1393
+ "successes": 31,
1394
+ "lscr": 82.5,
1395
+ "source": {
1396
+ "runId": "mora_v0_06_04465b5911a4",
1397
+ "perEpisodeFile": "per_task_eval/06_fruit_bowl_loading/Full/per_episode.jsonl",
1398
+ "summaryFile": "per_task_eval/06_fruit_bowl_loading/Full/summary.json"
1399
+ }
1400
+ }
1401
+ },
1402
+ "mean": {
1403
+ "status": "reported",
1404
+ "sr": 57.5,
1405
+ "srScope": "four-channel-mean",
1406
+ "srEpisodes": 200,
1407
+ "lscr": 81.625,
1408
+ "lscrScope": "four-channel-mean",
1409
+ "lscrEpisodes": 200,
1410
+ "efficiency": 0.9356844099213059,
1411
+ "efficiencyStatus": "reported"
1412
+ }
1413
+ },
1414
+ "07": {
1415
+ "channels": {
1416
+ "None": {
1417
+ "status": "reported",
1418
+ "episodes": 50,
1419
+ "successes": 33,
1420
+ "lscr": 77.5,
1421
+ "source": {
1422
+ "runId": "mora_v0_07_0a15bbe47bcb",
1423
+ "perEpisodeFile": "per_task_eval/07_citrus_plate_loading/None/per_episode.jsonl",
1424
+ "summaryFile": "per_task_eval/07_citrus_plate_loading/None/summary.json"
1425
+ }
1426
+ },
1427
+ "Equi.": {
1428
+ "status": "reported",
1429
+ "episodes": 50,
1430
+ "successes": 27,
1431
+ "lscr": 68.5,
1432
+ "source": {
1433
+ "runId": "mora_v0_07_0a15bbe47bcb",
1434
+ "perEpisodeFile": "per_task_eval/07_citrus_plate_loading/Equi/per_episode.jsonl",
1435
+ "summaryFile": "per_task_eval/07_citrus_plate_loading/Equi/summary.json"
1436
+ }
1437
+ },
1438
+ "Inv.": {
1439
+ "status": "reported",
1440
+ "episodes": 50,
1441
+ "successes": 31,
1442
+ "lscr": 80.5,
1443
+ "source": {
1444
+ "runId": "mora_v0_07_0a15bbe47bcb",
1445
+ "perEpisodeFile": "per_task_eval/07_citrus_plate_loading/Inv/per_episode.jsonl",
1446
+ "summaryFile": "per_task_eval/07_citrus_plate_loading/Inv/summary.json"
1447
+ }
1448
+ },
1449
+ "Full": {
1450
+ "status": "reported",
1451
+ "episodes": 50,
1452
+ "successes": 25,
1453
+ "lscr": 68.5,
1454
+ "source": {
1455
+ "runId": "mora_v0_07_0a15bbe47bcb",
1456
+ "perEpisodeFile": "per_task_eval/07_citrus_plate_loading/Full/per_episode.jsonl",
1457
+ "summaryFile": "per_task_eval/07_citrus_plate_loading/Full/summary.json"
1458
+ }
1459
+ }
1460
+ },
1461
+ "mean": {
1462
+ "status": "reported",
1463
+ "sr": 58.0,
1464
+ "srScope": "four-channel-mean",
1465
+ "srEpisodes": 200,
1466
+ "lscr": 73.75,
1467
+ "lscrScope": "four-channel-mean",
1468
+ "lscrEpisodes": 200,
1469
+ "efficiency": 0.8367441163243892,
1470
+ "efficiencyStatus": "reported"
1471
+ }
1472
+ },
1473
+ "08": {
1474
+ "channels": {
1475
+ "None": {
1476
+ "status": "reported",
1477
+ "episodes": 50,
1478
+ "successes": 35,
1479
+ "lscr": 82.0,
1480
+ "source": {
1481
+ "runId": "mora_v0_08_071ef9e79da9",
1482
+ "perEpisodeFile": "per_task_eval/08_frypan_stand_pour/None/per_episode.jsonl",
1483
+ "summaryFile": "per_task_eval/08_frypan_stand_pour/None/summary.json"
1484
+ }
1485
+ },
1486
+ "Equi.": {
1487
+ "status": "reported",
1488
+ "episodes": 50,
1489
+ "successes": 23,
1490
+ "lscr": 69.0,
1491
+ "source": {
1492
+ "runId": "mora_v0_08_071ef9e79da9",
1493
+ "perEpisodeFile": "per_task_eval/08_frypan_stand_pour/Equi/per_episode.jsonl",
1494
+ "summaryFile": "per_task_eval/08_frypan_stand_pour/Equi/summary.json"
1495
+ }
1496
+ },
1497
+ "Inv.": {
1498
+ "status": "reported",
1499
+ "episodes": 50,
1500
+ "successes": 27,
1501
+ "lscr": 80.0,
1502
+ "source": {
1503
+ "runId": "mora_v0_08_071ef9e79da9",
1504
+ "perEpisodeFile": "per_task_eval/08_frypan_stand_pour/Inv/per_episode.jsonl",
1505
+ "summaryFile": "per_task_eval/08_frypan_stand_pour/Inv/summary.json"
1506
+ }
1507
+ },
1508
+ "Full": {
1509
+ "status": "reported",
1510
+ "episodes": 50,
1511
+ "successes": 34,
1512
+ "lscr": 77.0,
1513
+ "source": {
1514
+ "runId": "mora_v0_08_071ef9e79da9",
1515
+ "perEpisodeFile": "per_task_eval/08_frypan_stand_pour/Full/per_episode.jsonl",
1516
+ "summaryFile": "per_task_eval/08_frypan_stand_pour/Full/summary.json"
1517
+ }
1518
+ }
1519
+ },
1520
+ "mean": {
1521
+ "status": "reported",
1522
+ "sr": 59.5,
1523
+ "srScope": "four-channel-mean",
1524
+ "srEpisodes": 200,
1525
+ "lscr": 77.0,
1526
+ "lscrScope": "four-channel-mean",
1527
+ "lscrEpisodes": 200,
1528
+ "efficiency": 1.2646990514942913,
1529
+ "efficiencyStatus": "reported"
1530
+ }
1531
+ },
1532
+ "09": {
1533
+ "channels": {
1534
+ "None": {
1535
+ "status": "reported",
1536
+ "episodes": 50,
1537
+ "successes": 41,
1538
+ "lscr": 1.0,
1539
+ "source": {
1540
+ "runId": "mora_v0_09_12606d72d3f9",
1541
+ "perEpisodeFile": "per_task_eval/09_cleaner_moisturizer_box_loading/None/per_episode.jsonl",
1542
+ "summaryFile": "per_task_eval/09_cleaner_moisturizer_box_loading/None/summary.json"
1543
+ }
1544
+ },
1545
+ "Equi.": {
1546
+ "status": "reported",
1547
+ "episodes": 50,
1548
+ "successes": 26,
1549
+ "lscr": 0.0,
1550
+ "source": {
1551
+ "runId": "mora_v0_09_12606d72d3f9",
1552
+ "perEpisodeFile": "per_task_eval/09_cleaner_moisturizer_box_loading/Equi/per_episode.jsonl",
1553
+ "summaryFile": "per_task_eval/09_cleaner_moisturizer_box_loading/Equi/summary.json"
1554
+ }
1555
+ },
1556
+ "Inv.": {
1557
+ "status": "reported",
1558
+ "episodes": 50,
1559
+ "successes": 30,
1560
+ "lscr": 2.0,
1561
+ "source": {
1562
+ "runId": "mora_v0_09_12606d72d3f9",
1563
+ "perEpisodeFile": "per_task_eval/09_cleaner_moisturizer_box_loading/Inv/per_episode.jsonl",
1564
+ "summaryFile": "per_task_eval/09_cleaner_moisturizer_box_loading/Inv/summary.json"
1565
+ }
1566
+ },
1567
+ "Full": {
1568
+ "status": "reported",
1569
+ "episodes": 50,
1570
+ "successes": 26,
1571
+ "lscr": 2.0,
1572
+ "source": {
1573
+ "runId": "mora_v0_09_12606d72d3f9",
1574
+ "perEpisodeFile": "per_task_eval/09_cleaner_moisturizer_box_loading/Full/per_episode.jsonl",
1575
+ "summaryFile": "per_task_eval/09_cleaner_moisturizer_box_loading/Full/summary.json"
1576
+ }
1577
+ }
1578
+ },
1579
+ "mean": {
1580
+ "status": "reported",
1581
+ "sr": 61.5,
1582
+ "srScope": "four-channel-mean",
1583
+ "srEpisodes": 200,
1584
+ "lscr": 1.25,
1585
+ "lscrScope": "four-channel-mean",
1586
+ "lscrEpisodes": 200,
1587
+ "efficiency": 0.9382837165331859,
1588
+ "efficiencyStatus": "reported"
1589
+ }
1590
+ }
1591
+ },
1592
+ "evaluation": {
1593
+ "schemaVersion": "1.0",
1594
+ "isTemplate": false,
1595
+ "protocol": {
1596
+ "benchmarkCommit": null,
1597
+ "taskCount": 26,
1598
+ "channels": [
1599
+ "None",
1600
+ "Equi.",
1601
+ "Inv.",
1602
+ "Full"
1603
+ ],
1604
+ "episodesPerChannel": 50,
1605
+ "seeds": {
1606
+ "03": {
1607
+ "None": {
1608
+ "base": 100300000,
1609
+ "first": 1,
1610
+ "last": 50
1611
+ },
1612
+ "Equi.": {
1613
+ "base": 100300000,
1614
+ "first": 1,
1615
+ "last": 50
1616
+ },
1617
+ "Inv.": {
1618
+ "base": 100300000,
1619
+ "first": 1,
1620
+ "last": 50
1621
+ },
1622
+ "Full": {
1623
+ "base": 200300000,
1624
+ "first": 1,
1625
+ "last": 50
1626
+ }
1627
+ },
1628
+ "06": {
1629
+ "None": {
1630
+ "base": 100600000,
1631
+ "first": 1,
1632
+ "last": 50
1633
+ },
1634
+ "Equi.": {
1635
+ "base": 100600000,
1636
+ "first": 1,
1637
+ "last": 50
1638
+ },
1639
+ "Inv.": {
1640
+ "base": 100600000,
1641
+ "first": 1,
1642
+ "last": 50
1643
+ },
1644
+ "Full": {
1645
+ "base": 200600000,
1646
+ "first": 1,
1647
+ "last": 50
1648
+ }
1649
+ },
1650
+ "07": {
1651
+ "None": {
1652
+ "base": 100700000,
1653
+ "first": 1,
1654
+ "last": 50
1655
+ },
1656
+ "Equi.": {
1657
+ "base": 100700000,
1658
+ "first": 1,
1659
+ "last": 50
1660
+ },
1661
+ "Inv.": {
1662
+ "base": 100700000,
1663
+ "first": 1,
1664
+ "last": 50
1665
+ },
1666
+ "Full": {
1667
+ "base": 200700000,
1668
+ "first": 1,
1669
+ "last": 50
1670
+ }
1671
+ },
1672
+ "08": {
1673
+ "None": {
1674
+ "base": 100800000,
1675
+ "first": 1,
1676
+ "last": 50
1677
+ },
1678
+ "Equi.": {
1679
+ "base": 100800000,
1680
+ "first": 1,
1681
+ "last": 50
1682
+ },
1683
+ "Inv.": {
1684
+ "base": 100800000,
1685
+ "first": 1,
1686
+ "last": 50
1687
+ },
1688
+ "Full": {
1689
+ "base": 200800000,
1690
+ "first": 1,
1691
+ "last": 50
1692
+ }
1693
+ },
1694
+ "09": {
1695
+ "None": {
1696
+ "base": 100900000,
1697
+ "first": 1,
1698
+ "last": 50
1699
+ },
1700
+ "Equi.": {
1701
+ "base": 100900000,
1702
+ "first": 1,
1703
+ "last": 50
1704
+ },
1705
+ "Inv.": {
1706
+ "base": 100900000,
1707
+ "first": 1,
1708
+ "last": 50
1709
+ },
1710
+ "Full": {
1711
+ "base": 200900000,
1712
+ "first": 1,
1713
+ "last": 50
1714
+ }
1715
+ },
1716
+ "12": {
1717
+ "None": {
1718
+ "base": 101200000,
1719
+ "first": 1,
1720
+ "last": 50
1721
+ },
1722
+ "Equi.": {
1723
+ "base": 101200000,
1724
+ "first": 1,
1725
+ "last": 50
1726
+ },
1727
+ "Inv.": {
1728
+ "base": 101200000,
1729
+ "first": 1,
1730
+ "last": 50
1731
+ },
1732
+ "Full": {
1733
+ "base": 201200000,
1734
+ "first": 1,
1735
+ "last": 50
1736
+ }
1737
+ },
1738
+ "21": {
1739
+ "None": {
1740
+ "base": 102100000,
1741
+ "first": 1,
1742
+ "last": 50
1743
+ },
1744
+ "Equi.": {
1745
+ "base": 102100000,
1746
+ "first": 1,
1747
+ "last": 50
1748
+ },
1749
+ "Inv.": {
1750
+ "base": 102100000,
1751
+ "first": 1,
1752
+ "last": 50
1753
+ },
1754
+ "Full": {
1755
+ "base": 202100000,
1756
+ "first": 1,
1757
+ "last": 50
1758
+ }
1759
+ },
1760
+ "22": {
1761
+ "None": {
1762
+ "base": 102200000,
1763
+ "first": 1,
1764
+ "last": 50
1765
+ },
1766
+ "Equi.": {
1767
+ "base": 102200000,
1768
+ "first": 1,
1769
+ "last": 50
1770
+ },
1771
+ "Inv.": {
1772
+ "base": 102200000,
1773
+ "first": 1,
1774
+ "last": 50
1775
+ },
1776
+ "Full": {
1777
+ "base": 202200000,
1778
+ "first": 1,
1779
+ "last": 50
1780
+ }
1781
+ },
1782
+ "24": {
1783
+ "None": {
1784
+ "base": 102400000,
1785
+ "first": 1,
1786
+ "last": 50
1787
+ },
1788
+ "Equi.": {
1789
+ "base": 102400000,
1790
+ "first": 1,
1791
+ "last": 50
1792
+ },
1793
+ "Inv.": {
1794
+ "base": 102400000,
1795
+ "first": 1,
1796
+ "last": 50
1797
+ },
1798
+ "Full": {
1799
+ "base": 202400000,
1800
+ "first": 1,
1801
+ "last": 50
1802
+ }
1803
+ },
1804
+ "26": {
1805
+ "None": {
1806
+ "base": 102600000,
1807
+ "first": 1,
1808
+ "last": 50
1809
+ },
1810
+ "Equi.": {
1811
+ "base": 102600000,
1812
+ "first": 1,
1813
+ "last": 50
1814
+ },
1815
+ "Inv.": {
1816
+ "base": 102600000,
1817
+ "first": 1,
1818
+ "last": 50
1819
+ },
1820
+ "Full": {
1821
+ "base": 202600000,
1822
+ "first": 1,
1823
+ "last": 50
1824
+ }
1825
+ },
1826
+ "27": {
1827
+ "None": {
1828
+ "base": 102700000,
1829
+ "first": 1,
1830
+ "last": 50
1831
+ },
1832
+ "Equi.": {
1833
+ "base": 102700000,
1834
+ "first": 1,
1835
+ "last": 50
1836
+ },
1837
+ "Inv.": {
1838
+ "base": 102700000,
1839
+ "first": 1,
1840
+ "last": 50
1841
+ },
1842
+ "Full": {
1843
+ "base": 202700000,
1844
+ "first": 1,
1845
+ "last": 50
1846
+ }
1847
+ },
1848
+ "32": {
1849
+ "None": {
1850
+ "base": 103200000,
1851
+ "first": 1,
1852
+ "last": 50
1853
+ },
1854
+ "Equi.": {
1855
+ "base": 103200000,
1856
+ "first": 1,
1857
+ "last": 50
1858
+ },
1859
+ "Inv.": {
1860
+ "base": 103200000,
1861
+ "first": 1,
1862
+ "last": 50
1863
+ },
1864
+ "Full": {
1865
+ "base": 203200000,
1866
+ "first": 1,
1867
+ "last": 50
1868
+ }
1869
+ },
1870
+ "34": {
1871
+ "None": {
1872
+ "base": 103400000,
1873
+ "first": 1,
1874
+ "last": 50
1875
+ },
1876
+ "Equi.": {
1877
+ "base": 103400000,
1878
+ "first": 1,
1879
+ "last": 50
1880
+ },
1881
+ "Inv.": {
1882
+ "base": 103400000,
1883
+ "first": 1,
1884
+ "last": 50
1885
+ },
1886
+ "Full": {
1887
+ "base": 203400000,
1888
+ "first": 1,
1889
+ "last": 50
1890
+ }
1891
+ },
1892
+ "42": {
1893
+ "None": {
1894
+ "base": 104200000,
1895
+ "first": 1,
1896
+ "last": 50
1897
+ },
1898
+ "Equi.": {
1899
+ "base": 104200000,
1900
+ "first": 1,
1901
+ "last": 50
1902
+ },
1903
+ "Inv.": {
1904
+ "base": 104200000,
1905
+ "first": 1,
1906
+ "last": 50
1907
+ },
1908
+ "Full": {
1909
+ "base": 204200000,
1910
+ "first": 1,
1911
+ "last": 50
1912
+ }
1913
+ },
1914
+ "43": {
1915
+ "None": {
1916
+ "base": 104300000,
1917
+ "first": 1,
1918
+ "last": 50
1919
+ },
1920
+ "Equi.": {
1921
+ "base": 104300000,
1922
+ "first": 1,
1923
+ "last": 50
1924
+ },
1925
+ "Inv.": {
1926
+ "base": 104300000,
1927
+ "first": 1,
1928
+ "last": 50
1929
+ },
1930
+ "Full": {
1931
+ "base": 204300000,
1932
+ "first": 1,
1933
+ "last": 50
1934
+ }
1935
+ },
1936
+ "44": {
1937
+ "None": {
1938
+ "base": 104400000,
1939
+ "first": 1,
1940
+ "last": 50
1941
+ },
1942
+ "Equi.": {
1943
+ "base": 104400000,
1944
+ "first": 1,
1945
+ "last": 50
1946
+ },
1947
+ "Inv.": {
1948
+ "base": 104400000,
1949
+ "first": 1,
1950
+ "last": 50
1951
+ },
1952
+ "Full": {
1953
+ "base": 204400000,
1954
+ "first": 1,
1955
+ "last": 50
1956
+ }
1957
+ },
1958
+ "51": {
1959
+ "None": {
1960
+ "base": 105100000,
1961
+ "first": 1,
1962
+ "last": 50
1963
+ },
1964
+ "Equi.": {
1965
+ "base": 105100000,
1966
+ "first": 1,
1967
+ "last": 50
1968
+ },
1969
+ "Inv.": {
1970
+ "base": 105100000,
1971
+ "first": 1,
1972
+ "last": 50
1973
+ },
1974
+ "Full": {
1975
+ "base": 205100000,
1976
+ "first": 1,
1977
+ "last": 50
1978
+ }
1979
+ },
1980
+ "60": {
1981
+ "None": {
1982
+ "base": 106000000,
1983
+ "first": 1,
1984
+ "last": 50
1985
+ },
1986
+ "Equi.": {
1987
+ "base": 106000000,
1988
+ "first": 1,
1989
+ "last": 50
1990
+ },
1991
+ "Inv.": {
1992
+ "base": 106000000,
1993
+ "first": 1,
1994
+ "last": 50
1995
+ },
1996
+ "Full": {
1997
+ "base": 206000000,
1998
+ "first": 1,
1999
+ "last": 50
2000
+ }
2001
+ },
2002
+ "61": {
2003
+ "None": {
2004
+ "base": 106100000,
2005
+ "first": 1,
2006
+ "last": 50
2007
+ },
2008
+ "Equi.": {
2009
+ "base": 106100000,
2010
+ "first": 1,
2011
+ "last": 50
2012
+ },
2013
+ "Inv.": {
2014
+ "base": 106100000,
2015
+ "first": 1,
2016
+ "last": 50
2017
+ },
2018
+ "Full": {
2019
+ "base": 206100000,
2020
+ "first": 1,
2021
+ "last": 50
2022
+ }
2023
+ },
2024
+ "62": {
2025
+ "None": {
2026
+ "base": 106200000,
2027
+ "first": 1,
2028
+ "last": 50
2029
+ },
2030
+ "Equi.": {
2031
+ "base": 106200000,
2032
+ "first": 1,
2033
+ "last": 50
2034
+ },
2035
+ "Inv.": {
2036
+ "base": 106200000,
2037
+ "first": 1,
2038
+ "last": 50
2039
+ },
2040
+ "Full": {
2041
+ "base": 206200000,
2042
+ "first": 1,
2043
+ "last": 50
2044
+ }
2045
+ },
2046
+ "64": {
2047
+ "None": {
2048
+ "base": 106400000,
2049
+ "first": 1,
2050
+ "last": 50
2051
+ },
2052
+ "Equi.": {
2053
+ "base": 106400000,
2054
+ "first": 1,
2055
+ "last": 50
2056
+ },
2057
+ "Inv.": {
2058
+ "base": 106400000,
2059
+ "first": 1,
2060
+ "last": 50
2061
+ },
2062
+ "Full": {
2063
+ "base": 206400000,
2064
+ "first": 1,
2065
+ "last": 50
2066
+ }
2067
+ },
2068
+ "67": {
2069
+ "None": {
2070
+ "base": 106700000,
2071
+ "first": 1,
2072
+ "last": 50
2073
+ },
2074
+ "Equi.": {
2075
+ "base": 106700000,
2076
+ "first": 1,
2077
+ "last": 50
2078
+ },
2079
+ "Inv.": {
2080
+ "base": 106700000,
2081
+ "first": 1,
2082
+ "last": 50
2083
+ },
2084
+ "Full": {
2085
+ "base": 206700000,
2086
+ "first": 1,
2087
+ "last": 50
2088
+ }
2089
+ },
2090
+ "73": {
2091
+ "None": {
2092
+ "base": 107300000,
2093
+ "first": 1,
2094
+ "last": 50
2095
+ },
2096
+ "Equi.": {
2097
+ "base": 107300000,
2098
+ "first": 1,
2099
+ "last": 50
2100
+ },
2101
+ "Inv.": {
2102
+ "base": 107300000,
2103
+ "first": 1,
2104
+ "last": 50
2105
+ },
2106
+ "Full": {
2107
+ "base": 207300000,
2108
+ "first": 1,
2109
+ "last": 50
2110
+ }
2111
+ },
2112
+ "76": {
2113
+ "None": {
2114
+ "base": 107600000,
2115
+ "first": 1,
2116
+ "last": 50
2117
+ },
2118
+ "Equi.": {
2119
+ "base": 107600000,
2120
+ "first": 1,
2121
+ "last": 50
2122
+ },
2123
+ "Inv.": {
2124
+ "base": 107600000,
2125
+ "first": 1,
2126
+ "last": 50
2127
+ },
2128
+ "Full": {
2129
+ "base": 207600000,
2130
+ "first": 1,
2131
+ "last": 50
2132
+ }
2133
+ },
2134
+ "79": {
2135
+ "None": {
2136
+ "base": 107900000,
2137
+ "first": 1,
2138
+ "last": 50
2139
+ },
2140
+ "Equi.": {
2141
+ "base": 107900000,
2142
+ "first": 1,
2143
+ "last": 50
2144
+ },
2145
+ "Inv.": {
2146
+ "base": 107900000,
2147
+ "first": 1,
2148
+ "last": 50
2149
+ },
2150
+ "Full": {
2151
+ "base": 207900000,
2152
+ "first": 1,
2153
+ "last": 50
2154
+ }
2155
+ },
2156
+ "80": {
2157
+ "None": {
2158
+ "base": 108000000,
2159
+ "first": 1,
2160
+ "last": 50
2161
+ },
2162
+ "Equi.": {
2163
+ "base": 108000000,
2164
+ "first": 1,
2165
+ "last": 50
2166
+ },
2167
+ "Inv.": {
2168
+ "base": 108000000,
2169
+ "first": 1,
2170
+ "last": 50
2171
+ },
2172
+ "Full": {
2173
+ "base": 208000000,
2174
+ "first": 1,
2175
+ "last": 50
2176
+ }
2177
+ }
2178
+ },
2179
+ "stepUnit": "physics_steps",
2180
+ "physicsHz": 60,
2181
+ "controlHz": 20,
2182
+ "executionStepBudget": {
2183
+ "03": {
2184
+ "None": 3246,
2185
+ "Equi.": 3246,
2186
+ "Inv.": 3246,
2187
+ "Full": 3246
2188
+ },
2189
+ "06": {
2190
+ "None": 1920,
2191
+ "Equi.": 1920,
2192
+ "Inv.": 1920,
2193
+ "Full": 1920
2194
+ },
2195
+ "07": {
2196
+ "None": 1806,
2197
+ "Equi.": 1806,
2198
+ "Inv.": 1806,
2199
+ "Full": 1806
2200
+ },
2201
+ "08": {
2202
+ "None": 3432,
2203
+ "Equi.": 3432,
2204
+ "Inv.": 3432,
2205
+ "Full": 3432
2206
+ },
2207
+ "09": {
2208
+ "None": 1926,
2209
+ "Equi.": 1926,
2210
+ "Inv.": 1926,
2211
+ "Full": 1926
2212
+ },
2213
+ "12": {
2214
+ "None": 3447,
2215
+ "Equi.": 3447,
2216
+ "Inv.": 3447,
2217
+ "Full": 3447
2218
+ },
2219
+ "21": {
2220
+ "None": 2481,
2221
+ "Equi.": 2481,
2222
+ "Inv.": 2481,
2223
+ "Full": 2481
2224
+ },
2225
+ "22": {
2226
+ "None": 2043,
2227
+ "Equi.": 2043,
2228
+ "Inv.": 2043,
2229
+ "Full": 2043
2230
+ },
2231
+ "24": {
2232
+ "None": 4626,
2233
+ "Equi.": 4626,
2234
+ "Inv.": 4626,
2235
+ "Full": 4626
2236
+ },
2237
+ "26": {
2238
+ "None": 2613,
2239
+ "Equi.": 2613,
2240
+ "Inv.": 2613,
2241
+ "Full": 2613
2242
+ },
2243
+ "27": {
2244
+ "None": 3570,
2245
+ "Equi.": 3570,
2246
+ "Inv.": 3570,
2247
+ "Full": 3570
2248
+ },
2249
+ "32": {
2250
+ "None": 2556,
2251
+ "Equi.": 2556,
2252
+ "Inv.": 2556,
2253
+ "Full": 2556
2254
+ },
2255
+ "34": {
2256
+ "None": 3240,
2257
+ "Equi.": 3240,
2258
+ "Inv.": 3240,
2259
+ "Full": 3240
2260
+ },
2261
+ "42": {
2262
+ "None": 2004,
2263
+ "Equi.": 2004,
2264
+ "Inv.": 2004,
2265
+ "Full": 2004
2266
+ },
2267
+ "43": {
2268
+ "None": 4125,
2269
+ "Equi.": 4125,
2270
+ "Inv.": 4125,
2271
+ "Full": 4125
2272
+ },
2273
+ "44": {
2274
+ "None": 3183,
2275
+ "Equi.": 3183,
2276
+ "Inv.": 3183,
2277
+ "Full": 3183
2278
+ },
2279
+ "51": {
2280
+ "None": 3069,
2281
+ "Equi.": 3069,
2282
+ "Inv.": 3069,
2283
+ "Full": 3069
2284
+ },
2285
+ "60": {
2286
+ "None": 2985,
2287
+ "Equi.": 2985,
2288
+ "Inv.": 2985,
2289
+ "Full": 2985
2290
+ },
2291
+ "61": {
2292
+ "None": 3102,
2293
+ "Equi.": 3102,
2294
+ "Inv.": 3102,
2295
+ "Full": 3102
2296
+ },
2297
+ "62": {
2298
+ "None": 3231,
2299
+ "Equi.": 3231,
2300
+ "Inv.": 3231,
2301
+ "Full": 3231
2302
+ },
2303
+ "64": {
2304
+ "None": 1995,
2305
+ "Equi.": 1995,
2306
+ "Inv.": 1995,
2307
+ "Full": 1995
2308
+ },
2309
+ "67": {
2310
+ "None": 2892,
2311
+ "Equi.": 2892,
2312
+ "Inv.": 2892,
2313
+ "Full": 2892
2314
+ },
2315
+ "73": {
2316
+ "None": 3639,
2317
+ "Equi.": 3639,
2318
+ "Inv.": 3639,
2319
+ "Full": 3639
2320
+ },
2321
+ "76": {
2322
+ "None": 2958,
2323
+ "Equi.": 2958,
2324
+ "Inv.": 2958,
2325
+ "Full": 2958
2326
+ },
2327
+ "79": {
2328
+ "None": 2043,
2329
+ "Equi.": 2043,
2330
+ "Inv.": 2043,
2331
+ "Full": 2043
2332
+ },
2333
+ "80": {
2334
+ "None": 1779,
2335
+ "Equi.": 1779,
2336
+ "Inv.": 1779,
2337
+ "Full": 1779
2338
+ }
2339
+ },
2340
+ "expertReference": "Per-task expert_time_step in the released scenes; validated against commit a06ee3e986df72e7faa846d3eb2afe6d5a88b48a.",
2341
+ "protocolChanges": "Task-specific Mora V0 checkpoints chosen retrospectively on recorded evaluation success. All four channels use the same selected checkpoint per task. Accepted episode indices 1–50 per channel. Reach-and-stop, 0.5 s dwell, execute 20 policy action steps per query. The native perturbation profiles and anchor selections are retained in the episode logs. The exact simulator checkout has not been independently attested for every worker."
2342
+ },
2343
+ "coverage": {
2344
+ "evaluatedTaskCount": 26,
2345
+ "evaluatedEpisodeCount": 5200,
2346
+ "completedTaskCount": {
2347
+ "None": 26,
2348
+ "Equi.": 26,
2349
+ "Inv.": 26,
2350
+ "Full": 26,
2351
+ "mean": 26
2352
+ },
2353
+ "expectedFullEpisodeCount": 1300,
2354
+ "expectedFourChannelEpisodeCount": 5200,
2355
+ "missingTaskChannels": {}
2356
+ },
2357
+ "aggregation": {
2358
+ "units": {
2359
+ "sr": "percent",
2360
+ "lscr": "percent",
2361
+ "efficiency": "ratio",
2362
+ "steps": "physics_steps"
2363
+ },
2364
+ "sr": "100 * successes / episodes per task/channel; equal channel mean per task; equal task mean over all 26 settings.",
2365
+ "lscr": "100 * mean episode latched_stage_completion_rate, including failures; equal channel mean per task; equal task mean over all 26 settings.",
2366
+ "efficiency": "expert_time_step / mean successful steps_to_stable_success. Pool successful episodes across all four channels before calculating task mean efficiency. Overall values average task efficiencies over tasks with success; report valid-task counts.",
2367
+ "missing": "null means unreported (NR). Efficiency without successes is null with no-success status. Valid-task counts are null when unreported, and 0 for an evaluated scope without successes.",
2368
+ "scope": "Overall channel aggregates require all 26 settings. Four-channel means require all four channels. Never substitute Full-only scores for means or include duplicate reruns."
2369
+ },
2370
+ "logs": {
2371
+ "repository": "Bench2Dex/Leaderboard",
2372
+ "policyPath": "mora_v0",
2373
+ "revision": null,
2374
+ "status": "not-archived"
2375
+ },
2376
+ "selectedCheckpoints": {
2377
+ "03": {
2378
+ "checkpointSha256": "ec7d439e37462bc3c60c4cd62a344639b81e41dfce05ed24133dccbb089348a0",
2379
+ "recipe": "raw v3, 2k",
2380
+ "embodiment": "multi_xarm7_with_ability"
2381
+ },
2382
+ "06": {
2383
+ "checkpointSha256": "04465b5911a47461d8808eeda7e7e3524a7be53be98cafd53a8d45f03bc80a13",
2384
+ "recipe": "hand24-clip, original split, 2k",
2385
+ "embodiment": "multi_ur5_rh56dfx_with_flange"
2386
+ },
2387
+ "07": {
2388
+ "checkpointSha256": "0a15bbe47bcb04b0fda16fb74cb1422ee357d0b9f24d7d4add9289e3b75ea1c8",
2389
+ "recipe": "all 50 demonstrations, GB256, 2k",
2390
+ "embodiment": "multi_ur5_rh5dg2_with_flange"
2391
+ },
2392
+ "08": {
2393
+ "checkpointSha256": "071ef9e79da9a1d215ef55f77a198c98bcfddb6211485d348b6a572c284fc4c6",
2394
+ "recipe": "all 50 demonstrations, GB256, 2k",
2395
+ "embodiment": "multi_ur5_schunk_hand_with_flange"
2396
+ },
2397
+ "09": {
2398
+ "checkpointSha256": "12606d72d3f9086389e403d5e1d3b2a665ad853f6d924338c4b232e728601ff4",
2399
+ "recipe": "all 50 demonstrations, GB512, 2.5k",
2400
+ "embodiment": "multi_xarm7_with_leap"
2401
+ },
2402
+ "12": {
2403
+ "checkpointSha256": "a7e364578da1e21d7b434f5171e19b7880fefc9516c7ac655aba67b6bfe292be",
2404
+ "recipe": "hand24-clip, all 50, left MANO aligned to EgoDex, GB512, 2.5k",
2405
+ "embodiment": "multi_ur5_rh56dfx_with_flange"
2406
+ },
2407
+ "21": {
2408
+ "checkpointSha256": "b1d200c9f34faa066c362d87613ab05bb4ac333f1805ad780c17e03001f99e21",
2409
+ "recipe": "left stereo, raw, 2k",
2410
+ "embodiment": "multi_ur5_wuji_with_flange"
2411
+ },
2412
+ "22": {
2413
+ "checkpointSha256": "7ad5f3956e12b0b138bf6fd58068553d6596c968ebf3683f13e8b2a6c99018a5",
2414
+ "recipe": "all 50 demonstrations, GB512, 2.5k",
2415
+ "embodiment": "multi_panda_with_orca"
2416
+ },
2417
+ "24": {
2418
+ "checkpointSha256": "923879d5d0cf19b98d43ca40094637cb51be72d9b6f6344d7ece9e074f977113",
2419
+ "recipe": "raw, 2k",
2420
+ "embodiment": "multi_rm_65_with_revo2"
2421
+ },
2422
+ "26": {
2423
+ "checkpointSha256": "b7f8e9b7a64f9c30ccd60f6ce330f92661c00181af3a64d4eb1f248b43c0414c",
2424
+ "recipe": "all 50 demonstrations, GB512, 2.5k",
2425
+ "embodiment": "multi_iiwa7_with_sharpa"
2426
+ },
2427
+ "27": {
2428
+ "checkpointSha256": "0e0aafb7f40796576b692b20c62383eb49229df22662590319897acc5a2c48e9",
2429
+ "recipe": "left stereo, raw, 2k",
2430
+ "embodiment": "multi_ur5_wuji_with_flange"
2431
+ },
2432
+ "32": {
2433
+ "checkpointSha256": "a13ca410c29e9419c2c0ce2e9ccc9f1563a9ea9bdc71f2b2436aeadeeda24404",
2434
+ "recipe": "all 50 demonstrations, GB512, 2.5k",
2435
+ "embodiment": "multi_iiwa7_with_sharpa"
2436
+ },
2437
+ "34": {
2438
+ "checkpointSha256": "f4d6423dad867f8ff89225c48286b2af6f73c5f9d16644614f8ac9607d8a0fb9",
2439
+ "recipe": "full18 training labels, active13 execution, 2k",
2440
+ "embodiment": "multi_ur5_rh5dg2_with_flange"
2441
+ },
2442
+ "42": {
2443
+ "checkpointSha256": "f7fb5f6fc3fb410901a5c2505051ab66f47ae48ac248c5de075f76413e61e933",
2444
+ "recipe": "raw, original split, 2k",
2445
+ "embodiment": "multi_ur5_rh56dfx_with_flange"
2446
+ },
2447
+ "43": {
2448
+ "checkpointSha256": "6aa35c9de9abdc18efadd725b70beceae8e3677e80991730ba95cd15700eda99",
2449
+ "recipe": "all 50 demonstrations, GB256, 2k",
2450
+ "embodiment": "multi_ur5_shadow_hand_with_flange"
2451
+ },
2452
+ "44": {
2453
+ "checkpointSha256": "60636292342382254dd1f63bcf7fc6ce0fc70490da550e9019769d8649d2cf36",
2454
+ "recipe": "all 50 demonstrations, GB256, 2k",
2455
+ "embodiment": "multi_ur5_schunk_hand_with_flange"
2456
+ },
2457
+ "51": {
2458
+ "checkpointSha256": "930eadc35cc098ef5f034e988464dc2be9a06dc7f0bcc26c9372b827469e9e06",
2459
+ "recipe": "all 50 demonstrations, GB512, 2.5k",
2460
+ "embodiment": "multi_xarm7_with_leap"
2461
+ },
2462
+ "60": {
2463
+ "checkpointSha256": "648585ddfe8f95e84b1744c86e14fd131f313b263967ed21eb5e091c20d3f57c",
2464
+ "recipe": "all 50 demonstrations, GB256, 2k",
2465
+ "embodiment": "multi_ur5_rh5dg2_with_flange"
2466
+ },
2467
+ "61": {
2468
+ "checkpointSha256": "bb96fa88d33c16ed524ce2cbbd23ab2c6d5fac9b761f775119932ca34ff9ec54",
2469
+ "recipe": "all 50 demonstrations, GB512, 2.5k",
2470
+ "embodiment": "multi_panda_with_orca"
2471
+ },
2472
+ "62": {
2473
+ "checkpointSha256": "67bb9aa3551f9c6a709773e4e08981abfbd2d563230043af7c6bdc9020901004",
2474
+ "recipe": "all 50 demonstrations, GB512, 2.5k",
2475
+ "embodiment": "multi_xarm7_with_leap"
2476
+ },
2477
+ "64": {
2478
+ "checkpointSha256": "dcaf852fdbc3d9f73ca916f61ce83398446abde23affed5f3144da93b2451bfc",
2479
+ "recipe": "arm14-clip, left stereo, 2k",
2480
+ "embodiment": "multi_panda_with_allegro"
2481
+ },
2482
+ "67": {
2483
+ "checkpointSha256": "ff2e261104ed4e927dfb9f71ab57b11c93a8bd32694c1b22250043236e58db4a",
2484
+ "recipe": "arm12-clip, 2k",
2485
+ "embodiment": "multi_rm_65_with_revo2"
2486
+ },
2487
+ "73": {
2488
+ "checkpointSha256": "5525c14a4fae8c50651c1f2823ff19828e9da28d5b0fa6fb942f194373849338",
2489
+ "recipe": "all 50 demonstrations, GB512, 2.5k",
2490
+ "embodiment": "multi_iiwa7_with_sharpa"
2491
+ },
2492
+ "76": {
2493
+ "checkpointSha256": "adab38bae3cc29bb3572b0748a3eb5bbd7b00a68ca94a58b12aea5e62fb597b5",
2494
+ "recipe": "all 50 demonstrations, GB256, 2k",
2495
+ "embodiment": "multi_ur5_shadow_hand_with_flange"
2496
+ },
2497
+ "79": {
2498
+ "checkpointSha256": "3ca6b6de9ea781832fdba3020256a1effe72c0f86112f60ee360ea9499615f31",
2499
+ "recipe": "all 50 demonstrations, GB256, 2k",
2500
+ "embodiment": "multi_xarm7_with_ability"
2501
+ },
2502
+ "80": {
2503
+ "checkpointSha256": "f1166cc6864bd50a416920c77ed38818e7d70fdf958608bcc20f07012fbcd502",
2504
+ "recipe": "all 50 demonstrations, GB512, 2.5k",
2505
+ "embodiment": "multi_jaka_zu7_dexhand021_with_flange"
2506
+ }
2507
+ },
2508
+ "recoveryDisclosure": {
2509
+ "initialErrors": 7,
2510
+ "tasks": [
2511
+ "34",
2512
+ "60"
2513
+ ],
2514
+ "successfulRetries": 3,
2515
+ "failedRetries": 4,
2516
+ "originalErrorsPreserved": true,
2517
+ "indices": [
2518
+ {
2519
+ "task": "34",
2520
+ "channel": "Equi",
2521
+ "index": 8,
2522
+ "seed": 103400007,
2523
+ "first_error": "ValueError: Ambiguous angle outside (-pi,pi); preserve source and exclude affected windows",
2524
+ "final_success": false
2525
+ },
2526
+ {
2527
+ "task": "34",
2528
+ "channel": "Inv",
2529
+ "index": 35,
2530
+ "seed": 103400034,
2531
+ "first_error": "ValueError: Ambiguous angle outside (-pi,pi); preserve source and exclude affected windows",
2532
+ "final_success": true
2533
+ },
2534
+ {
2535
+ "task": "34",
2536
+ "channel": "None",
2537
+ "index": 2,
2538
+ "seed": 103400001,
2539
+ "first_error": "ValueError: Ambiguous angle outside (-pi,pi); preserve source and exclude affected windows",
2540
+ "final_success": false
2541
+ },
2542
+ {
2543
+ "task": "60",
2544
+ "channel": "Equi",
2545
+ "index": 15,
2546
+ "seed": 106000014,
2547
+ "first_error": "ValueError: Ambiguous angle outside (-pi,pi); preserve source and exclude affected windows",
2548
+ "final_success": true
2549
+ },
2550
+ {
2551
+ "task": "60",
2552
+ "channel": "Equi",
2553
+ "index": 25,
2554
+ "seed": 106000024,
2555
+ "first_error": "ValueError: Ambiguous angle outside (-pi,pi); preserve source and exclude affected windows",
2556
+ "final_success": false
2557
+ },
2558
+ {
2559
+ "task": "60",
2560
+ "channel": "Equi",
2561
+ "index": 47,
2562
+ "seed": 106000046,
2563
+ "first_error": "ValueError: Ambiguous angle outside (-pi,pi); preserve source and exclude affected windows",
2564
+ "final_success": false
2565
+ },
2566
+ {
2567
+ "task": "60",
2568
+ "channel": "Inv",
2569
+ "index": 22,
2570
+ "seed": 106000021,
2571
+ "first_error": "ValueError: Ambiguous angle outside (-pi,pi); preserve source and exclude affected windows",
2572
+ "final_success": true
2573
+ }
2574
+ ]
2575
+ }
2576
+ }
2577
+ }
mora_v0/per_task_eval/03_wine_glass_plate_balance/Equi/native_shards/000/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/03_wine_glass_plate_balance/Equi/native_shards/000/summary.json ADDED
@@ -0,0 +1,211 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2000step",
5
+ "base_seed": 100300000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 25,
9
+ "successes": 6,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 3246,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.24,
50
+ "success_rate_ci": {
51
+ "mean": 0.24,
52
+ "ci_low": 0.08,
53
+ "ci_high": 0.4
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.7466666666666667
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 36.79444444444364
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.24,
64
+ "hard_violation_rate": 0.6,
65
+ "drop_rate": 0.6,
66
+ "high_speed_violation_rate": 0.0,
67
+ "safety_violation_step_rate": 0.0003069941270688735
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.24,
74
+ "at_end_success_rate": 0.24,
75
+ "task_family_success_rate": {
76
+ "table_setting": 0.24
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.64,
81
+ "current_chain_depth_progress_score": 0.64,
82
+ "latched_chain_depth_progress_score": 0.7466666666666667,
83
+ "avg_current_chain_depth": 1.92,
84
+ "avg_latched_chain_depth": 2.24
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.9808901429187638,
88
+ "expert_time_s": 36.06666666666666,
89
+ "expert_time_step": 2164,
90
+ "avg_steps_to_success": 2207.6666666666665,
91
+ "avg_policy_steps_to_success": 736.3333333333334,
92
+ "avg_policy_queries_to_success": 37.166666666666664
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.0003069941270688735,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_194_wineglass_1": 9,
98
+ "drop_to_table_obj_194_wineglass_2": 7,
99
+ "drop_to_table_obj_194_wineglass_3": 7
100
+ },
101
+ "robot_constraint_violation_rate": 0.36,
102
+ "finger_joint_limit_saturation_rate": 0.08,
103
+ "joint_limit_active_step_rate": 0.0017084890549919915,
104
+ "finger_joint_limit_saturation_step_rate": 0.00022690870261612387,
105
+ "max_joint_limit_excess_rad": 0.13941732048988342,
106
+ "robot_constraint_diagnostics": {
107
+ "joint_limit_soft_counts": {
108
+ "middle_q2": 21,
109
+ "multi_ring_q2": 105,
110
+ "pinky_q1": 2
111
+ },
112
+ "joint_limit_hard_counts": {
113
+ "middle_q2": 1,
114
+ "multi_pinky_q1": 4,
115
+ "multi_ring_q1": 1,
116
+ "multi_ring_q2": 5,
117
+ "pinky_q1": 15,
118
+ "ring_q1": 2,
119
+ "ring_q2": 2,
120
+ "thumb_q2": 6
121
+ },
122
+ "joint_limit_hard_episode_events": 31,
123
+ "finger_joint_limit_saturation_counts": {
124
+ "multi_ring_q2": 17
125
+ },
126
+ "joint_limit_max_excess_by_joint": {
127
+ "index_q1": 0.004180565942078829,
128
+ "index_q2": 1.0939534611509316e-08,
129
+ "middle_q1": 0.0019478725735098124,
130
+ "middle_q2": 0.1328805834054947,
131
+ "multi_index_q1": 0.00401646364480257,
132
+ "multi_index_q2": 9.207280271539275e-09,
133
+ "multi_middle_q1": 0.0015486484626308084,
134
+ "multi_middle_q2": 0.002593185054138303,
135
+ "multi_pinky_q1": 0.007836827076971531,
136
+ "multi_pinky_q2": 0.0021768026053905487,
137
+ "multi_ring_q1": 0.005249276291579008,
138
+ "multi_ring_q2": 0.13941732048988342,
139
+ "pinky_q1": 0.017454849556088448,
140
+ "pinky_q2": 0.004159054718911648,
141
+ "ring_q1": 0.005172125529497862,
142
+ "ring_q2": 0.00843200646340847,
143
+ "thumb_q2": 0.006736241281032562
144
+ },
145
+ "coverage": {
146
+ "hard_limit_source": {
147
+ "available": 25,
148
+ "unavailable": 0,
149
+ "unknown": 0
150
+ },
151
+ "joint_role_manifest": {
152
+ "available": 25,
153
+ "unavailable": 0,
154
+ "unknown": 0
155
+ }
156
+ },
157
+ "hard_limit_sources": {
158
+ "runtime.joint_pos_limits": 25
159
+ },
160
+ "soft_limit_sources": {
161
+ "runtime.soft_joint_pos_limits": 25
162
+ },
163
+ "unknown_runtime_joint_names": [],
164
+ "baseline_checks": {
165
+ "post_home": {
166
+ "checked": 25,
167
+ "hard_violation_episodes": 0
168
+ },
169
+ "post_reset": {
170
+ "checked": 25,
171
+ "hard_violation_episodes": 0
172
+ }
173
+ }
174
+ }
175
+ },
176
+ "robot_motion": {
177
+ "joint_acceleration_rms": 11.456243853848612,
178
+ "joint_effort_rms": 102.24249267396151,
179
+ "joint_jerk_rms": 1135.0961523421936,
180
+ "joint_velocity_rms": 0.8151439393878812
181
+ },
182
+ "grasp": {
183
+ "kinematic_grasp_stability_index": 0.9244543996511804,
184
+ "mean_kinematic_grasp_stability": 0.6282927060920159,
185
+ "per_object_best_kinematic_grasp_stability": {
186
+ "obj_194_wineglass_1": 0.7999688330339871,
187
+ "obj_194_wineglass_2": 0.67574321738512,
188
+ "obj_194_wineglass_3": 0.8145494098944459
189
+ },
190
+ "grasp_gsi_diagnostics": {
191
+ "mean_anti_slip_score": 0.8829518583977488,
192
+ "mean_duration_score": 0.5489593799710433,
193
+ "mean_hold_score": 0.23280988246040624,
194
+ "mean_kinematic_stability_score": 0.1595432324238229,
195
+ "mean_legacy_gsi": 0.42070864973458966,
196
+ "mean_lift_score": 0.2598179514068027,
197
+ "mean_motion_stability_score": 0.2125134812565939
198
+ }
199
+ },
200
+ "tool": {
201
+ "tool_selection_accuracy": null,
202
+ "tool_switch_success_rate": null,
203
+ "tool_switch_total": 0,
204
+ "tool_switch_successful": 0
205
+ },
206
+ "runtime": {
207
+ "avg_steps": 2996.8,
208
+ "error_count": 0
209
+ }
210
+ }
211
+ }
mora_v0/per_task_eval/03_wine_glass_plate_balance/Equi/native_shards/001/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/03_wine_glass_plate_balance/Equi/native_shards/001/summary.json ADDED
@@ -0,0 +1,218 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2000step",
5
+ "base_seed": 100300000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 25,
9
+ "successes": 7,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 3246,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.28,
50
+ "success_rate_ci": {
51
+ "mean": 0.28,
52
+ "ci_low": 0.12,
53
+ "ci_high": 0.48
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.7466666666666667
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 36.88095238095157
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.12,
64
+ "hard_violation_rate": 0.76,
65
+ "drop_rate": 0.76,
66
+ "high_speed_violation_rate": 0.08,
67
+ "safety_violation_step_rate": 0.00047349765956871125
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.28,
74
+ "at_end_success_rate": 0.28,
75
+ "task_family_success_rate": {
76
+ "table_setting": 0.28
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.6666666666666666,
81
+ "current_chain_depth_progress_score": 0.6666666666666666,
82
+ "latched_chain_depth_progress_score": 0.7466666666666667,
83
+ "avg_current_chain_depth": 2,
84
+ "avg_latched_chain_depth": 2.24
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.9801296076862106,
88
+ "expert_time_s": 36.06666666666666,
89
+ "expert_time_step": 2164,
90
+ "avg_steps_to_success": 2212.8571428571427,
91
+ "avg_policy_steps_to_success": 737.8571428571429,
92
+ "avg_policy_queries_to_success": 37.285714285714285
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.0004464406504504992,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_194_wineglass_1": 8,
98
+ "drop_to_table_obj_194_wineglass_2": 9,
99
+ "drop_to_table_obj_194_wineglass_3": 13,
100
+ "high_speed_obj_194_wineglass_1": 2,
101
+ "high_speed_obj_194_wineglass_2": 1
102
+ },
103
+ "robot_constraint_violation_rate": 0.44,
104
+ "finger_joint_limit_saturation_rate": 0.04,
105
+ "joint_limit_active_step_rate": 0.0013934359695879218,
106
+ "finger_joint_limit_saturation_step_rate": 0.00013528504559106038,
107
+ "max_joint_limit_excess_rad": 0.18246252834796906,
108
+ "robot_constraint_diagnostics": {
109
+ "joint_limit_soft_counts": {
110
+ "middle_q1": 18,
111
+ "multi_pinky_q1": 36,
112
+ "multi_ring_q1": 17,
113
+ "pinky_q1": 11,
114
+ "pinky_q2": 42,
115
+ "ring_q2": 54
116
+ },
117
+ "joint_limit_hard_counts": {
118
+ "middle_q1": 5,
119
+ "multi_pinky_q1": 13,
120
+ "multi_ring_q1": 7,
121
+ "multi_ring_q2": 6,
122
+ "pinky_q1": 20,
123
+ "pinky_q2": 5,
124
+ "ring_q1": 2,
125
+ "ring_q2": 11,
126
+ "thumb_q2": 2
127
+ },
128
+ "joint_limit_hard_episode_events": 49,
129
+ "finger_joint_limit_saturation_counts": {
130
+ "pinky_q2": 8,
131
+ "ring_q2": 10
132
+ },
133
+ "joint_limit_max_excess_by_joint": {
134
+ "index_q1": 0.004260599613189697,
135
+ "index_q2": 1.1244551068045894e-08,
136
+ "middle_q1": 0.044337060302495956,
137
+ "middle_q2": 0.001714093261398375,
138
+ "multi_index_q1": 0.003545554354786873,
139
+ "multi_index_q2": 1.2713139874165336e-08,
140
+ "multi_middle_q1": 0.0016508514527231455,
141
+ "multi_middle_q2": 0.0033921687863767147,
142
+ "multi_pinky_q1": 0.09601803123950958,
143
+ "multi_pinky_q2": 0.0005938089452683926,
144
+ "multi_ring_q1": 0.06733377277851105,
145
+ "multi_ring_q2": 0.006712598260492086,
146
+ "pinky_q1": 0.031240170821547508,
147
+ "pinky_q2": 0.18246252834796906,
148
+ "ring_q1": 0.007420097012072802,
149
+ "ring_q2": 0.1773361712694168,
150
+ "thumb_q2": 0.005348311737179756
151
+ },
152
+ "coverage": {
153
+ "hard_limit_source": {
154
+ "available": 25,
155
+ "unavailable": 0,
156
+ "unknown": 0
157
+ },
158
+ "joint_role_manifest": {
159
+ "available": 25,
160
+ "unavailable": 0,
161
+ "unknown": 0
162
+ }
163
+ },
164
+ "hard_limit_sources": {
165
+ "runtime.joint_pos_limits": 25
166
+ },
167
+ "soft_limit_sources": {
168
+ "runtime.soft_joint_pos_limits": 25
169
+ },
170
+ "unknown_runtime_joint_names": [],
171
+ "baseline_checks": {
172
+ "post_home": {
173
+ "checked": 25,
174
+ "hard_violation_episodes": 0
175
+ },
176
+ "post_reset": {
177
+ "checked": 25,
178
+ "hard_violation_episodes": 0
179
+ }
180
+ }
181
+ }
182
+ },
183
+ "robot_motion": {
184
+ "joint_acceleration_rms": 11.33750392229561,
185
+ "joint_effort_rms": 103.88307482831449,
186
+ "joint_jerk_rms": 1118.4515983565038,
187
+ "joint_velocity_rms": 0.8572011226245809
188
+ },
189
+ "grasp": {
190
+ "kinematic_grasp_stability_index": 0.9223325063637366,
191
+ "mean_kinematic_grasp_stability": 0.6219607718276341,
192
+ "per_object_best_kinematic_grasp_stability": {
193
+ "obj_194_wineglass_1": 0.8353911109649078,
194
+ "obj_194_wineglass_2": 0.8139377714980759,
195
+ "obj_194_wineglass_3": 0.6900333628574978
196
+ },
197
+ "grasp_gsi_diagnostics": {
198
+ "mean_anti_slip_score": 0.8728989725545493,
199
+ "mean_duration_score": 0.5823526811455243,
200
+ "mean_hold_score": 0.2837506618860159,
201
+ "mean_kinematic_stability_score": 0.17239131587061304,
202
+ "mean_legacy_gsi": 0.45724544567089087,
203
+ "mean_lift_score": 0.32383227305184425,
204
+ "mean_motion_stability_score": 0.26464678072443165
205
+ }
206
+ },
207
+ "tool": {
208
+ "tool_selection_accuracy": null,
209
+ "tool_switch_success_rate": null,
210
+ "tool_switch_total": 0,
211
+ "tool_switch_successful": 0
212
+ },
213
+ "runtime": {
214
+ "avg_steps": 2956.72,
215
+ "error_count": 0
216
+ }
217
+ }
218
+ }
mora_v0/per_task_eval/03_wine_glass_plate_balance/Equi/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/03_wine_glass_plate_balance/Equi/summary.json ADDED
@@ -0,0 +1,222 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2000step",
5
+ "base_seed": 100300000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 50,
9
+ "successes": 13,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 3246,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.26,
50
+ "success_rate_ci": {
51
+ "mean": 0.26,
52
+ "ci_low": 0.14,
53
+ "ci_high": 0.38
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.7466666666666667
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 36.841025641024835
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.18,
64
+ "hard_violation_rate": 0.68,
65
+ "drop_rate": 0.68,
66
+ "high_speed_violation_rate": 0.04,
67
+ "safety_violation_step_rate": 0.0003896854297961542
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.26,
74
+ "at_end_success_rate": 0.26,
75
+ "task_family_success_rate": {
76
+ "table_setting": 0.26
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.6533333333333333,
81
+ "current_chain_depth_progress_score": 0.6533333333333333,
82
+ "latched_chain_depth_progress_score": 0.7466666666666667,
83
+ "avg_current_chain_depth": 1.96,
84
+ "avg_latched_chain_depth": 2.24
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.980480623947389,
88
+ "expert_time_s": 36.06666666666666,
89
+ "expert_time_step": 2164,
90
+ "avg_steps_to_success": 2210.4615384615386,
91
+ "avg_policy_steps_to_success": 737.1538461538462,
92
+ "avg_policy_queries_to_success": 37.23076923076923
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.0003762480011824937,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_194_wineglass_1": 17,
98
+ "drop_to_table_obj_194_wineglass_2": 16,
99
+ "drop_to_table_obj_194_wineglass_3": 20,
100
+ "high_speed_obj_194_wineglass_1": 2,
101
+ "high_speed_obj_194_wineglass_2": 1
102
+ },
103
+ "robot_constraint_violation_rate": 0.4,
104
+ "finger_joint_limit_saturation_rate": 0.06,
105
+ "joint_limit_active_step_rate": 0.0015520230048777866,
106
+ "finger_joint_limit_saturation_step_rate": 0.00018140528628441663,
107
+ "max_joint_limit_excess_rad": 0.18246252834796906,
108
+ "robot_constraint_diagnostics": {
109
+ "joint_limit_soft_counts": {
110
+ "middle_q1": 18,
111
+ "middle_q2": 21,
112
+ "multi_pinky_q1": 36,
113
+ "multi_ring_q1": 17,
114
+ "multi_ring_q2": 105,
115
+ "pinky_q1": 13,
116
+ "pinky_q2": 42,
117
+ "ring_q2": 54
118
+ },
119
+ "joint_limit_hard_counts": {
120
+ "middle_q1": 5,
121
+ "middle_q2": 1,
122
+ "multi_pinky_q1": 17,
123
+ "multi_ring_q1": 8,
124
+ "multi_ring_q2": 11,
125
+ "pinky_q1": 35,
126
+ "pinky_q2": 5,
127
+ "ring_q1": 4,
128
+ "ring_q2": 13,
129
+ "thumb_q2": 8
130
+ },
131
+ "joint_limit_hard_episode_events": 80,
132
+ "finger_joint_limit_saturation_counts": {
133
+ "multi_ring_q2": 17,
134
+ "pinky_q2": 8,
135
+ "ring_q2": 10
136
+ },
137
+ "joint_limit_max_excess_by_joint": {
138
+ "index_q1": 0.004260599613189697,
139
+ "index_q2": 1.1244551068045894e-08,
140
+ "middle_q1": 0.044337060302495956,
141
+ "middle_q2": 0.1328805834054947,
142
+ "multi_index_q1": 0.00401646364480257,
143
+ "multi_index_q2": 1.2713139874165336e-08,
144
+ "multi_middle_q1": 0.0016508514527231455,
145
+ "multi_middle_q2": 0.0033921687863767147,
146
+ "multi_pinky_q1": 0.09601803123950958,
147
+ "multi_pinky_q2": 0.0021768026053905487,
148
+ "multi_ring_q1": 0.06733377277851105,
149
+ "multi_ring_q2": 0.13941732048988342,
150
+ "pinky_q1": 0.031240170821547508,
151
+ "pinky_q2": 0.18246252834796906,
152
+ "ring_q1": 0.007420097012072802,
153
+ "ring_q2": 0.1773361712694168,
154
+ "thumb_q2": 0.006736241281032562
155
+ },
156
+ "coverage": {
157
+ "hard_limit_source": {
158
+ "available": 50,
159
+ "unavailable": 0,
160
+ "unknown": 0
161
+ },
162
+ "joint_role_manifest": {
163
+ "available": 50,
164
+ "unavailable": 0,
165
+ "unknown": 0
166
+ }
167
+ },
168
+ "hard_limit_sources": {
169
+ "runtime.joint_pos_limits": 50
170
+ },
171
+ "soft_limit_sources": {
172
+ "runtime.soft_joint_pos_limits": 50
173
+ },
174
+ "unknown_runtime_joint_names": [],
175
+ "baseline_checks": {
176
+ "post_home": {
177
+ "checked": 50,
178
+ "hard_violation_episodes": 0
179
+ },
180
+ "post_reset": {
181
+ "checked": 50,
182
+ "hard_violation_episodes": 0
183
+ }
184
+ }
185
+ }
186
+ },
187
+ "robot_motion": {
188
+ "joint_acceleration_rms": 11.396873888072111,
189
+ "joint_effort_rms": 103.062783751138,
190
+ "joint_jerk_rms": 1126.7738753493488,
191
+ "joint_velocity_rms": 0.836172531006231
192
+ },
193
+ "grasp": {
194
+ "kinematic_grasp_stability_index": 0.9233934530074585,
195
+ "mean_kinematic_grasp_stability": 0.6251267389598251,
196
+ "per_object_best_kinematic_grasp_stability": {
197
+ "obj_194_wineglass_1": 0.8176799719994474,
198
+ "obj_194_wineglass_2": 0.744840494441598,
199
+ "obj_194_wineglass_3": 0.7522913863759719
200
+ },
201
+ "grasp_gsi_diagnostics": {
202
+ "mean_anti_slip_score": 0.877925415476149,
203
+ "mean_duration_score": 0.5656560305582838,
204
+ "mean_hold_score": 0.25828027217321103,
205
+ "mean_kinematic_stability_score": 0.16596727414721796,
206
+ "mean_legacy_gsi": 0.43897704770274026,
207
+ "mean_lift_score": 0.29182511222932345,
208
+ "mean_motion_stability_score": 0.23858013099051278
209
+ }
210
+ },
211
+ "tool": {
212
+ "tool_selection_accuracy": null,
213
+ "tool_switch_success_rate": null,
214
+ "tool_switch_total": 0,
215
+ "tool_switch_successful": 0
216
+ },
217
+ "runtime": {
218
+ "avg_steps": 2976.76,
219
+ "error_count": 0
220
+ }
221
+ }
222
+ }
mora_v0/per_task_eval/03_wine_glass_plate_balance/Full/native_shards/000/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/03_wine_glass_plate_balance/Full/native_shards/000/summary.json ADDED
@@ -0,0 +1,216 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2000step",
5
+ "base_seed": 200300000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 25,
9
+ "successes": 6,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 3246,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.24,
50
+ "success_rate_ci": {
51
+ "mean": 0.24,
52
+ "ci_low": 0.08,
53
+ "ci_high": 0.4
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.8133333333333334
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 35.15555555555484
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.16,
64
+ "hard_violation_rate": 0.76,
65
+ "drop_rate": 0.76,
66
+ "high_speed_violation_rate": 0.04,
67
+ "safety_violation_step_rate": 0.02320731871384367
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.24,
74
+ "at_end_success_rate": 0.24,
75
+ "task_family_success_rate": {
76
+ "table_setting": 0.24
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.6933333333333334,
81
+ "current_chain_depth_progress_score": 0.6933333333333334,
82
+ "latched_chain_depth_progress_score": 0.8133333333333334,
83
+ "avg_current_chain_depth": 2.08,
84
+ "avg_latched_chain_depth": 2.44
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 1.0282869702133737,
88
+ "expert_time_s": 36.06666666666666,
89
+ "expert_time_step": 2164,
90
+ "avg_steps_to_success": 2109.3333333333335,
91
+ "avg_policy_steps_to_success": 703.3333333333334,
92
+ "avg_policy_queries_to_success": 35.666666666666664
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.00044396609713440063,
96
+ "violation_counts": {
97
+ "drop_obj_194_wineglass_2": 1,
98
+ "drop_to_table_obj_194_wineglass_1": 11,
99
+ "drop_to_table_obj_194_wineglass_2": 8,
100
+ "drop_to_table_obj_194_wineglass_3": 12,
101
+ "high_speed_obj_194_wineglass_2": 1
102
+ },
103
+ "robot_constraint_violation_rate": 0.28,
104
+ "finger_joint_limit_saturation_rate": 0.08,
105
+ "joint_limit_active_step_rate": 0.0022736445580519306,
106
+ "finger_joint_limit_saturation_step_rate": 0.0009686533028386924,
107
+ "max_joint_limit_excess_rad": 0.2497359812259674,
108
+ "robot_constraint_diagnostics": {
109
+ "joint_limit_soft_counts": {
110
+ "index_q1": 1,
111
+ "multi_ring_q1": 46,
112
+ "multi_ring_q2": 99,
113
+ "pinky_q1": 8,
114
+ "ring_q1": 15,
115
+ "ring_q2": 2
116
+ },
117
+ "joint_limit_hard_counts": {
118
+ "index_q1": 2,
119
+ "middle_q1": 1,
120
+ "multi_ring_q1": 2,
121
+ "multi_ring_q2": 2,
122
+ "pinky_q1": 4,
123
+ "ring_q1": 5,
124
+ "ring_q2": 4
125
+ },
126
+ "joint_limit_hard_episode_events": 16,
127
+ "finger_joint_limit_saturation_counts": {
128
+ "multi_ring_q1": 21,
129
+ "multi_ring_q2": 51
130
+ },
131
+ "joint_limit_max_excess_by_joint": {
132
+ "index_q1": 0.013271281495690346,
133
+ "index_q2": 1.1998781523914204e-08,
134
+ "middle_q1": 0.007092274259775877,
135
+ "middle_q2": 0.0020781345665454865,
136
+ "multi_index_q1": 0.00142211327329278,
137
+ "multi_index_q2": 1.0218990098564973e-08,
138
+ "multi_middle_q1": 0.0015896966215223074,
139
+ "multi_middle_q2": 0.0005197598948143423,
140
+ "multi_pinky_q1": 0.002687940141186118,
141
+ "multi_pinky_q2": 0.0012490669032558799,
142
+ "multi_ring_q1": 0.21524950861930847,
143
+ "multi_ring_q2": 0.2497359812259674,
144
+ "pinky_q1": 0.054866448044776917,
145
+ "pinky_q2": 0.004172260407358408,
146
+ "ring_q1": 0.029724949970841408,
147
+ "ring_q2": 0.047779981046915054,
148
+ "thumb_q2": 0.0024876203387975693
149
+ },
150
+ "coverage": {
151
+ "hard_limit_source": {
152
+ "available": 25,
153
+ "unavailable": 0,
154
+ "unknown": 0
155
+ },
156
+ "joint_role_manifest": {
157
+ "available": 25,
158
+ "unavailable": 0,
159
+ "unknown": 0
160
+ }
161
+ },
162
+ "hard_limit_sources": {
163
+ "runtime.joint_pos_limits": 25
164
+ },
165
+ "soft_limit_sources": {
166
+ "runtime.soft_joint_pos_limits": 25
167
+ },
168
+ "unknown_runtime_joint_names": [],
169
+ "baseline_checks": {
170
+ "post_home": {
171
+ "checked": 25,
172
+ "hard_violation_episodes": 0
173
+ },
174
+ "post_reset": {
175
+ "checked": 25,
176
+ "hard_violation_episodes": 0
177
+ }
178
+ }
179
+ }
180
+ },
181
+ "robot_motion": {
182
+ "joint_acceleration_rms": 10.82781852000208,
183
+ "joint_effort_rms": 103.44365129607935,
184
+ "joint_jerk_rms": 1065.6110760870683,
185
+ "joint_velocity_rms": 0.8253091377828877
186
+ },
187
+ "grasp": {
188
+ "kinematic_grasp_stability_index": 0.943439022558679,
189
+ "mean_kinematic_grasp_stability": 0.6455984081053279,
190
+ "per_object_best_kinematic_grasp_stability": {
191
+ "obj_194_wineglass_1": 0.8137191000824423,
192
+ "obj_194_wineglass_2": 0.897560613235321,
193
+ "obj_194_wineglass_3": 0.7330838308815779
194
+ },
195
+ "grasp_gsi_diagnostics": {
196
+ "mean_anti_slip_score": 0.8771214792507683,
197
+ "mean_duration_score": 0.5846970090165837,
198
+ "mean_hold_score": 0.28906614484818327,
199
+ "mean_kinematic_stability_score": 0.21411216110558026,
200
+ "mean_legacy_gsi": 0.46021592480854395,
201
+ "mean_lift_score": 0.32396642813677884,
202
+ "mean_motion_stability_score": 0.2668202478653398
203
+ }
204
+ },
205
+ "tool": {
206
+ "tool_selection_accuracy": null,
207
+ "tool_switch_success_rate": null,
208
+ "tool_switch_total": 0,
209
+ "tool_switch_successful": 0
210
+ },
211
+ "runtime": {
212
+ "avg_steps": 2973.2,
213
+ "error_count": 0
214
+ }
215
+ }
216
+ }
mora_v0/per_task_eval/03_wine_glass_plate_balance/Full/native_shards/001/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/03_wine_glass_plate_balance/Full/native_shards/001/summary.json ADDED
@@ -0,0 +1,219 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2000step",
5
+ "base_seed": 200300000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 25,
9
+ "successes": 6,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 3246,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.24,
50
+ "success_rate_ci": {
51
+ "mean": 0.24,
52
+ "ci_low": 0.08,
53
+ "ci_high": 0.4
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.7999999999999999
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 36.22499999999923
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.16,
64
+ "hard_violation_rate": 0.76,
65
+ "drop_rate": 0.76,
66
+ "high_speed_violation_rate": 0.04,
67
+ "safety_violation_step_rate": 0.0136251087465703
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.24,
74
+ "at_end_success_rate": 0.24,
75
+ "task_family_success_rate": {
76
+ "table_setting": 0.24
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.6799999999999999,
81
+ "current_chain_depth_progress_score": 0.6799999999999999,
82
+ "latched_chain_depth_progress_score": 0.7999999999999999,
83
+ "avg_current_chain_depth": 2.04,
84
+ "avg_latched_chain_depth": 2.4
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.9973083647861969,
88
+ "expert_time_s": 36.06666666666666,
89
+ "expert_time_step": 2164,
90
+ "avg_steps_to_success": 2173.5,
91
+ "avg_policy_steps_to_success": 725,
92
+ "avg_policy_queries_to_success": 36.833333333333336
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.0004149099913002744,
96
+ "violation_counts": {
97
+ "drop_obj_194_wineglass_3": 1,
98
+ "drop_to_table_obj_194_wineglass_1": 9,
99
+ "drop_to_table_obj_194_wineglass_2": 9,
100
+ "drop_to_table_obj_194_wineglass_3": 11,
101
+ "high_speed_obj_194_wineglass_3": 1
102
+ },
103
+ "robot_constraint_violation_rate": 0.48,
104
+ "finger_joint_limit_saturation_rate": 0.04,
105
+ "joint_limit_active_step_rate": 0.025095362377032727,
106
+ "finger_joint_limit_saturation_step_rate": 0.022364986950411565,
107
+ "max_joint_limit_excess_rad": 0.43434858322143555,
108
+ "robot_constraint_diagnostics": {
109
+ "joint_limit_soft_counts": {
110
+ "middle_q2": 27,
111
+ "multi_middle_q1": 1806,
112
+ "multi_middle_q2": 4,
113
+ "multi_ring_q2": 20,
114
+ "pinky_q1": 3,
115
+ "ring_q1": 15,
116
+ "ring_q2": 1
117
+ },
118
+ "joint_limit_hard_counts": {
119
+ "middle_q2": 2,
120
+ "multi_middle_q1": 4,
121
+ "multi_middle_q2": 2,
122
+ "multi_pinky_q2": 4,
123
+ "multi_ring_q1": 52,
124
+ "multi_ring_q2": 8,
125
+ "pinky_q1": 9,
126
+ "pinky_q2": 1,
127
+ "ring_q1": 5,
128
+ "ring_q2": 6
129
+ },
130
+ "joint_limit_hard_episode_events": 37,
131
+ "finger_joint_limit_saturation_counts": {
132
+ "multi_middle_q1": 1671
133
+ },
134
+ "joint_limit_max_excess_by_joint": {
135
+ "index_q1": 0.002813324797898531,
136
+ "index_q2": 8.890562064323149e-09,
137
+ "middle_q1": 0.0016501120990142226,
138
+ "middle_q2": 0.13230398297309875,
139
+ "multi_index_q1": 0.0035281586460769176,
140
+ "multi_index_q2": 1.1711183134366365e-08,
141
+ "multi_middle_q1": 0.43434858322143555,
142
+ "multi_middle_q2": 0.01762932538986206,
143
+ "multi_pinky_q1": 0.004636600613594055,
144
+ "multi_pinky_q2": 0.0073623149655759335,
145
+ "multi_ring_q1": 0.00907111819833517,
146
+ "multi_ring_q2": 0.05921969190239906,
147
+ "pinky_q1": 0.020613856613636017,
148
+ "pinky_q2": 0.006702502723783255,
149
+ "ring_q1": 0.0579492449760437,
150
+ "ring_q2": 0.012042106129229069,
151
+ "thumb_q2": 0.004173769615590572
152
+ },
153
+ "coverage": {
154
+ "hard_limit_source": {
155
+ "available": 25,
156
+ "unavailable": 0,
157
+ "unknown": 0
158
+ },
159
+ "joint_role_manifest": {
160
+ "available": 25,
161
+ "unavailable": 0,
162
+ "unknown": 0
163
+ }
164
+ },
165
+ "hard_limit_sources": {
166
+ "runtime.joint_pos_limits": 25
167
+ },
168
+ "soft_limit_sources": {
169
+ "runtime.soft_joint_pos_limits": 25
170
+ },
171
+ "unknown_runtime_joint_names": [],
172
+ "baseline_checks": {
173
+ "post_home": {
174
+ "checked": 25,
175
+ "hard_violation_episodes": 0
176
+ },
177
+ "post_reset": {
178
+ "checked": 25,
179
+ "hard_violation_episodes": 0
180
+ }
181
+ }
182
+ }
183
+ },
184
+ "robot_motion": {
185
+ "joint_acceleration_rms": 10.732362104236469,
186
+ "joint_effort_rms": 103.04903259370309,
187
+ "joint_jerk_rms": 1051.705383202551,
188
+ "joint_velocity_rms": 0.8168488635857162
189
+ },
190
+ "grasp": {
191
+ "kinematic_grasp_stability_index": 0.9383987186853804,
192
+ "mean_kinematic_grasp_stability": 0.6425980960661591,
193
+ "per_object_best_kinematic_grasp_stability": {
194
+ "obj_194_wineglass_1": 0.8627625439492401,
195
+ "obj_194_wineglass_2": 0.8858216955568163,
196
+ "obj_194_wineglass_3": 0.6752287511421491
197
+ },
198
+ "grasp_gsi_diagnostics": {
199
+ "mean_anti_slip_score": 0.883254675175651,
200
+ "mean_duration_score": 0.5868394704545722,
201
+ "mean_hold_score": 0.322934014983721,
202
+ "mean_kinematic_stability_score": 0.22669959628451342,
203
+ "mean_legacy_gsi": 0.4817140643689989,
204
+ "mean_lift_score": 0.3660836358362872,
205
+ "mean_motion_stability_score": 0.29997671841636253
206
+ }
207
+ },
208
+ "tool": {
209
+ "tool_selection_accuracy": null,
210
+ "tool_switch_success_rate": null,
211
+ "tool_switch_total": 0,
212
+ "tool_switch_successful": 0
213
+ },
214
+ "runtime": {
215
+ "avg_steps": 2988.6,
216
+ "error_count": 0
217
+ }
218
+ }
219
+ }
mora_v0/per_task_eval/03_wine_glass_plate_balance/Full/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/03_wine_glass_plate_balance/Full/summary.json ADDED
@@ -0,0 +1,227 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2000step",
5
+ "base_seed": 200300000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 50,
9
+ "successes": 12,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 3246,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.24,
50
+ "success_rate_ci": {
51
+ "mean": 0.24,
52
+ "ci_low": 0.12,
53
+ "ci_high": 0.36
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.8066666666666666
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 35.690277777777034
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.16,
64
+ "hard_violation_rate": 0.76,
65
+ "drop_rate": 0.76,
66
+ "high_speed_violation_rate": 0.04,
67
+ "safety_violation_step_rate": 0.018403837767117313
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.24,
74
+ "at_end_success_rate": 0.24,
75
+ "task_family_success_rate": {
76
+ "table_setting": 0.24
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.6866666666666666,
81
+ "current_chain_depth_progress_score": 0.6866666666666666,
82
+ "latched_chain_depth_progress_score": 0.8066666666666666,
83
+ "avg_current_chain_depth": 2.06,
84
+ "avg_latched_chain_depth": 2.42
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 1.0127976674997852,
88
+ "expert_time_s": 36.06666666666666,
89
+ "expert_time_step": 2164,
90
+ "avg_steps_to_success": 2141.4166666666665,
91
+ "avg_policy_steps_to_success": 714.1666666666666,
92
+ "avg_policy_queries_to_success": 36.25
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.00042940051662249655,
96
+ "violation_counts": {
97
+ "drop_obj_194_wineglass_2": 1,
98
+ "drop_obj_194_wineglass_3": 1,
99
+ "drop_to_table_obj_194_wineglass_1": 20,
100
+ "drop_to_table_obj_194_wineglass_2": 17,
101
+ "drop_to_table_obj_194_wineglass_3": 23,
102
+ "high_speed_obj_194_wineglass_2": 1,
103
+ "high_speed_obj_194_wineglass_3": 1
104
+ },
105
+ "robot_constraint_violation_rate": 0.38,
106
+ "finger_joint_limit_saturation_rate": 0.06,
107
+ "joint_limit_active_step_rate": 0.013713978999630986,
108
+ "finger_joint_limit_saturation_step_rate": 0.011694454694890805,
109
+ "max_joint_limit_excess_rad": 0.43434858322143555,
110
+ "robot_constraint_diagnostics": {
111
+ "joint_limit_soft_counts": {
112
+ "index_q1": 1,
113
+ "middle_q2": 27,
114
+ "multi_middle_q1": 1806,
115
+ "multi_middle_q2": 4,
116
+ "multi_ring_q1": 46,
117
+ "multi_ring_q2": 119,
118
+ "pinky_q1": 11,
119
+ "ring_q1": 30,
120
+ "ring_q2": 3
121
+ },
122
+ "joint_limit_hard_counts": {
123
+ "index_q1": 2,
124
+ "middle_q1": 1,
125
+ "middle_q2": 2,
126
+ "multi_middle_q1": 4,
127
+ "multi_middle_q2": 2,
128
+ "multi_pinky_q2": 4,
129
+ "multi_ring_q1": 54,
130
+ "multi_ring_q2": 10,
131
+ "pinky_q1": 13,
132
+ "pinky_q2": 1,
133
+ "ring_q1": 10,
134
+ "ring_q2": 10
135
+ },
136
+ "joint_limit_hard_episode_events": 53,
137
+ "finger_joint_limit_saturation_counts": {
138
+ "multi_middle_q1": 1671,
139
+ "multi_ring_q1": 21,
140
+ "multi_ring_q2": 51
141
+ },
142
+ "joint_limit_max_excess_by_joint": {
143
+ "index_q1": 0.013271281495690346,
144
+ "index_q2": 1.1998781523914204e-08,
145
+ "middle_q1": 0.007092274259775877,
146
+ "middle_q2": 0.13230398297309875,
147
+ "multi_index_q1": 0.0035281586460769176,
148
+ "multi_index_q2": 1.1711183134366365e-08,
149
+ "multi_middle_q1": 0.43434858322143555,
150
+ "multi_middle_q2": 0.01762932538986206,
151
+ "multi_pinky_q1": 0.004636600613594055,
152
+ "multi_pinky_q2": 0.0073623149655759335,
153
+ "multi_ring_q1": 0.21524950861930847,
154
+ "multi_ring_q2": 0.2497359812259674,
155
+ "pinky_q1": 0.054866448044776917,
156
+ "pinky_q2": 0.006702502723783255,
157
+ "ring_q1": 0.0579492449760437,
158
+ "ring_q2": 0.047779981046915054,
159
+ "thumb_q2": 0.004173769615590572
160
+ },
161
+ "coverage": {
162
+ "hard_limit_source": {
163
+ "available": 50,
164
+ "unavailable": 0,
165
+ "unknown": 0
166
+ },
167
+ "joint_role_manifest": {
168
+ "available": 50,
169
+ "unavailable": 0,
170
+ "unknown": 0
171
+ }
172
+ },
173
+ "hard_limit_sources": {
174
+ "runtime.joint_pos_limits": 50
175
+ },
176
+ "soft_limit_sources": {
177
+ "runtime.soft_joint_pos_limits": 50
178
+ },
179
+ "unknown_runtime_joint_names": [],
180
+ "baseline_checks": {
181
+ "post_home": {
182
+ "checked": 50,
183
+ "hard_violation_episodes": 0
184
+ },
185
+ "post_reset": {
186
+ "checked": 50,
187
+ "hard_violation_episodes": 0
188
+ }
189
+ }
190
+ }
191
+ },
192
+ "robot_motion": {
193
+ "joint_acceleration_rms": 10.780090312119274,
194
+ "joint_effort_rms": 103.24634194489121,
195
+ "joint_jerk_rms": 1058.6582296448098,
196
+ "joint_velocity_rms": 0.821079000684302
197
+ },
198
+ "grasp": {
199
+ "kinematic_grasp_stability_index": 0.9409188706220296,
200
+ "mean_kinematic_grasp_stability": 0.6440982520857435,
201
+ "per_object_best_kinematic_grasp_stability": {
202
+ "obj_194_wineglass_1": 0.8382408220158413,
203
+ "obj_194_wineglass_2": 0.8916911543960687,
204
+ "obj_194_wineglass_3": 0.7041562910118635
205
+ },
206
+ "grasp_gsi_diagnostics": {
207
+ "mean_anti_slip_score": 0.8801880772132097,
208
+ "mean_duration_score": 0.5857682397355779,
209
+ "mean_hold_score": 0.30600007991595213,
210
+ "mean_kinematic_stability_score": 0.22040587869504685,
211
+ "mean_legacy_gsi": 0.4709649945887714,
212
+ "mean_lift_score": 0.345025031986533,
213
+ "mean_motion_stability_score": 0.28339848314085114
214
+ }
215
+ },
216
+ "tool": {
217
+ "tool_selection_accuracy": null,
218
+ "tool_switch_success_rate": null,
219
+ "tool_switch_total": 0,
220
+ "tool_switch_successful": 0
221
+ },
222
+ "runtime": {
223
+ "avg_steps": 2980.9,
224
+ "error_count": 0
225
+ }
226
+ }
227
+ }
mora_v0/per_task_eval/03_wine_glass_plate_balance/Inv/native_shards/000/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/03_wine_glass_plate_balance/Inv/native_shards/000/summary.json ADDED
@@ -0,0 +1,209 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2000step",
5
+ "base_seed": 100300000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 25,
9
+ "successes": 8,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 3246,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.32,
50
+ "success_rate_ci": {
51
+ "mean": 0.32,
52
+ "ci_low": 0.16,
53
+ "ci_high": 0.52
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.7466666666666667
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 35.86041666666591
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.32,
64
+ "hard_violation_rate": 0.56,
65
+ "drop_rate": 0.56,
66
+ "high_speed_violation_rate": 0.0,
67
+ "safety_violation_step_rate": 0.000303888390082188
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.36,
74
+ "at_end_success_rate": 0.32,
75
+ "task_family_success_rate": {
76
+ "table_setting": 0.32
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.7066666666666667,
81
+ "current_chain_depth_progress_score": 0.7066666666666667,
82
+ "latched_chain_depth_progress_score": 0.7466666666666667,
83
+ "avg_current_chain_depth": 2.12,
84
+ "avg_latched_chain_depth": 2.24
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 1.0109124169725334,
88
+ "expert_time_s": 36.06666666666666,
89
+ "expert_time_step": 2164,
90
+ "avg_steps_to_success": 2151.625,
91
+ "avg_policy_steps_to_success": 717.375,
92
+ "avg_policy_queries_to_success": 36.125
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.000303888390082188,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_194_wineglass_1": 9,
98
+ "drop_to_table_obj_194_wineglass_2": 5,
99
+ "drop_to_table_obj_194_wineglass_3": 8
100
+ },
101
+ "robot_constraint_violation_rate": 0.36,
102
+ "finger_joint_limit_saturation_rate": 0.0,
103
+ "joint_limit_active_step_rate": 0.0006354029974445749,
104
+ "finger_joint_limit_saturation_step_rate": 0.0,
105
+ "max_joint_limit_excess_rad": 0.06844606250524521,
106
+ "robot_constraint_diagnostics": {
107
+ "joint_limit_soft_counts": {
108
+ "multi_ring_q1": 34,
109
+ "multi_ring_q2": 3,
110
+ "pinky_q1": 9,
111
+ "ring_q2": 3
112
+ },
113
+ "joint_limit_hard_counts": {
114
+ "index_q1": 3,
115
+ "multi_pinky_q1": 2,
116
+ "multi_ring_q1": 83,
117
+ "multi_ring_q2": 1,
118
+ "pinky_q1": 3,
119
+ "ring_q1": 5,
120
+ "ring_q2": 8
121
+ },
122
+ "joint_limit_hard_episode_events": 101,
123
+ "finger_joint_limit_saturation_counts": {},
124
+ "joint_limit_max_excess_by_joint": {
125
+ "index_q1": 0.005986154079437256,
126
+ "index_q2": 1.1073232109026776e-08,
127
+ "middle_q1": 0.0038528216537088156,
128
+ "middle_q2": 0.0016824461054056883,
129
+ "multi_index_q1": 0.00133789936080575,
130
+ "multi_index_q2": 1.124725823586914e-08,
131
+ "multi_middle_q1": 0.0022346919868141413,
132
+ "multi_middle_q2": 0.002004175679758191,
133
+ "multi_pinky_q1": 0.007192529737949371,
134
+ "multi_pinky_q2": 0.00046597878099419177,
135
+ "multi_ring_q1": 0.025935903191566467,
136
+ "multi_ring_q2": 0.028830135241150856,
137
+ "pinky_q1": 0.06844606250524521,
138
+ "pinky_q2": 0.0046754265204072,
139
+ "ring_q1": 0.00973603967577219,
140
+ "ring_q2": 0.041158277541399,
141
+ "thumb_q2": 0.0036329044960439205
142
+ },
143
+ "coverage": {
144
+ "hard_limit_source": {
145
+ "available": 25,
146
+ "unavailable": 0,
147
+ "unknown": 0
148
+ },
149
+ "joint_role_manifest": {
150
+ "available": 25,
151
+ "unavailable": 0,
152
+ "unknown": 0
153
+ }
154
+ },
155
+ "hard_limit_sources": {
156
+ "runtime.joint_pos_limits": 25
157
+ },
158
+ "soft_limit_sources": {
159
+ "runtime.soft_joint_pos_limits": 25
160
+ },
161
+ "unknown_runtime_joint_names": [],
162
+ "baseline_checks": {
163
+ "post_home": {
164
+ "checked": 25,
165
+ "hard_violation_episodes": 0
166
+ },
167
+ "post_reset": {
168
+ "checked": 25,
169
+ "hard_violation_episodes": 0
170
+ }
171
+ }
172
+ }
173
+ },
174
+ "robot_motion": {
175
+ "joint_acceleration_rms": 11.017166685893278,
176
+ "joint_effort_rms": 104.69523107442613,
177
+ "joint_jerk_rms": 1086.509516715968,
178
+ "joint_velocity_rms": 0.8311616689297702
179
+ },
180
+ "grasp": {
181
+ "kinematic_grasp_stability_index": 0.9213454341143336,
182
+ "mean_kinematic_grasp_stability": 0.6057444524002165,
183
+ "per_object_best_kinematic_grasp_stability": {
184
+ "obj_194_wineglass_1": 0.7989414542709276,
185
+ "obj_194_wineglass_2": 0.7626891603152245,
186
+ "obj_194_wineglass_3": 0.7008127745806
187
+ },
188
+ "grasp_gsi_diagnostics": {
189
+ "mean_anti_slip_score": 0.8825358829170674,
190
+ "mean_duration_score": 0.5420681252964504,
191
+ "mean_hold_score": 0.2289597049471307,
192
+ "mean_kinematic_stability_score": 0.14251864709014514,
193
+ "mean_legacy_gsi": 0.4170608678092251,
194
+ "mean_lift_score": 0.25624363206139106,
195
+ "mean_motion_stability_score": 0.20748525467667997
196
+ }
197
+ },
198
+ "tool": {
199
+ "tool_selection_accuracy": null,
200
+ "tool_switch_success_rate": null,
201
+ "tool_switch_total": 0,
202
+ "tool_switch_successful": 0
203
+ },
204
+ "runtime": {
205
+ "avg_steps": 2895.8,
206
+ "error_count": 0
207
+ }
208
+ }
209
+ }
mora_v0/per_task_eval/03_wine_glass_plate_balance/Inv/native_shards/001/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/03_wine_glass_plate_balance/Inv/native_shards/001/summary.json ADDED
@@ -0,0 +1,212 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2000step",
5
+ "base_seed": 100300000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 25,
9
+ "successes": 5,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 3246,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.2,
50
+ "success_rate_ci": {
51
+ "mean": 0.2,
52
+ "ci_low": 0.04,
53
+ "ci_high": 0.36
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.72
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 35.25999999999928
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.08,
64
+ "hard_violation_rate": 0.88,
65
+ "drop_rate": 0.88,
66
+ "high_speed_violation_rate": 0.12,
67
+ "safety_violation_step_rate": 0.0005827968952819943
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.2,
74
+ "at_end_success_rate": 0.2,
75
+ "task_family_success_rate": {
76
+ "table_setting": 0.2
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.64,
81
+ "current_chain_depth_progress_score": 0.64,
82
+ "latched_chain_depth_progress_score": 0.72,
83
+ "avg_current_chain_depth": 1.92,
84
+ "avg_latched_chain_depth": 2.16
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 1.0274867958195641,
88
+ "expert_time_s": 36.06666666666666,
89
+ "expert_time_step": 2164,
90
+ "avg_steps_to_success": 2115.6,
91
+ "avg_policy_steps_to_success": 705.6,
92
+ "avg_policy_queries_to_success": 35.6
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.0004900792073962224,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_194_wineglass_1": 14,
98
+ "drop_to_table_obj_194_wineglass_2": 11,
99
+ "drop_to_table_obj_194_wineglass_3": 7,
100
+ "high_speed_distractor_00_clutter_040_large_marker": 1,
101
+ "high_speed_distractor_01_clutter_054_softball": 1,
102
+ "high_speed_obj_029_plate_1": 2,
103
+ "high_speed_obj_194_wineglass_1": 1
104
+ },
105
+ "robot_constraint_violation_rate": 0.52,
106
+ "finger_joint_limit_saturation_rate": 0.0,
107
+ "joint_limit_active_step_rate": 0.0005960422792656759,
108
+ "finger_joint_limit_saturation_step_rate": 0.0,
109
+ "max_joint_limit_excess_rad": 0.08711928874254227,
110
+ "robot_constraint_diagnostics": {
111
+ "joint_limit_soft_counts": {
112
+ "multi_pinky_q1": 11,
113
+ "pinky_q1": 14,
114
+ "ring_q1": 19,
115
+ "ring_q2": 9
116
+ },
117
+ "joint_limit_hard_counts": {
118
+ "index_q1": 1,
119
+ "multi_pinky_q1": 9,
120
+ "multi_ring_q2": 1,
121
+ "pinky_q1": 11,
122
+ "ring_q1": 9,
123
+ "ring_q2": 5
124
+ },
125
+ "joint_limit_hard_episode_events": 31,
126
+ "finger_joint_limit_saturation_counts": {},
127
+ "joint_limit_max_excess_by_joint": {
128
+ "index_q1": 0.005373318679630756,
129
+ "index_q2": 9.464217853860646e-09,
130
+ "middle_q1": 0.002761028939858079,
131
+ "middle_q2": 0.002844219096004963,
132
+ "multi_index_q1": 0.001624695141799748,
133
+ "multi_index_q2": 8.842273580000892e-09,
134
+ "multi_middle_q1": 0.0009412128711119294,
135
+ "multi_middle_q2": 0.00016801584570202976,
136
+ "multi_pinky_q1": 0.019029716029763222,
137
+ "multi_pinky_q2": 0.0006420841091312468,
138
+ "multi_ring_q1": 0.004608642775565386,
139
+ "multi_ring_q2": 0.006039965897798538,
140
+ "pinky_q1": 0.05777708441019058,
141
+ "pinky_q2": 0.0037078747991472483,
142
+ "ring_q1": 0.021077001467347145,
143
+ "ring_q2": 0.08711928874254227,
144
+ "thumb_q2": 0.0017954156501218677
145
+ },
146
+ "coverage": {
147
+ "hard_limit_source": {
148
+ "available": 25,
149
+ "unavailable": 0,
150
+ "unknown": 0
151
+ },
152
+ "joint_role_manifest": {
153
+ "available": 25,
154
+ "unavailable": 0,
155
+ "unknown": 0
156
+ }
157
+ },
158
+ "hard_limit_sources": {
159
+ "runtime.joint_pos_limits": 25
160
+ },
161
+ "soft_limit_sources": {
162
+ "runtime.soft_joint_pos_limits": 25
163
+ },
164
+ "unknown_runtime_joint_names": [],
165
+ "baseline_checks": {
166
+ "post_home": {
167
+ "checked": 25,
168
+ "hard_violation_episodes": 0
169
+ },
170
+ "post_reset": {
171
+ "checked": 25,
172
+ "hard_violation_episodes": 0
173
+ }
174
+ }
175
+ }
176
+ },
177
+ "robot_motion": {
178
+ "joint_acceleration_rms": 11.09772054573395,
179
+ "joint_effort_rms": 104.33174524853648,
180
+ "joint_jerk_rms": 1091.318641991692,
181
+ "joint_velocity_rms": 0.8470688325338257
182
+ },
183
+ "grasp": {
184
+ "kinematic_grasp_stability_index": 0.9516568712739121,
185
+ "mean_kinematic_grasp_stability": 0.6626155171988392,
186
+ "per_object_best_kinematic_grasp_stability": {
187
+ "obj_194_wineglass_1": 0.8677156370846281,
188
+ "obj_194_wineglass_2": 0.8606041112027155,
189
+ "obj_194_wineglass_3": 0.7339740893920681
190
+ },
191
+ "grasp_gsi_diagnostics": {
192
+ "mean_anti_slip_score": 0.9034928505248686,
193
+ "mean_duration_score": 0.6043426973469636,
194
+ "mean_hold_score": 0.2992105182164681,
195
+ "mean_kinematic_stability_score": 0.21992621941903043,
196
+ "mean_legacy_gsi": 0.4757104353750226,
197
+ "mean_lift_score": 0.3363127669690339,
198
+ "mean_motion_stability_score": 0.27829412350110905
199
+ }
200
+ },
201
+ "tool": {
202
+ "tool_selection_accuracy": null,
203
+ "tool_switch_success_rate": null,
204
+ "tool_switch_total": 0,
205
+ "tool_switch_successful": 0
206
+ },
207
+ "runtime": {
208
+ "avg_steps": 3019.92,
209
+ "error_count": 0
210
+ }
211
+ }
212
+ }
mora_v0/per_task_eval/03_wine_glass_plate_balance/Inv/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/03_wine_glass_plate_balance/Inv/summary.json ADDED
@@ -0,0 +1,215 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2000step",
5
+ "base_seed": 100300000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 50,
9
+ "successes": 13,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 3246,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.26,
50
+ "success_rate_ci": {
51
+ "mean": 0.26,
52
+ "ci_low": 0.14,
53
+ "ci_high": 0.38
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.7333333333333333
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 35.62948717948644
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.2,
64
+ "hard_violation_rate": 0.72,
65
+ "drop_rate": 0.72,
66
+ "high_speed_violation_rate": 0.06,
67
+ "safety_violation_step_rate": 0.0004462685860723631
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.28,
74
+ "at_end_success_rate": 0.26,
75
+ "task_family_success_rate": {
76
+ "table_setting": 0.26
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.6733333333333333,
81
+ "current_chain_depth_progress_score": 0.6733333333333333,
82
+ "latched_chain_depth_progress_score": 0.7333333333333333,
83
+ "avg_current_chain_depth": 2.02,
84
+ "avg_latched_chain_depth": 2.2
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 1.0172871780675452,
88
+ "expert_time_s": 36.06666666666666,
89
+ "expert_time_step": 2164,
90
+ "avg_steps_to_success": 2137.769230769231,
91
+ "avg_policy_steps_to_success": 712.8461538461538,
92
+ "avg_policy_queries_to_success": 35.92307692307692
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.0003989370693677186,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_194_wineglass_1": 23,
98
+ "drop_to_table_obj_194_wineglass_2": 16,
99
+ "drop_to_table_obj_194_wineglass_3": 15,
100
+ "high_speed_distractor_00_clutter_040_large_marker": 1,
101
+ "high_speed_distractor_01_clutter_054_softball": 1,
102
+ "high_speed_obj_029_plate_1": 2,
103
+ "high_speed_obj_194_wineglass_1": 1
104
+ },
105
+ "robot_constraint_violation_rate": 0.44,
106
+ "finger_joint_limit_saturation_rate": 0.0,
107
+ "joint_limit_active_step_rate": 0.0006153097171603794,
108
+ "finger_joint_limit_saturation_step_rate": 0.0,
109
+ "max_joint_limit_excess_rad": 0.08711928874254227,
110
+ "robot_constraint_diagnostics": {
111
+ "joint_limit_soft_counts": {
112
+ "multi_pinky_q1": 11,
113
+ "multi_ring_q1": 34,
114
+ "multi_ring_q2": 3,
115
+ "pinky_q1": 23,
116
+ "ring_q1": 19,
117
+ "ring_q2": 12
118
+ },
119
+ "joint_limit_hard_counts": {
120
+ "index_q1": 4,
121
+ "multi_pinky_q1": 11,
122
+ "multi_ring_q1": 83,
123
+ "multi_ring_q2": 2,
124
+ "pinky_q1": 14,
125
+ "ring_q1": 14,
126
+ "ring_q2": 13
127
+ },
128
+ "joint_limit_hard_episode_events": 132,
129
+ "finger_joint_limit_saturation_counts": {},
130
+ "joint_limit_max_excess_by_joint": {
131
+ "index_q1": 0.005986154079437256,
132
+ "index_q2": 1.1073232109026776e-08,
133
+ "middle_q1": 0.0038528216537088156,
134
+ "middle_q2": 0.002844219096004963,
135
+ "multi_index_q1": 0.001624695141799748,
136
+ "multi_index_q2": 1.124725823586914e-08,
137
+ "multi_middle_q1": 0.0022346919868141413,
138
+ "multi_middle_q2": 0.002004175679758191,
139
+ "multi_pinky_q1": 0.019029716029763222,
140
+ "multi_pinky_q2": 0.0006420841091312468,
141
+ "multi_ring_q1": 0.025935903191566467,
142
+ "multi_ring_q2": 0.028830135241150856,
143
+ "pinky_q1": 0.06844606250524521,
144
+ "pinky_q2": 0.0046754265204072,
145
+ "ring_q1": 0.021077001467347145,
146
+ "ring_q2": 0.08711928874254227,
147
+ "thumb_q2": 0.0036329044960439205
148
+ },
149
+ "coverage": {
150
+ "hard_limit_source": {
151
+ "available": 50,
152
+ "unavailable": 0,
153
+ "unknown": 0
154
+ },
155
+ "joint_role_manifest": {
156
+ "available": 50,
157
+ "unavailable": 0,
158
+ "unknown": 0
159
+ }
160
+ },
161
+ "hard_limit_sources": {
162
+ "runtime.joint_pos_limits": 50
163
+ },
164
+ "soft_limit_sources": {
165
+ "runtime.soft_joint_pos_limits": 50
166
+ },
167
+ "unknown_runtime_joint_names": [],
168
+ "baseline_checks": {
169
+ "post_home": {
170
+ "checked": 50,
171
+ "hard_violation_episodes": 0
172
+ },
173
+ "post_reset": {
174
+ "checked": 50,
175
+ "hard_violation_episodes": 0
176
+ }
177
+ }
178
+ }
179
+ },
180
+ "robot_motion": {
181
+ "joint_acceleration_rms": 11.057443615813613,
182
+ "joint_effort_rms": 104.5134881614813,
183
+ "joint_jerk_rms": 1088.91407935383,
184
+ "joint_velocity_rms": 0.8391152507317979
185
+ },
186
+ "grasp": {
187
+ "kinematic_grasp_stability_index": 0.9365011526941228,
188
+ "mean_kinematic_grasp_stability": 0.6341799847995279,
189
+ "per_object_best_kinematic_grasp_stability": {
190
+ "obj_194_wineglass_1": 0.8333285456777778,
191
+ "obj_194_wineglass_2": 0.81164663575897,
192
+ "obj_194_wineglass_3": 0.7173934319863341
193
+ },
194
+ "grasp_gsi_diagnostics": {
195
+ "mean_anti_slip_score": 0.893014366720968,
196
+ "mean_duration_score": 0.573205411321707,
197
+ "mean_hold_score": 0.2640851115817994,
198
+ "mean_kinematic_stability_score": 0.18122243325458778,
199
+ "mean_legacy_gsi": 0.44638565159212384,
200
+ "mean_lift_score": 0.29627819951521245,
201
+ "mean_motion_stability_score": 0.24288968908889452
202
+ }
203
+ },
204
+ "tool": {
205
+ "tool_selection_accuracy": null,
206
+ "tool_switch_success_rate": null,
207
+ "tool_switch_total": 0,
208
+ "tool_switch_successful": 0
209
+ },
210
+ "runtime": {
211
+ "avg_steps": 2957.86,
212
+ "error_count": 0
213
+ }
214
+ }
215
+ }
mora_v0/per_task_eval/03_wine_glass_plate_balance/None/native_shards/000/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/03_wine_glass_plate_balance/None/native_shards/000/summary.json ADDED
@@ -0,0 +1,206 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2000step",
5
+ "base_seed": 100300000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 25,
9
+ "successes": 11,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 3246,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.44,
50
+ "success_rate_ci": {
51
+ "mean": 0.44,
52
+ "ci_low": 0.24,
53
+ "ci_high": 0.64
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.7733333333333333
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 38.05606060605973
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.32,
64
+ "hard_violation_rate": 0.56,
65
+ "drop_rate": 0.56,
66
+ "high_speed_violation_rate": 0.0,
67
+ "safety_violation_step_rate": 0.0003117869644704582
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.44,
74
+ "at_end_success_rate": 0.44,
75
+ "task_family_success_rate": {
76
+ "table_setting": 0.44
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.7333333333333333,
81
+ "current_chain_depth_progress_score": 0.7333333333333333,
82
+ "latched_chain_depth_progress_score": 0.7733333333333333,
83
+ "avg_current_chain_depth": 2.2,
84
+ "avg_latched_chain_depth": 2.32
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.9514750680189635,
88
+ "expert_time_s": 36.06666666666666,
89
+ "expert_time_step": 2164,
90
+ "avg_steps_to_success": 2283.3636363636365,
91
+ "avg_policy_steps_to_success": 761.3636363636364,
92
+ "avg_policy_queries_to_success": 38.45454545454545
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.0003117869644704582,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_194_wineglass_1": 6,
98
+ "drop_to_table_obj_194_wineglass_2": 7,
99
+ "drop_to_table_obj_194_wineglass_3": 9
100
+ },
101
+ "robot_constraint_violation_rate": 0.24,
102
+ "finger_joint_limit_saturation_rate": 0.0,
103
+ "joint_limit_active_step_rate": 0.0004251640424597157,
104
+ "finger_joint_limit_saturation_step_rate": 0.0,
105
+ "max_joint_limit_excess_rad": 0.07819271087646484,
106
+ "robot_constraint_diagnostics": {
107
+ "joint_limit_soft_counts": {
108
+ "multi_pinky_q2": 11,
109
+ "multi_ring_q2": 16,
110
+ "pinky_q1": 12,
111
+ "ring_q2": 6
112
+ },
113
+ "joint_limit_hard_counts": {
114
+ "multi_pinky_q2": 2,
115
+ "multi_ring_q2": 3,
116
+ "pinky_q1": 2,
117
+ "pinky_q2": 1,
118
+ "ring_q1": 5,
119
+ "ring_q2": 4,
120
+ "thumb_q2": 1
121
+ },
122
+ "joint_limit_hard_episode_events": 13,
123
+ "finger_joint_limit_saturation_counts": {},
124
+ "joint_limit_max_excess_by_joint": {
125
+ "index_q1": 0.003644155338406563,
126
+ "index_q2": 1.3093375272887897e-08,
127
+ "middle_q1": 0.0019025503424927592,
128
+ "middle_q2": 0.0008063862333074212,
129
+ "multi_index_q2": 3.4527258829797347e-09,
130
+ "multi_pinky_q1": 0.002839241409674287,
131
+ "multi_pinky_q2": 0.054891206324100494,
132
+ "multi_ring_q1": 0.003343795193359256,
133
+ "multi_ring_q2": 0.07194944471120834,
134
+ "pinky_q1": 0.07819271087646484,
135
+ "pinky_q2": 0.0056563434191048145,
136
+ "ring_q1": 0.006019518710672855,
137
+ "ring_q2": 0.05637228116393089,
138
+ "thumb_q2": 0.0054933312349021435
139
+ },
140
+ "coverage": {
141
+ "hard_limit_source": {
142
+ "available": 25,
143
+ "unavailable": 0,
144
+ "unknown": 0
145
+ },
146
+ "joint_role_manifest": {
147
+ "available": 25,
148
+ "unavailable": 0,
149
+ "unknown": 0
150
+ }
151
+ },
152
+ "hard_limit_sources": {
153
+ "runtime.joint_pos_limits": 25
154
+ },
155
+ "soft_limit_sources": {
156
+ "runtime.soft_joint_pos_limits": 25
157
+ },
158
+ "unknown_runtime_joint_names": [],
159
+ "baseline_checks": {
160
+ "post_home": {
161
+ "checked": 25,
162
+ "hard_violation_episodes": 0
163
+ },
164
+ "post_reset": {
165
+ "checked": 25,
166
+ "hard_violation_episodes": 0
167
+ }
168
+ }
169
+ }
170
+ },
171
+ "robot_motion": {
172
+ "joint_acceleration_rms": 10.39194568912505,
173
+ "joint_effort_rms": 102.95838166456693,
174
+ "joint_jerk_rms": 1025.5955439770876,
175
+ "joint_velocity_rms": 0.8137788147733255
176
+ },
177
+ "grasp": {
178
+ "kinematic_grasp_stability_index": 0.9192030611131372,
179
+ "mean_kinematic_grasp_stability": 0.6067561205698627,
180
+ "per_object_best_kinematic_grasp_stability": {
181
+ "obj_194_wineglass_1": 0.8083203085823123,
182
+ "obj_194_wineglass_2": 0.6865198968222266,
183
+ "obj_194_wineglass_3": 0.771268188809384
184
+ },
185
+ "grasp_gsi_diagnostics": {
186
+ "mean_anti_slip_score": 0.9430330340106704,
187
+ "mean_duration_score": 0.5035289384807057,
188
+ "mean_hold_score": 0.20022418942984027,
189
+ "mean_kinematic_stability_score": 0.15090493387925563,
190
+ "mean_legacy_gsi": 0.4057562781564782,
191
+ "mean_lift_score": 0.22865894407589785,
192
+ "mean_motion_stability_score": 0.18339606978000664
193
+ }
194
+ },
195
+ "tool": {
196
+ "tool_selection_accuracy": null,
197
+ "tool_switch_success_rate": null,
198
+ "tool_switch_total": 0,
199
+ "tool_switch_successful": 0
200
+ },
201
+ "runtime": {
202
+ "avg_steps": 2822.44,
203
+ "error_count": 0
204
+ }
205
+ }
206
+ }
mora_v0/per_task_eval/03_wine_glass_plate_balance/None/native_shards/001/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/03_wine_glass_plate_balance/None/native_shards/001/summary.json ADDED
@@ -0,0 +1,219 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2000step",
5
+ "base_seed": 100300000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 25,
9
+ "successes": 7,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 3246,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.28,
50
+ "success_rate_ci": {
51
+ "mean": 0.28,
52
+ "ci_low": 0.12,
53
+ "ci_high": 0.44099999999998546
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.6799999999999999
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 36.483333333332546
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.28,
64
+ "hard_violation_rate": 0.64,
65
+ "drop_rate": 0.64,
66
+ "high_speed_violation_rate": 0.04,
67
+ "safety_violation_step_rate": 0.0003796558690729617
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.32,
74
+ "at_end_success_rate": 0.28,
75
+ "task_family_success_rate": {
76
+ "table_setting": 0.28
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.6,
81
+ "current_chain_depth_progress_score": 0.6,
82
+ "latched_chain_depth_progress_score": 0.6799999999999999,
83
+ "avg_current_chain_depth": 1.8,
84
+ "avg_latched_chain_depth": 2.04
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.9934126052192729,
88
+ "expert_time_s": 36.06666666666666,
89
+ "expert_time_step": 2164,
90
+ "avg_steps_to_success": 2189,
91
+ "avg_policy_steps_to_success": 729.8571428571429,
92
+ "avg_policy_queries_to_success": 36.714285714285715
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.0003796558690729617,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_194_wineglass_1": 10,
98
+ "drop_to_table_obj_194_wineglass_2": 9,
99
+ "drop_to_table_obj_194_wineglass_3": 8,
100
+ "high_speed_obj_194_wineglass_1": 1
101
+ },
102
+ "robot_constraint_violation_rate": 0.44,
103
+ "finger_joint_limit_saturation_rate": 0.04,
104
+ "joint_limit_active_step_rate": 0.004935526297948503,
105
+ "finger_joint_limit_saturation_step_rate": 0.003769440414367263,
106
+ "max_joint_limit_excess_rad": 1.0974076986312866,
107
+ "robot_constraint_diagnostics": {
108
+ "joint_limit_soft_counts": {
109
+ "index_q1": 19,
110
+ "middle_q1": 291,
111
+ "pinky_q1": 19,
112
+ "pinky_q2": 194,
113
+ "ring_q1": 295,
114
+ "ring_q2": 247
115
+ },
116
+ "joint_limit_hard_counts": {
117
+ "index_q1": 1,
118
+ "middle_q1": 11,
119
+ "multi_ring_q1": 2,
120
+ "pinky_q1": 8,
121
+ "pinky_q2": 4,
122
+ "ring_q1": 11,
123
+ "ring_q2": 5
124
+ },
125
+ "joint_limit_hard_episode_events": 30,
126
+ "finger_joint_limit_saturation_counts": {
127
+ "middle_q1": 191,
128
+ "pinky_q2": 167,
129
+ "ring_q1": 248,
130
+ "ring_q2": 202
131
+ },
132
+ "joint_limit_max_excess_by_joint": {
133
+ "index_q1": 0.06295124441385269,
134
+ "index_q2": 8.24686541278652e-09,
135
+ "middle_q1": 0.4764763116836548,
136
+ "middle_q2": 0.0037700049579143524,
137
+ "multi_index_q1": 0.0020975673105567694,
138
+ "multi_index_q2": 1.1288080692395397e-08,
139
+ "multi_middle_q1": 0.0033539971336722374,
140
+ "multi_middle_q2": 0.000765387318097055,
141
+ "multi_pinky_q1": 0.0025628528092056513,
142
+ "multi_pinky_q2": 0.0007166645373217762,
143
+ "multi_ring_q1": 0.006885673385113478,
144
+ "multi_ring_q2": 0.004084001760929823,
145
+ "pinky_q1": 0.07525113970041275,
146
+ "pinky_q2": 1.0903534889221191,
147
+ "ring_q1": 1.0974076986312866,
148
+ "ring_q2": 0.5466373562812805,
149
+ "thumb_q2": 0.002625890774652362,
150
+ "xarm_l_joint2": 0.00027060508728027344,
151
+ "xarm_l_joint6": 0.0010161399841308594
152
+ },
153
+ "coverage": {
154
+ "hard_limit_source": {
155
+ "available": 25,
156
+ "unavailable": 0,
157
+ "unknown": 0
158
+ },
159
+ "joint_role_manifest": {
160
+ "available": 25,
161
+ "unavailable": 0,
162
+ "unknown": 0
163
+ }
164
+ },
165
+ "hard_limit_sources": {
166
+ "runtime.joint_pos_limits": 25
167
+ },
168
+ "soft_limit_sources": {
169
+ "runtime.soft_joint_pos_limits": 25
170
+ },
171
+ "unknown_runtime_joint_names": [],
172
+ "baseline_checks": {
173
+ "post_home": {
174
+ "checked": 25,
175
+ "hard_violation_episodes": 0
176
+ },
177
+ "post_reset": {
178
+ "checked": 25,
179
+ "hard_violation_episodes": 0
180
+ }
181
+ }
182
+ }
183
+ },
184
+ "robot_motion": {
185
+ "joint_acceleration_rms": 11.215371964285216,
186
+ "joint_effort_rms": 105.28881781599972,
187
+ "joint_jerk_rms": 1106.030261872567,
188
+ "joint_velocity_rms": 0.8420777265519906
189
+ },
190
+ "grasp": {
191
+ "kinematic_grasp_stability_index": 0.9167570760106218,
192
+ "mean_kinematic_grasp_stability": 0.5428977876664652,
193
+ "per_object_best_kinematic_grasp_stability": {
194
+ "obj_194_wineglass_1": 0.822465683129762,
195
+ "obj_194_wineglass_2": 0.777464579796114,
196
+ "obj_194_wineglass_3": 0.7204316935058916
197
+ },
198
+ "grasp_gsi_diagnostics": {
199
+ "mean_anti_slip_score": 0.855692904868314,
200
+ "mean_duration_score": 0.5714501970077265,
201
+ "mean_hold_score": 0.2661784624351442,
202
+ "mean_kinematic_stability_score": 0.1460353002872876,
203
+ "mean_legacy_gsi": 0.4403028932240471,
204
+ "mean_lift_score": 0.31114590754466953,
205
+ "mean_motion_stability_score": 0.24407516537539067
206
+ }
207
+ },
208
+ "tool": {
209
+ "tool_selection_accuracy": null,
210
+ "tool_switch_success_rate": null,
211
+ "tool_switch_total": 0,
212
+ "tool_switch_successful": 0
213
+ },
214
+ "runtime": {
215
+ "avg_steps": 2950.04,
216
+ "error_count": 0
217
+ }
218
+ }
219
+ }
mora_v0/per_task_eval/03_wine_glass_plate_balance/None/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/03_wine_glass_plate_balance/None/summary.json ADDED
@@ -0,0 +1,224 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2000step",
5
+ "base_seed": 100300000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 50,
9
+ "successes": 18,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 3246,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.36,
50
+ "success_rate_ci": {
51
+ "mean": 0.36,
52
+ "ci_low": 0.24,
53
+ "ci_high": 0.5
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.7266666666666667
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 37.444444444443604
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.3,
64
+ "hard_violation_rate": 0.6,
65
+ "drop_rate": 0.6,
66
+ "high_speed_violation_rate": 0.02,
67
+ "safety_violation_step_rate": 0.00034647153389877485
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.38,
74
+ "at_end_success_rate": 0.36,
75
+ "task_family_success_rate": {
76
+ "table_setting": 0.36
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.6666666666666666,
81
+ "current_chain_depth_progress_score": 0.6666666666666666,
82
+ "latched_chain_depth_progress_score": 0.7266666666666667,
83
+ "avg_current_chain_depth": 2,
84
+ "avg_latched_chain_depth": 2.18
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.9677841102635283,
88
+ "expert_time_s": 36.06666666666666,
89
+ "expert_time_step": 2164,
90
+ "avg_steps_to_success": 2246.6666666666665,
91
+ "avg_policy_steps_to_success": 749.1111111111111,
92
+ "avg_policy_queries_to_success": 37.77777777777778
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.00034647153389877485,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_194_wineglass_1": 16,
98
+ "drop_to_table_obj_194_wineglass_2": 16,
99
+ "drop_to_table_obj_194_wineglass_3": 17,
100
+ "high_speed_obj_194_wineglass_1": 1
101
+ },
102
+ "robot_constraint_violation_rate": 0.34,
103
+ "finger_joint_limit_saturation_rate": 0.02,
104
+ "joint_limit_active_step_rate": 0.002730195687122346,
105
+ "finger_joint_limit_saturation_step_rate": 0.0019263817284771883,
106
+ "max_joint_limit_excess_rad": 1.0974076986312866,
107
+ "robot_constraint_diagnostics": {
108
+ "joint_limit_soft_counts": {
109
+ "index_q1": 19,
110
+ "middle_q1": 291,
111
+ "multi_pinky_q2": 11,
112
+ "multi_ring_q2": 16,
113
+ "pinky_q1": 31,
114
+ "pinky_q2": 194,
115
+ "ring_q1": 295,
116
+ "ring_q2": 253
117
+ },
118
+ "joint_limit_hard_counts": {
119
+ "index_q1": 1,
120
+ "middle_q1": 11,
121
+ "multi_pinky_q2": 2,
122
+ "multi_ring_q1": 2,
123
+ "multi_ring_q2": 3,
124
+ "pinky_q1": 10,
125
+ "pinky_q2": 5,
126
+ "ring_q1": 16,
127
+ "ring_q2": 9,
128
+ "thumb_q2": 1
129
+ },
130
+ "joint_limit_hard_episode_events": 43,
131
+ "finger_joint_limit_saturation_counts": {
132
+ "middle_q1": 191,
133
+ "pinky_q2": 167,
134
+ "ring_q1": 248,
135
+ "ring_q2": 202
136
+ },
137
+ "joint_limit_max_excess_by_joint": {
138
+ "index_q1": 0.06295124441385269,
139
+ "index_q2": 1.3093375272887897e-08,
140
+ "middle_q1": 0.4764763116836548,
141
+ "middle_q2": 0.0037700049579143524,
142
+ "multi_index_q1": 0.0020975673105567694,
143
+ "multi_index_q2": 1.1288080692395397e-08,
144
+ "multi_middle_q1": 0.0033539971336722374,
145
+ "multi_middle_q2": 0.000765387318097055,
146
+ "multi_pinky_q1": 0.002839241409674287,
147
+ "multi_pinky_q2": 0.054891206324100494,
148
+ "multi_ring_q1": 0.006885673385113478,
149
+ "multi_ring_q2": 0.07194944471120834,
150
+ "pinky_q1": 0.07819271087646484,
151
+ "pinky_q2": 1.0903534889221191,
152
+ "ring_q1": 1.0974076986312866,
153
+ "ring_q2": 0.5466373562812805,
154
+ "thumb_q2": 0.0054933312349021435,
155
+ "xarm_l_joint2": 0.00027060508728027344,
156
+ "xarm_l_joint6": 0.0010161399841308594
157
+ },
158
+ "coverage": {
159
+ "hard_limit_source": {
160
+ "available": 50,
161
+ "unavailable": 0,
162
+ "unknown": 0
163
+ },
164
+ "joint_role_manifest": {
165
+ "available": 50,
166
+ "unavailable": 0,
167
+ "unknown": 0
168
+ }
169
+ },
170
+ "hard_limit_sources": {
171
+ "runtime.joint_pos_limits": 50
172
+ },
173
+ "soft_limit_sources": {
174
+ "runtime.soft_joint_pos_limits": 50
175
+ },
176
+ "unknown_runtime_joint_names": [],
177
+ "baseline_checks": {
178
+ "post_home": {
179
+ "checked": 50,
180
+ "hard_violation_episodes": 0
181
+ },
182
+ "post_reset": {
183
+ "checked": 50,
184
+ "hard_violation_episodes": 0
185
+ }
186
+ }
187
+ }
188
+ },
189
+ "robot_motion": {
190
+ "joint_acceleration_rms": 10.803658826705133,
191
+ "joint_effort_rms": 104.12359974028332,
192
+ "joint_jerk_rms": 1065.8129029248273,
193
+ "joint_velocity_rms": 0.8279282706626581
194
+ },
195
+ "grasp": {
196
+ "kinematic_grasp_stability_index": 0.9179800685618795,
197
+ "mean_kinematic_grasp_stability": 0.574826954118164,
198
+ "per_object_best_kinematic_grasp_stability": {
199
+ "obj_194_wineglass_1": 0.8153929958560372,
200
+ "obj_194_wineglass_2": 0.7319922383091704,
201
+ "obj_194_wineglass_3": 0.7458499411576378
202
+ },
203
+ "grasp_gsi_diagnostics": {
204
+ "mean_anti_slip_score": 0.8993629694394921,
205
+ "mean_duration_score": 0.5374895677442161,
206
+ "mean_hold_score": 0.23320132593249224,
207
+ "mean_kinematic_stability_score": 0.14847011708327162,
208
+ "mean_legacy_gsi": 0.4230295856902626,
209
+ "mean_lift_score": 0.2699024258102837,
210
+ "mean_motion_stability_score": 0.21373561757769866
211
+ }
212
+ },
213
+ "tool": {
214
+ "tool_selection_accuracy": null,
215
+ "tool_switch_success_rate": null,
216
+ "tool_switch_total": 0,
217
+ "tool_switch_successful": 0
218
+ },
219
+ "runtime": {
220
+ "avg_steps": 2886.24,
221
+ "error_count": 0
222
+ }
223
+ }
224
+ }
mora_v0/per_task_eval/06_fruit_bowl_loading/Equi/native_shards/000/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/06_fruit_bowl_loading/Equi/native_shards/000/summary.json ADDED
@@ -0,0 +1,268 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2k",
5
+ "base_seed": 100600000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 25,
9
+ "successes": 6,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 1920,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.24,
50
+ "success_rate_ci": {
51
+ "mean": 0.24,
52
+ "ci_low": 0.08,
53
+ "ci_high": 0.4
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.66
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 22.916666666666647
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.24,
64
+ "hard_violation_rate": 0.68,
65
+ "drop_rate": 0.6,
66
+ "high_speed_violation_rate": 0.52,
67
+ "safety_violation_step_rate": 0.000894254415381176
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.24,
74
+ "at_end_success_rate": 0.24,
75
+ "task_family_success_rate": {
76
+ "rigid_pick_place": 0.24
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.6,
81
+ "current_chain_depth_progress_score": 0.6666666666666666,
82
+ "latched_chain_depth_progress_score": 0.7066666666666667,
83
+ "avg_current_chain_depth": 2,
84
+ "avg_latched_chain_depth": 2.12
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.9318877300427653,
88
+ "expert_time_s": 21.333333333333332,
89
+ "expert_time_step": 1280,
90
+ "avg_steps_to_success": 1375,
91
+ "avg_policy_steps_to_success": 458.6666666666667,
92
+ "avg_policy_queries_to_success": 23.333333333333332
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.000782472613458529,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_011_banana_2": 2,
98
+ "drop_to_table_obj_013_apple_3": 7,
99
+ "drop_to_table_obj_013_apple_4": 12,
100
+ "high_speed_obj_013_apple_3": 11,
101
+ "high_speed_obj_013_apple_4": 3
102
+ },
103
+ "robot_constraint_violation_rate": 1.0,
104
+ "finger_joint_limit_saturation_rate": 1.0,
105
+ "joint_limit_active_step_rate": 0.14804381846635367,
106
+ "finger_joint_limit_saturation_step_rate": 0.09016320143080707,
107
+ "max_joint_limit_excess_rad": 1.5275760889053345,
108
+ "robot_constraint_diagnostics": {
109
+ "joint_limit_soft_counts": {
110
+ "left_index_1_joint": 665,
111
+ "left_index_2_joint": 35,
112
+ "left_little_1_joint": 762,
113
+ "left_little_2_joint": 30,
114
+ "left_middle_1_joint": 1118,
115
+ "left_middle_2_joint": 74,
116
+ "left_ring_1_joint": 864,
117
+ "left_ring_2_joint": 106,
118
+ "left_thumb_2_joint": 260,
119
+ "left_thumb_3_joint": 35,
120
+ "left_thumb_4_joint": 72,
121
+ "right_index_1_joint": 900,
122
+ "right_index_2_joint": 6,
123
+ "right_little_1_joint": 776,
124
+ "right_little_2_joint": 62,
125
+ "right_middle_1_joint": 1234,
126
+ "right_middle_2_joint": 6,
127
+ "right_ring_1_joint": 977,
128
+ "right_ring_2_joint": 42,
129
+ "right_thumb_1_joint": 13,
130
+ "right_thumb_2_joint": 3293,
131
+ "right_thumb_3_joint": 16,
132
+ "right_thumb_4_joint": 63
133
+ },
134
+ "joint_limit_hard_counts": {
135
+ "left_index_1_joint": 24,
136
+ "left_index_2_joint": 23,
137
+ "left_little_1_joint": 21,
138
+ "left_little_2_joint": 27,
139
+ "left_middle_1_joint": 27,
140
+ "left_middle_2_joint": 42,
141
+ "left_ring_1_joint": 27,
142
+ "left_ring_2_joint": 38,
143
+ "left_thumb_2_joint": 10,
144
+ "left_thumb_3_joint": 2,
145
+ "left_thumb_4_joint": 4,
146
+ "right_index_1_joint": 33,
147
+ "right_index_2_joint": 16,
148
+ "right_little_1_joint": 19,
149
+ "right_little_2_joint": 5,
150
+ "right_middle_1_joint": 45,
151
+ "right_middle_2_joint": 17,
152
+ "right_ring_1_joint": 37,
153
+ "right_ring_2_joint": 18,
154
+ "right_thumb_1_joint": 2,
155
+ "right_thumb_2_joint": 36,
156
+ "right_thumb_3_joint": 1,
157
+ "right_thumb_4_joint": 7
158
+ },
159
+ "joint_limit_hard_episode_events": 118,
160
+ "finger_joint_limit_saturation_counts": {
161
+ "left_index_1_joint": 357,
162
+ "left_little_1_joint": 538,
163
+ "left_middle_1_joint": 705,
164
+ "left_ring_1_joint": 500,
165
+ "left_thumb_2_joint": 104,
166
+ "left_thumb_3_joint": 20,
167
+ "left_thumb_4_joint": 34,
168
+ "right_index_1_joint": 509,
169
+ "right_little_1_joint": 620,
170
+ "right_little_2_joint": 17,
171
+ "right_middle_1_joint": 671,
172
+ "right_ring_1_joint": 627,
173
+ "right_thumb_2_joint": 1731,
174
+ "right_thumb_3_joint": 11,
175
+ "right_thumb_4_joint": 31
176
+ },
177
+ "joint_limit_max_excess_by_joint": {
178
+ "left_index_1_joint": 1.0526398420333862,
179
+ "left_index_2_joint": 0.10732898116111755,
180
+ "left_little_1_joint": 1.174699306488037,
181
+ "left_little_2_joint": 0.045096974819898605,
182
+ "left_middle_1_joint": 1.2612115144729614,
183
+ "left_middle_2_joint": 0.038757774978876114,
184
+ "left_ring_1_joint": 1.131105899810791,
185
+ "left_ring_2_joint": 0.09605004638433456,
186
+ "left_thumb_2_joint": 0.4470149576663971,
187
+ "left_thumb_3_joint": 0.4703221917152405,
188
+ "left_thumb_4_joint": 0.8424203395843506,
189
+ "right_index_1_joint": 0.7172218561172485,
190
+ "right_index_2_joint": 0.03866291418671608,
191
+ "right_little_1_joint": 1.5275760889053345,
192
+ "right_little_2_joint": 0.3231325149536133,
193
+ "right_middle_1_joint": 1.1832196712493896,
194
+ "right_middle_2_joint": 0.02850760892033577,
195
+ "right_ring_1_joint": 1.3478960990905762,
196
+ "right_ring_2_joint": 0.11619895696640015,
197
+ "right_thumb_1_joint": 0.15416760742664337,
198
+ "right_thumb_2_joint": 0.3947981297969818,
199
+ "right_thumb_3_joint": 0.40242546796798706,
200
+ "right_thumb_4_joint": 0.8027892112731934
201
+ },
202
+ "coverage": {
203
+ "hard_limit_source": {
204
+ "available": 25,
205
+ "unavailable": 0,
206
+ "unknown": 0
207
+ },
208
+ "joint_role_manifest": {
209
+ "available": 25,
210
+ "unavailable": 0,
211
+ "unknown": 0
212
+ }
213
+ },
214
+ "hard_limit_sources": {
215
+ "runtime.joint_pos_limits": 25
216
+ },
217
+ "soft_limit_sources": {
218
+ "runtime.soft_joint_pos_limits": 25
219
+ },
220
+ "unknown_runtime_joint_names": [],
221
+ "baseline_checks": {
222
+ "post_home": {
223
+ "checked": 25,
224
+ "hard_violation_episodes": 0
225
+ },
226
+ "post_reset": {
227
+ "checked": 25,
228
+ "hard_violation_episodes": 0
229
+ }
230
+ }
231
+ }
232
+ },
233
+ "robot_motion": {
234
+ "joint_acceleration_rms": 50.838360945623435,
235
+ "joint_effort_rms": 34.88074494259113,
236
+ "joint_jerk_rms": 5196.921538756235,
237
+ "joint_velocity_rms": 1.0061809872517755
238
+ },
239
+ "grasp": {
240
+ "kinematic_grasp_stability_index": 0.9341063404561603,
241
+ "mean_kinematic_grasp_stability": 0.5427542506339144,
242
+ "per_object_best_kinematic_grasp_stability": {
243
+ "obj_011_banana_2": 0.48367521084542536,
244
+ "obj_013_apple_3": 0.8784060520663695,
245
+ "obj_013_apple_4": 0.3129012029201666
246
+ },
247
+ "grasp_gsi_diagnostics": {
248
+ "mean_anti_slip_score": 0.9348683781702667,
249
+ "mean_duration_score": 0.1848949229147757,
250
+ "mean_hold_score": 0.16160924784374955,
251
+ "mean_kinematic_stability_score": 0.12318636414938677,
252
+ "mean_legacy_gsi": 0.32161938108959603,
253
+ "mean_lift_score": 0.2280393263113819,
254
+ "mean_motion_stability_score": 0.15038141445261013
255
+ }
256
+ },
257
+ "tool": {
258
+ "tool_selection_accuracy": null,
259
+ "tool_switch_success_rate": null,
260
+ "tool_switch_total": 0,
261
+ "tool_switch_successful": 0
262
+ },
263
+ "runtime": {
264
+ "avg_steps": 1789.2,
265
+ "error_count": 0
266
+ }
267
+ }
268
+ }
mora_v0/per_task_eval/06_fruit_bowl_loading/Equi/native_shards/001/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/06_fruit_bowl_loading/Equi/native_shards/001/summary.json ADDED
@@ -0,0 +1,268 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2k",
5
+ "base_seed": 100600000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 25,
9
+ "successes": 11,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 1920,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.44,
50
+ "success_rate_ci": {
51
+ "mean": 0.44,
52
+ "ci_low": 0.24,
53
+ "ci_high": 0.64
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.79
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 22.712121212121204
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.36,
64
+ "hard_violation_rate": 0.52,
65
+ "drop_rate": 0.44,
66
+ "high_speed_violation_rate": 0.28,
67
+ "safety_violation_step_rate": 0.000740386911870074
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.48,
74
+ "at_end_success_rate": 0.44,
75
+ "task_family_success_rate": {
76
+ "rigid_pick_place": 0.44
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.71,
81
+ "current_chain_depth_progress_score": 0.7733333333333333,
82
+ "latched_chain_depth_progress_score": 0.84,
83
+ "avg_current_chain_depth": 2.32,
84
+ "avg_latched_chain_depth": 2.52
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.940863229349727,
88
+ "expert_time_s": 21.333333333333332,
89
+ "expert_time_step": 1280,
90
+ "avg_steps_to_success": 1362.7272727272727,
91
+ "avg_policy_steps_to_success": 454.54545454545456,
92
+ "avg_policy_queries_to_success": 23.181818181818183
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.0005970862192500597,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_011_banana_2": 1,
98
+ "drop_to_table_obj_013_apple_3": 7,
99
+ "drop_to_table_obj_013_apple_4": 8,
100
+ "high_speed_obj_013_apple_3": 4,
101
+ "high_speed_obj_013_apple_4": 5
102
+ },
103
+ "robot_constraint_violation_rate": 1.0,
104
+ "finger_joint_limit_saturation_rate": 1.0,
105
+ "joint_limit_active_step_rate": 0.13522808693575353,
106
+ "finger_joint_limit_saturation_step_rate": 0.08125149271554813,
107
+ "max_joint_limit_excess_rad": 1.6511658430099487,
108
+ "robot_constraint_diagnostics": {
109
+ "joint_limit_soft_counts": {
110
+ "left_index_1_joint": 375,
111
+ "left_index_2_joint": 11,
112
+ "left_little_1_joint": 380,
113
+ "left_little_2_joint": 97,
114
+ "left_middle_1_joint": 522,
115
+ "left_middle_2_joint": 26,
116
+ "left_ring_1_joint": 492,
117
+ "left_ring_2_joint": 38,
118
+ "left_thumb_2_joint": 18,
119
+ "left_thumb_3_joint": 15,
120
+ "left_thumb_4_joint": 28,
121
+ "right_index_1_joint": 959,
122
+ "right_index_2_joint": 23,
123
+ "right_little_1_joint": 945,
124
+ "right_little_2_joint": 141,
125
+ "right_middle_1_joint": 1408,
126
+ "right_middle_2_joint": 34,
127
+ "right_ring_1_joint": 1249,
128
+ "right_ring_2_joint": 18,
129
+ "right_thumb_1_joint": 33,
130
+ "right_thumb_2_joint": 3262,
131
+ "right_thumb_3_joint": 48,
132
+ "right_thumb_4_joint": 174
133
+ },
134
+ "joint_limit_hard_counts": {
135
+ "left_index_1_joint": 21,
136
+ "left_index_2_joint": 18,
137
+ "left_little_1_joint": 14,
138
+ "left_little_2_joint": 26,
139
+ "left_middle_1_joint": 19,
140
+ "left_middle_2_joint": 44,
141
+ "left_ring_1_joint": 16,
142
+ "left_ring_2_joint": 41,
143
+ "left_thumb_2_joint": 1,
144
+ "left_thumb_3_joint": 1,
145
+ "left_thumb_4_joint": 1,
146
+ "right_index_1_joint": 50,
147
+ "right_index_2_joint": 9,
148
+ "right_little_1_joint": 36,
149
+ "right_little_2_joint": 11,
150
+ "right_middle_1_joint": 55,
151
+ "right_middle_2_joint": 17,
152
+ "right_ring_1_joint": 37,
153
+ "right_ring_2_joint": 14,
154
+ "right_thumb_1_joint": 3,
155
+ "right_thumb_2_joint": 45,
156
+ "right_thumb_3_joint": 5,
157
+ "right_thumb_4_joint": 12
158
+ },
159
+ "joint_limit_hard_episode_events": 132,
160
+ "finger_joint_limit_saturation_counts": {
161
+ "left_index_1_joint": 190,
162
+ "left_little_1_joint": 235,
163
+ "left_little_2_joint": 14,
164
+ "left_middle_1_joint": 303,
165
+ "left_ring_1_joint": 285,
166
+ "left_thumb_4_joint": 21,
167
+ "right_index_1_joint": 516,
168
+ "right_little_1_joint": 706,
169
+ "right_little_2_joint": 80,
170
+ "right_middle_1_joint": 859,
171
+ "right_ring_1_joint": 856,
172
+ "right_thumb_2_joint": 1881,
173
+ "right_thumb_3_joint": 7,
174
+ "right_thumb_4_joint": 94
175
+ },
176
+ "joint_limit_max_excess_by_joint": {
177
+ "left_index_1_joint": 0.708935022354126,
178
+ "left_index_2_joint": 0.01621445082128048,
179
+ "left_little_1_joint": 0.97555011510849,
180
+ "left_little_2_joint": 0.280036062002182,
181
+ "left_middle_1_joint": 0.8640082478523254,
182
+ "left_middle_2_joint": 0.02650519274175167,
183
+ "left_ring_1_joint": 0.980859100818634,
184
+ "left_ring_2_joint": 0.03758540377020836,
185
+ "left_thumb_1_joint": 1.4206803825800307e-05,
186
+ "left_thumb_2_joint": 0.07785109430551529,
187
+ "left_thumb_3_joint": 0.14139962196350098,
188
+ "left_thumb_4_joint": 0.5698095560073853,
189
+ "right_index_1_joint": 1.0677776336669922,
190
+ "right_index_2_joint": 0.13556255400180817,
191
+ "right_little_1_joint": 1.6511658430099487,
192
+ "right_little_2_joint": 0.530315637588501,
193
+ "right_middle_1_joint": 1.1244797706604004,
194
+ "right_middle_2_joint": 0.1444265991449356,
195
+ "right_ring_1_joint": 1.5612239837646484,
196
+ "right_ring_2_joint": 0.058094825595617294,
197
+ "right_thumb_1_joint": 0.13134731352329254,
198
+ "right_thumb_2_joint": 0.45885607600212097,
199
+ "right_thumb_3_joint": 0.26032817363739014,
200
+ "right_thumb_4_joint": 1.2807923555374146
201
+ },
202
+ "coverage": {
203
+ "hard_limit_source": {
204
+ "available": 25,
205
+ "unavailable": 0,
206
+ "unknown": 0
207
+ },
208
+ "joint_role_manifest": {
209
+ "available": 25,
210
+ "unavailable": 0,
211
+ "unknown": 0
212
+ }
213
+ },
214
+ "hard_limit_sources": {
215
+ "runtime.joint_pos_limits": 25
216
+ },
217
+ "soft_limit_sources": {
218
+ "runtime.soft_joint_pos_limits": 25
219
+ },
220
+ "unknown_runtime_joint_names": [],
221
+ "baseline_checks": {
222
+ "post_home": {
223
+ "checked": 25,
224
+ "hard_violation_episodes": 0
225
+ },
226
+ "post_reset": {
227
+ "checked": 25,
228
+ "hard_violation_episodes": 0
229
+ }
230
+ }
231
+ }
232
+ },
233
+ "robot_motion": {
234
+ "joint_acceleration_rms": 49.7102411448741,
235
+ "joint_effort_rms": 35.25884275139252,
236
+ "joint_jerk_rms": 5066.589718337765,
237
+ "joint_velocity_rms": 1.0171209289854322
238
+ },
239
+ "grasp": {
240
+ "kinematic_grasp_stability_index": 0.9191116836736373,
241
+ "mean_kinematic_grasp_stability": 0.4626109246846702,
242
+ "per_object_best_kinematic_grasp_stability": {
243
+ "obj_011_banana_2": 0.5078038782368343,
244
+ "obj_013_apple_3": 0.7692682883887582,
245
+ "obj_013_apple_4": 0.2182270372482906
246
+ },
247
+ "grasp_gsi_diagnostics": {
248
+ "mean_anti_slip_score": 0.9408106949176454,
249
+ "mean_duration_score": 0.1483243978861346,
250
+ "mean_hold_score": 0.13450883227066304,
251
+ "mean_kinematic_stability_score": 0.095777797236748,
252
+ "mean_legacy_gsi": 0.3018672918602893,
253
+ "mean_lift_score": 0.20325207116603772,
254
+ "mean_motion_stability_score": 0.1281106244979097
255
+ }
256
+ },
257
+ "tool": {
258
+ "tool_selection_accuracy": null,
259
+ "tool_switch_success_rate": null,
260
+ "tool_switch_total": 0,
261
+ "tool_switch_successful": 0
262
+ },
263
+ "runtime": {
264
+ "avg_steps": 1674.8,
265
+ "error_count": 0
266
+ }
267
+ }
268
+ }
mora_v0/per_task_eval/06_fruit_bowl_loading/Equi/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/06_fruit_bowl_loading/Equi/summary.json ADDED
@@ -0,0 +1,270 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2k",
5
+ "base_seed": 100600000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 50,
9
+ "successes": 17,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 1920,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.34,
50
+ "success_rate_ci": {
51
+ "mean": 0.34,
52
+ "ci_low": 0.22,
53
+ "ci_high": 0.48
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.725
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 22.784313725490183
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.3,
64
+ "hard_violation_rate": 0.6,
65
+ "drop_rate": 0.52,
66
+ "high_speed_violation_rate": 0.4,
67
+ "safety_violation_step_rate": 0.0008198614318706697
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.36,
74
+ "at_end_success_rate": 0.34,
75
+ "task_family_success_rate": {
76
+ "rigid_pick_place": 0.34
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.655,
81
+ "current_chain_depth_progress_score": 0.72,
82
+ "latched_chain_depth_progress_score": 0.7733333333333333,
83
+ "avg_current_chain_depth": 2.16,
84
+ "avg_latched_chain_depth": 2.32
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.937695406064917,
88
+ "expert_time_s": 21.333333333333332,
89
+ "expert_time_step": 1280,
90
+ "avg_steps_to_success": 1367.0588235294117,
91
+ "avg_policy_steps_to_success": 456,
92
+ "avg_policy_queries_to_success": 23.235294117647058
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.0006928406466512702,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_011_banana_2": 3,
98
+ "drop_to_table_obj_013_apple_3": 14,
99
+ "drop_to_table_obj_013_apple_4": 20,
100
+ "high_speed_obj_013_apple_3": 15,
101
+ "high_speed_obj_013_apple_4": 8
102
+ },
103
+ "robot_constraint_violation_rate": 1.0,
104
+ "finger_joint_limit_saturation_rate": 1.0,
105
+ "joint_limit_active_step_rate": 0.1418475750577367,
106
+ "finger_joint_limit_saturation_step_rate": 0.08585450346420323,
107
+ "max_joint_limit_excess_rad": 1.6511658430099487,
108
+ "robot_constraint_diagnostics": {
109
+ "joint_limit_soft_counts": {
110
+ "left_index_1_joint": 1040,
111
+ "left_index_2_joint": 46,
112
+ "left_little_1_joint": 1142,
113
+ "left_little_2_joint": 127,
114
+ "left_middle_1_joint": 1640,
115
+ "left_middle_2_joint": 100,
116
+ "left_ring_1_joint": 1356,
117
+ "left_ring_2_joint": 144,
118
+ "left_thumb_2_joint": 278,
119
+ "left_thumb_3_joint": 50,
120
+ "left_thumb_4_joint": 100,
121
+ "right_index_1_joint": 1859,
122
+ "right_index_2_joint": 29,
123
+ "right_little_1_joint": 1721,
124
+ "right_little_2_joint": 203,
125
+ "right_middle_1_joint": 2642,
126
+ "right_middle_2_joint": 40,
127
+ "right_ring_1_joint": 2226,
128
+ "right_ring_2_joint": 60,
129
+ "right_thumb_1_joint": 46,
130
+ "right_thumb_2_joint": 6555,
131
+ "right_thumb_3_joint": 64,
132
+ "right_thumb_4_joint": 237
133
+ },
134
+ "joint_limit_hard_counts": {
135
+ "left_index_1_joint": 45,
136
+ "left_index_2_joint": 41,
137
+ "left_little_1_joint": 35,
138
+ "left_little_2_joint": 53,
139
+ "left_middle_1_joint": 46,
140
+ "left_middle_2_joint": 86,
141
+ "left_ring_1_joint": 43,
142
+ "left_ring_2_joint": 79,
143
+ "left_thumb_2_joint": 11,
144
+ "left_thumb_3_joint": 3,
145
+ "left_thumb_4_joint": 5,
146
+ "right_index_1_joint": 83,
147
+ "right_index_2_joint": 25,
148
+ "right_little_1_joint": 55,
149
+ "right_little_2_joint": 16,
150
+ "right_middle_1_joint": 100,
151
+ "right_middle_2_joint": 34,
152
+ "right_ring_1_joint": 74,
153
+ "right_ring_2_joint": 32,
154
+ "right_thumb_1_joint": 5,
155
+ "right_thumb_2_joint": 81,
156
+ "right_thumb_3_joint": 6,
157
+ "right_thumb_4_joint": 19
158
+ },
159
+ "joint_limit_hard_episode_events": 250,
160
+ "finger_joint_limit_saturation_counts": {
161
+ "left_index_1_joint": 547,
162
+ "left_little_1_joint": 773,
163
+ "left_little_2_joint": 14,
164
+ "left_middle_1_joint": 1008,
165
+ "left_ring_1_joint": 785,
166
+ "left_thumb_2_joint": 104,
167
+ "left_thumb_3_joint": 20,
168
+ "left_thumb_4_joint": 55,
169
+ "right_index_1_joint": 1025,
170
+ "right_little_1_joint": 1326,
171
+ "right_little_2_joint": 97,
172
+ "right_middle_1_joint": 1530,
173
+ "right_ring_1_joint": 1483,
174
+ "right_thumb_2_joint": 3612,
175
+ "right_thumb_3_joint": 18,
176
+ "right_thumb_4_joint": 125
177
+ },
178
+ "joint_limit_max_excess_by_joint": {
179
+ "left_index_1_joint": 1.0526398420333862,
180
+ "left_index_2_joint": 0.10732898116111755,
181
+ "left_little_1_joint": 1.174699306488037,
182
+ "left_little_2_joint": 0.280036062002182,
183
+ "left_middle_1_joint": 1.2612115144729614,
184
+ "left_middle_2_joint": 0.038757774978876114,
185
+ "left_ring_1_joint": 1.131105899810791,
186
+ "left_ring_2_joint": 0.09605004638433456,
187
+ "left_thumb_1_joint": 1.4206803825800307e-05,
188
+ "left_thumb_2_joint": 0.4470149576663971,
189
+ "left_thumb_3_joint": 0.4703221917152405,
190
+ "left_thumb_4_joint": 0.8424203395843506,
191
+ "right_index_1_joint": 1.0677776336669922,
192
+ "right_index_2_joint": 0.13556255400180817,
193
+ "right_little_1_joint": 1.6511658430099487,
194
+ "right_little_2_joint": 0.530315637588501,
195
+ "right_middle_1_joint": 1.1832196712493896,
196
+ "right_middle_2_joint": 0.1444265991449356,
197
+ "right_ring_1_joint": 1.5612239837646484,
198
+ "right_ring_2_joint": 0.11619895696640015,
199
+ "right_thumb_1_joint": 0.15416760742664337,
200
+ "right_thumb_2_joint": 0.45885607600212097,
201
+ "right_thumb_3_joint": 0.40242546796798706,
202
+ "right_thumb_4_joint": 1.2807923555374146
203
+ },
204
+ "coverage": {
205
+ "hard_limit_source": {
206
+ "available": 50,
207
+ "unavailable": 0,
208
+ "unknown": 0
209
+ },
210
+ "joint_role_manifest": {
211
+ "available": 50,
212
+ "unavailable": 0,
213
+ "unknown": 0
214
+ }
215
+ },
216
+ "hard_limit_sources": {
217
+ "runtime.joint_pos_limits": 50
218
+ },
219
+ "soft_limit_sources": {
220
+ "runtime.soft_joint_pos_limits": 50
221
+ },
222
+ "unknown_runtime_joint_names": [],
223
+ "baseline_checks": {
224
+ "post_home": {
225
+ "checked": 50,
226
+ "hard_violation_episodes": 0
227
+ },
228
+ "post_reset": {
229
+ "checked": 50,
230
+ "hard_violation_episodes": 0
231
+ }
232
+ }
233
+ }
234
+ },
235
+ "robot_motion": {
236
+ "joint_acceleration_rms": 50.27430104524876,
237
+ "joint_effort_rms": 35.06979384699182,
238
+ "joint_jerk_rms": 5131.755628547,
239
+ "joint_velocity_rms": 1.0116509581186037
240
+ },
241
+ "grasp": {
242
+ "kinematic_grasp_stability_index": 0.9266090120648988,
243
+ "mean_kinematic_grasp_stability": 0.5026825876592923,
244
+ "per_object_best_kinematic_grasp_stability": {
245
+ "obj_011_banana_2": 0.4957395445411298,
246
+ "obj_013_apple_3": 0.8238371702275638,
247
+ "obj_013_apple_4": 0.2655641200842286
248
+ },
249
+ "grasp_gsi_diagnostics": {
250
+ "mean_anti_slip_score": 0.937839536543956,
251
+ "mean_duration_score": 0.16660966040045516,
252
+ "mean_hold_score": 0.1480590400572063,
253
+ "mean_kinematic_stability_score": 0.10948208069306738,
254
+ "mean_legacy_gsi": 0.3117433364749427,
255
+ "mean_lift_score": 0.2156456987387098,
256
+ "mean_motion_stability_score": 0.1392460194752599
257
+ }
258
+ },
259
+ "tool": {
260
+ "tool_selection_accuracy": null,
261
+ "tool_switch_success_rate": null,
262
+ "tool_switch_total": 0,
263
+ "tool_switch_successful": 0
264
+ },
265
+ "runtime": {
266
+ "avg_steps": 1732,
267
+ "error_count": 0
268
+ }
269
+ }
270
+ }
mora_v0/per_task_eval/06_fruit_bowl_loading/Full/native_shards/000/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/06_fruit_bowl_loading/Full/native_shards/000/summary.json ADDED
@@ -0,0 +1,264 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2k",
5
+ "base_seed": 200600000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 25,
9
+ "successes": 14,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 1920,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.56,
50
+ "success_rate_ci": {
51
+ "mean": 0.56,
52
+ "ci_low": 0.36,
53
+ "ci_high": 0.76
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.81
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 22.898809523809504
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.52,
64
+ "hard_violation_rate": 0.36,
65
+ "drop_rate": 0.32,
66
+ "high_speed_violation_rate": 0.24,
67
+ "safety_violation_step_rate": 0.000545161690001239
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.56,
74
+ "at_end_success_rate": 0.56,
75
+ "task_family_success_rate": {
76
+ "rigid_pick_place": 0.56
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.78,
81
+ "current_chain_depth_progress_score": 0.84,
82
+ "latched_chain_depth_progress_score": 0.84,
83
+ "avg_current_chain_depth": 2.52,
84
+ "avg_latched_chain_depth": 2.52
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.9322802198497361,
88
+ "expert_time_s": 21.333333333333332,
89
+ "expert_time_step": 1280,
90
+ "avg_steps_to_success": 1373.9285714285713,
91
+ "avg_policy_steps_to_success": 458.35714285714283,
92
+ "avg_policy_queries_to_success": 23.285714285714285
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.0004956015363647627,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_011_banana_2": 1,
98
+ "drop_to_table_obj_013_apple_3": 4,
99
+ "drop_to_table_obj_013_apple_4": 7,
100
+ "high_speed_obj_013_apple_3": 2,
101
+ "high_speed_obj_013_apple_4": 6
102
+ },
103
+ "robot_constraint_violation_rate": 1.0,
104
+ "finger_joint_limit_saturation_rate": 0.96,
105
+ "joint_limit_active_step_rate": 0.1378267872630405,
106
+ "finger_joint_limit_saturation_step_rate": 0.08507000371701152,
107
+ "max_joint_limit_excess_rad": 1.5180509090423584,
108
+ "robot_constraint_diagnostics": {
109
+ "joint_limit_soft_counts": {
110
+ "left_index_1_joint": 410,
111
+ "left_index_2_joint": 9,
112
+ "left_little_1_joint": 544,
113
+ "left_little_2_joint": 71,
114
+ "left_middle_1_joint": 630,
115
+ "left_middle_2_joint": 66,
116
+ "left_ring_1_joint": 645,
117
+ "left_ring_2_joint": 77,
118
+ "left_thumb_2_joint": 36,
119
+ "left_thumb_4_joint": 40,
120
+ "right_index_1_joint": 1013,
121
+ "right_index_2_joint": 24,
122
+ "right_little_1_joint": 934,
123
+ "right_little_2_joint": 84,
124
+ "right_middle_1_joint": 1535,
125
+ "right_middle_2_joint": 30,
126
+ "right_ring_1_joint": 1306,
127
+ "right_ring_2_joint": 51,
128
+ "right_thumb_1_joint": 49,
129
+ "right_thumb_2_joint": 2965,
130
+ "right_thumb_3_joint": 34,
131
+ "right_thumb_4_joint": 203
132
+ },
133
+ "joint_limit_hard_counts": {
134
+ "left_index_1_joint": 18,
135
+ "left_index_2_joint": 22,
136
+ "left_little_1_joint": 16,
137
+ "left_little_2_joint": 40,
138
+ "left_middle_1_joint": 24,
139
+ "left_middle_2_joint": 54,
140
+ "left_ring_1_joint": 18,
141
+ "left_ring_2_joint": 66,
142
+ "left_thumb_2_joint": 2,
143
+ "left_thumb_4_joint": 1,
144
+ "right_index_1_joint": 39,
145
+ "right_index_2_joint": 22,
146
+ "right_little_1_joint": 31,
147
+ "right_little_2_joint": 7,
148
+ "right_middle_1_joint": 41,
149
+ "right_middle_2_joint": 18,
150
+ "right_ring_1_joint": 40,
151
+ "right_ring_2_joint": 10,
152
+ "right_thumb_1_joint": 3,
153
+ "right_thumb_2_joint": 42,
154
+ "right_thumb_3_joint": 4,
155
+ "right_thumb_4_joint": 15
156
+ },
157
+ "joint_limit_hard_episode_events": 105,
158
+ "finger_joint_limit_saturation_counts": {
159
+ "left_index_1_joint": 220,
160
+ "left_little_1_joint": 356,
161
+ "left_middle_1_joint": 339,
162
+ "left_ring_1_joint": 392,
163
+ "left_thumb_4_joint": 23,
164
+ "right_index_1_joint": 638,
165
+ "right_little_1_joint": 811,
166
+ "right_little_2_joint": 36,
167
+ "right_middle_1_joint": 1024,
168
+ "right_ring_1_joint": 835,
169
+ "right_thumb_1_joint": 12,
170
+ "right_thumb_2_joint": 1530,
171
+ "right_thumb_4_joint": 147
172
+ },
173
+ "joint_limit_max_excess_by_joint": {
174
+ "left_index_1_joint": 0.9070860743522644,
175
+ "left_index_2_joint": 0.02048588916659355,
176
+ "left_little_1_joint": 1.042426586151123,
177
+ "left_little_2_joint": 0.10967608541250229,
178
+ "left_middle_1_joint": 1.1422985792160034,
179
+ "left_middle_2_joint": 0.05379383638501167,
180
+ "left_ring_1_joint": 0.8085528612136841,
181
+ "left_ring_2_joint": 0.05745309963822365,
182
+ "left_thumb_2_joint": 0.024611053988337517,
183
+ "left_thumb_3_joint": 1.6463045540149324e-05,
184
+ "left_thumb_4_joint": 0.2448093593120575,
185
+ "right_index_1_joint": 0.8882948160171509,
186
+ "right_index_2_joint": 0.04815778136253357,
187
+ "right_little_1_joint": 1.5180509090423584,
188
+ "right_little_2_joint": 0.3379167318344116,
189
+ "right_middle_1_joint": 1.2586828470230103,
190
+ "right_middle_2_joint": 0.05448639765381813,
191
+ "right_ring_1_joint": 1.0764539241790771,
192
+ "right_ring_2_joint": 0.10288409888744354,
193
+ "right_thumb_1_joint": 0.17422093451023102,
194
+ "right_thumb_2_joint": 0.43493086099624634,
195
+ "right_thumb_3_joint": 0.12095694988965988,
196
+ "right_thumb_4_joint": 1.0856343507766724
197
+ },
198
+ "coverage": {
199
+ "hard_limit_source": {
200
+ "available": 25,
201
+ "unavailable": 0,
202
+ "unknown": 0
203
+ },
204
+ "joint_role_manifest": {
205
+ "available": 25,
206
+ "unavailable": 0,
207
+ "unknown": 0
208
+ }
209
+ },
210
+ "hard_limit_sources": {
211
+ "runtime.joint_pos_limits": 25
212
+ },
213
+ "soft_limit_sources": {
214
+ "runtime.soft_joint_pos_limits": 25
215
+ },
216
+ "unknown_runtime_joint_names": [],
217
+ "baseline_checks": {
218
+ "post_home": {
219
+ "checked": 25,
220
+ "hard_violation_episodes": 0
221
+ },
222
+ "post_reset": {
223
+ "checked": 25,
224
+ "hard_violation_episodes": 0
225
+ }
226
+ }
227
+ }
228
+ },
229
+ "robot_motion": {
230
+ "joint_acceleration_rms": 51.65924788465825,
231
+ "joint_effort_rms": 36.68149282461523,
232
+ "joint_jerk_rms": 5270.7563664429745,
233
+ "joint_velocity_rms": 1.0603761545736994
234
+ },
235
+ "grasp": {
236
+ "kinematic_grasp_stability_index": 0.8639595697590946,
237
+ "mean_kinematic_grasp_stability": 0.3881919838925285,
238
+ "per_object_best_kinematic_grasp_stability": {
239
+ "obj_011_banana_2": 0.49019570247181576,
240
+ "obj_013_apple_3": 0.7959182256531453,
241
+ "obj_013_apple_4": 0.2508246115878469
242
+ },
243
+ "grasp_gsi_diagnostics": {
244
+ "mean_anti_slip_score": 0.9157243119757378,
245
+ "mean_duration_score": 0.1406257896859748,
246
+ "mean_hold_score": 0.12090395032830215,
247
+ "mean_kinematic_stability_score": 0.08145982913929092,
248
+ "mean_legacy_gsi": 0.2873174165966608,
249
+ "mean_lift_score": 0.1935216084320529,
250
+ "mean_motion_stability_score": 0.11363898629031277
251
+ }
252
+ },
253
+ "tool": {
254
+ "tool_selection_accuracy": null,
255
+ "tool_switch_success_rate": null,
256
+ "tool_switch_total": 0,
257
+ "tool_switch_successful": 0
258
+ },
259
+ "runtime": {
260
+ "avg_steps": 1614.2,
261
+ "error_count": 0
262
+ }
263
+ }
264
+ }
mora_v0/per_task_eval/06_fruit_bowl_loading/Full/native_shards/001/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/06_fruit_bowl_loading/Full/native_shards/001/summary.json ADDED
@@ -0,0 +1,259 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2k",
5
+ "base_seed": 200600000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 25,
9
+ "successes": 17,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 1920,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.68,
50
+ "success_rate_ci": {
51
+ "mean": 0.68,
52
+ "ci_low": 0.48,
53
+ "ci_high": 0.84
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.84
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 22.736274509803913
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.52,
64
+ "hard_violation_rate": 0.4,
65
+ "drop_rate": 0.28,
66
+ "high_speed_violation_rate": 0.32,
67
+ "safety_violation_step_rate": 0.0005447329511556121
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.68,
74
+ "at_end_success_rate": 0.68,
75
+ "task_family_success_rate": {
76
+ "rigid_pick_place": 0.68
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.79,
81
+ "current_chain_depth_progress_score": 0.8133333333333334,
82
+ "latched_chain_depth_progress_score": 0.8533333333333333,
83
+ "avg_current_chain_depth": 2.44,
84
+ "avg_latched_chain_depth": 2.56
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.9393577217699899,
88
+ "expert_time_s": 21.333333333333332,
89
+ "expert_time_step": 1280,
90
+ "avg_steps_to_success": 1364.1764705882354,
91
+ "avg_policy_steps_to_success": 455.05882352941177,
92
+ "avg_policy_queries_to_success": 23.294117647058822
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.0005187932868148687,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_011_banana_2": 2,
98
+ "drop_to_table_obj_013_apple_3": 2,
99
+ "drop_to_table_obj_013_apple_4": 7,
100
+ "high_speed_obj_013_apple_3": 4,
101
+ "high_speed_obj_013_apple_4": 5
102
+ },
103
+ "robot_constraint_violation_rate": 1.0,
104
+ "finger_joint_limit_saturation_rate": 0.92,
105
+ "joint_limit_active_step_rate": 0.1512282431065342,
106
+ "finger_joint_limit_saturation_step_rate": 0.09405722289953568,
107
+ "max_joint_limit_excess_rad": 1.4776114225387573,
108
+ "robot_constraint_diagnostics": {
109
+ "joint_limit_soft_counts": {
110
+ "left_index_1_joint": 504,
111
+ "left_index_2_joint": 22,
112
+ "left_little_1_joint": 747,
113
+ "left_little_2_joint": 178,
114
+ "left_middle_1_joint": 726,
115
+ "left_middle_2_joint": 46,
116
+ "left_ring_1_joint": 765,
117
+ "left_ring_2_joint": 88,
118
+ "left_thumb_2_joint": 55,
119
+ "right_index_1_joint": 1019,
120
+ "right_index_2_joint": 26,
121
+ "right_little_1_joint": 412,
122
+ "right_little_2_joint": 53,
123
+ "right_middle_1_joint": 1247,
124
+ "right_middle_2_joint": 9,
125
+ "right_ring_1_joint": 830,
126
+ "right_ring_2_joint": 9,
127
+ "right_thumb_1_joint": 32,
128
+ "right_thumb_2_joint": 3378,
129
+ "right_thumb_4_joint": 73
130
+ },
131
+ "joint_limit_hard_counts": {
132
+ "left_index_1_joint": 21,
133
+ "left_index_2_joint": 44,
134
+ "left_little_1_joint": 17,
135
+ "left_little_2_joint": 74,
136
+ "left_middle_1_joint": 23,
137
+ "left_middle_2_joint": 83,
138
+ "left_ring_1_joint": 18,
139
+ "left_ring_2_joint": 85,
140
+ "left_thumb_2_joint": 3,
141
+ "right_index_1_joint": 33,
142
+ "right_index_2_joint": 22,
143
+ "right_little_1_joint": 16,
144
+ "right_little_2_joint": 5,
145
+ "right_middle_1_joint": 40,
146
+ "right_middle_2_joint": 6,
147
+ "right_ring_1_joint": 30,
148
+ "right_ring_2_joint": 4,
149
+ "right_thumb_1_joint": 2,
150
+ "right_thumb_2_joint": 49,
151
+ "right_thumb_4_joint": 3
152
+ },
153
+ "joint_limit_hard_episode_events": 118,
154
+ "finger_joint_limit_saturation_counts": {
155
+ "left_index_1_joint": 218,
156
+ "left_little_1_joint": 546,
157
+ "left_little_2_joint": 12,
158
+ "left_middle_1_joint": 453,
159
+ "left_ring_1_joint": 564,
160
+ "right_index_1_joint": 649,
161
+ "right_little_1_joint": 365,
162
+ "right_little_2_joint": 10,
163
+ "right_middle_1_joint": 880,
164
+ "right_ring_1_joint": 553,
165
+ "right_thumb_2_joint": 1846,
166
+ "right_thumb_4_joint": 46
167
+ },
168
+ "joint_limit_max_excess_by_joint": {
169
+ "left_index_1_joint": 0.7977405786514282,
170
+ "left_index_2_joint": 0.03154894337058067,
171
+ "left_little_1_joint": 1.1651893854141235,
172
+ "left_little_2_joint": 0.16055350005626678,
173
+ "left_middle_1_joint": 0.8965175747871399,
174
+ "left_middle_2_joint": 0.06627637892961502,
175
+ "left_ring_1_joint": 1.1758384704589844,
176
+ "left_ring_2_joint": 0.06529927998781204,
177
+ "left_thumb_2_joint": 0.07512982189655304,
178
+ "left_thumb_3_joint": 0.0007201156113296747,
179
+ "left_thumb_4_joint": 0.0008341372013092041,
180
+ "right_index_1_joint": 0.954833984375,
181
+ "right_index_2_joint": 0.06537564843893051,
182
+ "right_little_1_joint": 1.4776114225387573,
183
+ "right_little_2_joint": 0.28001129627227783,
184
+ "right_middle_1_joint": 1.1718026399612427,
185
+ "right_middle_2_joint": 0.05367403104901314,
186
+ "right_ring_1_joint": 1.303239345550537,
187
+ "right_ring_2_joint": 0.07121134549379349,
188
+ "right_thumb_1_joint": 0.1457221806049347,
189
+ "right_thumb_2_joint": 0.5781301856040955,
190
+ "right_thumb_3_joint": 0.001836868585087359,
191
+ "right_thumb_4_joint": 0.9554686546325684
192
+ },
193
+ "coverage": {
194
+ "hard_limit_source": {
195
+ "available": 25,
196
+ "unavailable": 0,
197
+ "unknown": 0
198
+ },
199
+ "joint_role_manifest": {
200
+ "available": 25,
201
+ "unavailable": 0,
202
+ "unknown": 0
203
+ }
204
+ },
205
+ "hard_limit_sources": {
206
+ "runtime.joint_pos_limits": 25
207
+ },
208
+ "soft_limit_sources": {
209
+ "runtime.soft_joint_pos_limits": 25
210
+ },
211
+ "unknown_runtime_joint_names": [],
212
+ "baseline_checks": {
213
+ "post_home": {
214
+ "checked": 25,
215
+ "hard_violation_episodes": 0
216
+ },
217
+ "post_reset": {
218
+ "checked": 25,
219
+ "hard_violation_episodes": 0
220
+ }
221
+ }
222
+ }
223
+ },
224
+ "robot_motion": {
225
+ "joint_acceleration_rms": 54.08777076822663,
226
+ "joint_effort_rms": 36.989707260962085,
227
+ "joint_jerk_rms": 5518.08697204552,
228
+ "joint_velocity_rms": 1.0934302868902577
229
+ },
230
+ "grasp": {
231
+ "kinematic_grasp_stability_index": 0.8712518781975677,
232
+ "mean_kinematic_grasp_stability": 0.39287186183174133,
233
+ "per_object_best_kinematic_grasp_stability": {
234
+ "obj_011_banana_2": 0.567250866710615,
235
+ "obj_013_apple_3": 0.8349299057223357,
236
+ "obj_013_apple_4": 0.22999477550696135
237
+ },
238
+ "grasp_gsi_diagnostics": {
239
+ "mean_anti_slip_score": 0.9360174843516146,
240
+ "mean_duration_score": 0.12728174294825076,
241
+ "mean_hold_score": 0.12168500239868697,
242
+ "mean_kinematic_stability_score": 0.08228473834442745,
243
+ "mean_legacy_gsi": 0.2897310970493175,
244
+ "mean_lift_score": 0.19680355552780082,
245
+ "mean_motion_stability_score": 0.11509703936851659
246
+ }
247
+ },
248
+ "tool": {
249
+ "tool_selection_accuracy": null,
250
+ "tool_switch_success_rate": null,
251
+ "tool_switch_total": 0,
252
+ "tool_switch_successful": 0
253
+ },
254
+ "runtime": {
255
+ "avg_steps": 1542.04,
256
+ "error_count": 0
257
+ }
258
+ }
259
+ }
mora_v0/per_task_eval/06_fruit_bowl_loading/Full/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/06_fruit_bowl_loading/Full/summary.json ADDED
@@ -0,0 +1,265 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2k",
5
+ "base_seed": 200600000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 50,
9
+ "successes": 31,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 1920,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.62,
50
+ "success_rate_ci": {
51
+ "mean": 0.62,
52
+ "ci_low": 0.48,
53
+ "ci_high": 0.76
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.825
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 22.809677419354824
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.52,
64
+ "hard_violation_rate": 0.38,
65
+ "drop_rate": 0.3,
66
+ "high_speed_violation_rate": 0.28,
67
+ "safety_violation_step_rate": 0.0005449522216308012
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.62,
74
+ "at_end_success_rate": 0.62,
75
+ "task_family_success_rate": {
76
+ "rigid_pick_place": 0.62
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.785,
81
+ "current_chain_depth_progress_score": 0.8266666666666667,
82
+ "latched_chain_depth_progress_score": 0.8466666666666667,
83
+ "avg_current_chain_depth": 2.48,
84
+ "avg_latched_chain_depth": 2.54
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.9361614305801979,
88
+ "expert_time_s": 21.333333333333332,
89
+ "expert_time_step": 1280,
90
+ "avg_steps_to_success": 1368.5806451612902,
91
+ "avg_policy_steps_to_success": 456.5483870967742,
92
+ "avg_policy_queries_to_success": 23.29032258064516
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.000506932299191443,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_011_banana_2": 3,
98
+ "drop_to_table_obj_013_apple_3": 6,
99
+ "drop_to_table_obj_013_apple_4": 14,
100
+ "high_speed_obj_013_apple_3": 6,
101
+ "high_speed_obj_013_apple_4": 11
102
+ },
103
+ "robot_constraint_violation_rate": 1.0,
104
+ "finger_joint_limit_saturation_rate": 0.94,
105
+ "joint_limit_active_step_rate": 0.14437431880972296,
106
+ "finger_joint_limit_saturation_step_rate": 0.0894608774998099,
107
+ "max_joint_limit_excess_rad": 1.5180509090423584,
108
+ "robot_constraint_diagnostics": {
109
+ "joint_limit_soft_counts": {
110
+ "left_index_1_joint": 914,
111
+ "left_index_2_joint": 31,
112
+ "left_little_1_joint": 1291,
113
+ "left_little_2_joint": 249,
114
+ "left_middle_1_joint": 1356,
115
+ "left_middle_2_joint": 112,
116
+ "left_ring_1_joint": 1410,
117
+ "left_ring_2_joint": 165,
118
+ "left_thumb_2_joint": 91,
119
+ "left_thumb_4_joint": 40,
120
+ "right_index_1_joint": 2032,
121
+ "right_index_2_joint": 50,
122
+ "right_little_1_joint": 1346,
123
+ "right_little_2_joint": 137,
124
+ "right_middle_1_joint": 2782,
125
+ "right_middle_2_joint": 39,
126
+ "right_ring_1_joint": 2136,
127
+ "right_ring_2_joint": 60,
128
+ "right_thumb_1_joint": 81,
129
+ "right_thumb_2_joint": 6343,
130
+ "right_thumb_3_joint": 34,
131
+ "right_thumb_4_joint": 276
132
+ },
133
+ "joint_limit_hard_counts": {
134
+ "left_index_1_joint": 39,
135
+ "left_index_2_joint": 66,
136
+ "left_little_1_joint": 33,
137
+ "left_little_2_joint": 114,
138
+ "left_middle_1_joint": 47,
139
+ "left_middle_2_joint": 137,
140
+ "left_ring_1_joint": 36,
141
+ "left_ring_2_joint": 151,
142
+ "left_thumb_2_joint": 5,
143
+ "left_thumb_4_joint": 1,
144
+ "right_index_1_joint": 72,
145
+ "right_index_2_joint": 44,
146
+ "right_little_1_joint": 47,
147
+ "right_little_2_joint": 12,
148
+ "right_middle_1_joint": 81,
149
+ "right_middle_2_joint": 24,
150
+ "right_ring_1_joint": 70,
151
+ "right_ring_2_joint": 14,
152
+ "right_thumb_1_joint": 5,
153
+ "right_thumb_2_joint": 91,
154
+ "right_thumb_3_joint": 4,
155
+ "right_thumb_4_joint": 18
156
+ },
157
+ "joint_limit_hard_episode_events": 223,
158
+ "finger_joint_limit_saturation_counts": {
159
+ "left_index_1_joint": 438,
160
+ "left_little_1_joint": 902,
161
+ "left_little_2_joint": 12,
162
+ "left_middle_1_joint": 792,
163
+ "left_ring_1_joint": 956,
164
+ "left_thumb_4_joint": 23,
165
+ "right_index_1_joint": 1287,
166
+ "right_little_1_joint": 1176,
167
+ "right_little_2_joint": 46,
168
+ "right_middle_1_joint": 1904,
169
+ "right_ring_1_joint": 1388,
170
+ "right_thumb_1_joint": 12,
171
+ "right_thumb_2_joint": 3376,
172
+ "right_thumb_4_joint": 193
173
+ },
174
+ "joint_limit_max_excess_by_joint": {
175
+ "left_index_1_joint": 0.9070860743522644,
176
+ "left_index_2_joint": 0.03154894337058067,
177
+ "left_little_1_joint": 1.1651893854141235,
178
+ "left_little_2_joint": 0.16055350005626678,
179
+ "left_middle_1_joint": 1.1422985792160034,
180
+ "left_middle_2_joint": 0.06627637892961502,
181
+ "left_ring_1_joint": 1.1758384704589844,
182
+ "left_ring_2_joint": 0.06529927998781204,
183
+ "left_thumb_2_joint": 0.07512982189655304,
184
+ "left_thumb_3_joint": 0.0007201156113296747,
185
+ "left_thumb_4_joint": 0.2448093593120575,
186
+ "right_index_1_joint": 0.954833984375,
187
+ "right_index_2_joint": 0.06537564843893051,
188
+ "right_little_1_joint": 1.5180509090423584,
189
+ "right_little_2_joint": 0.3379167318344116,
190
+ "right_middle_1_joint": 1.2586828470230103,
191
+ "right_middle_2_joint": 0.05448639765381813,
192
+ "right_ring_1_joint": 1.303239345550537,
193
+ "right_ring_2_joint": 0.10288409888744354,
194
+ "right_thumb_1_joint": 0.17422093451023102,
195
+ "right_thumb_2_joint": 0.5781301856040955,
196
+ "right_thumb_3_joint": 0.12095694988965988,
197
+ "right_thumb_4_joint": 1.0856343507766724
198
+ },
199
+ "coverage": {
200
+ "hard_limit_source": {
201
+ "available": 50,
202
+ "unavailable": 0,
203
+ "unknown": 0
204
+ },
205
+ "joint_role_manifest": {
206
+ "available": 50,
207
+ "unavailable": 0,
208
+ "unknown": 0
209
+ }
210
+ },
211
+ "hard_limit_sources": {
212
+ "runtime.joint_pos_limits": 50
213
+ },
214
+ "soft_limit_sources": {
215
+ "runtime.soft_joint_pos_limits": 50
216
+ },
217
+ "unknown_runtime_joint_names": [],
218
+ "baseline_checks": {
219
+ "post_home": {
220
+ "checked": 50,
221
+ "hard_violation_episodes": 0
222
+ },
223
+ "post_reset": {
224
+ "checked": 50,
225
+ "hard_violation_episodes": 0
226
+ }
227
+ }
228
+ }
229
+ },
230
+ "robot_motion": {
231
+ "joint_acceleration_rms": 52.87350932644244,
232
+ "joint_effort_rms": 36.835600042788656,
233
+ "joint_jerk_rms": 5394.421669244247,
234
+ "joint_velocity_rms": 1.0769032207319786
235
+ },
236
+ "grasp": {
237
+ "kinematic_grasp_stability_index": 0.8676057239783311,
238
+ "mean_kinematic_grasp_stability": 0.39053192286213495,
239
+ "per_object_best_kinematic_grasp_stability": {
240
+ "obj_011_banana_2": 0.5287232845912153,
241
+ "obj_013_apple_3": 0.8154240656877405,
242
+ "obj_013_apple_4": 0.24040969354740413
243
+ },
244
+ "grasp_gsi_diagnostics": {
245
+ "mean_anti_slip_score": 0.9258708981636762,
246
+ "mean_duration_score": 0.13395376631711278,
247
+ "mean_hold_score": 0.12129447636349455,
248
+ "mean_kinematic_stability_score": 0.08187228374185919,
249
+ "mean_legacy_gsi": 0.28852425682298916,
250
+ "mean_lift_score": 0.19516258197992686,
251
+ "mean_motion_stability_score": 0.11436801282941468
252
+ }
253
+ },
254
+ "tool": {
255
+ "tool_selection_accuracy": null,
256
+ "tool_switch_success_rate": null,
257
+ "tool_switch_total": 0,
258
+ "tool_switch_successful": 0
259
+ },
260
+ "runtime": {
261
+ "avg_steps": 1578.12,
262
+ "error_count": 0
263
+ }
264
+ }
265
+ }
mora_v0/per_task_eval/06_fruit_bowl_loading/Inv/native_shards/000/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/06_fruit_bowl_loading/Inv/native_shards/000/summary.json ADDED
@@ -0,0 +1,268 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2k",
5
+ "base_seed": 100600000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 25,
9
+ "successes": 11,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 1920,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.44,
50
+ "success_rate_ci": {
51
+ "mean": 0.44,
52
+ "ci_low": 0.24,
53
+ "ci_high": 0.64
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.76
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 23.27878787878784
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.4,
64
+ "hard_violation_rate": 0.52,
65
+ "drop_rate": 0.48,
66
+ "high_speed_violation_rate": 0.32,
67
+ "safety_violation_step_rate": 0.0006154720196951046
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.44,
74
+ "at_end_success_rate": 0.44,
75
+ "task_family_success_rate": {
76
+ "rigid_pick_place": 0.44
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.68,
81
+ "current_chain_depth_progress_score": 0.7466666666666667,
82
+ "latched_chain_depth_progress_score": 0.8133333333333334,
83
+ "avg_current_chain_depth": 2.24,
84
+ "avg_latched_chain_depth": 2.44
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.9213150671398289,
88
+ "expert_time_s": 21.333333333333332,
89
+ "expert_time_step": 1280,
90
+ "avg_steps_to_success": 1396.7272727272727,
91
+ "avg_policy_steps_to_success": 465.90909090909093,
92
+ "avg_policy_queries_to_success": 23.818181818181817
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.0005681280181800966,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_013_apple_3": 7,
98
+ "drop_to_table_obj_013_apple_4": 9,
99
+ "high_speed_obj_013_apple_3": 4,
100
+ "high_speed_obj_013_apple_4": 4
101
+ },
102
+ "robot_constraint_violation_rate": 1.0,
103
+ "finger_joint_limit_saturation_rate": 1.0,
104
+ "joint_limit_active_step_rate": 0.15135877284348073,
105
+ "finger_joint_limit_saturation_step_rate": 0.08706561878609981,
106
+ "max_joint_limit_excess_rad": 1.7565630674362183,
107
+ "robot_constraint_diagnostics": {
108
+ "joint_limit_soft_counts": {
109
+ "left_index_1_joint": 577,
110
+ "left_index_2_joint": 18,
111
+ "left_little_1_joint": 610,
112
+ "left_little_2_joint": 153,
113
+ "left_middle_1_joint": 814,
114
+ "left_middle_2_joint": 86,
115
+ "left_ring_1_joint": 756,
116
+ "left_ring_2_joint": 99,
117
+ "left_thumb_2_joint": 55,
118
+ "left_thumb_3_joint": 24,
119
+ "left_thumb_4_joint": 3,
120
+ "right_index_1_joint": 970,
121
+ "right_index_2_joint": 49,
122
+ "right_little_1_joint": 1033,
123
+ "right_little_2_joint": 249,
124
+ "right_middle_1_joint": 1388,
125
+ "right_middle_2_joint": 25,
126
+ "right_ring_1_joint": 1113,
127
+ "right_ring_2_joint": 10,
128
+ "right_thumb_1_joint": 61,
129
+ "right_thumb_2_joint": 3817,
130
+ "right_thumb_3_joint": 71,
131
+ "right_thumb_4_joint": 236
132
+ },
133
+ "joint_limit_hard_counts": {
134
+ "left_index_1_joint": 31,
135
+ "left_index_2_joint": 29,
136
+ "left_little_1_joint": 19,
137
+ "left_little_2_joint": 60,
138
+ "left_middle_1_joint": 25,
139
+ "left_middle_2_joint": 67,
140
+ "left_ring_1_joint": 22,
141
+ "left_ring_2_joint": 55,
142
+ "left_thumb_2_joint": 3,
143
+ "left_thumb_3_joint": 1,
144
+ "left_thumb_4_joint": 1,
145
+ "right_index_1_joint": 40,
146
+ "right_index_2_joint": 28,
147
+ "right_little_1_joint": 38,
148
+ "right_little_2_joint": 11,
149
+ "right_middle_1_joint": 46,
150
+ "right_middle_2_joint": 24,
151
+ "right_ring_1_joint": 46,
152
+ "right_ring_2_joint": 2,
153
+ "right_thumb_1_joint": 4,
154
+ "right_thumb_2_joint": 46,
155
+ "right_thumb_3_joint": 6,
156
+ "right_thumb_4_joint": 15
157
+ },
158
+ "joint_limit_hard_episode_events": 132,
159
+ "finger_joint_limit_saturation_counts": {
160
+ "left_index_1_joint": 214,
161
+ "left_little_1_joint": 368,
162
+ "left_little_2_joint": 45,
163
+ "left_middle_1_joint": 445,
164
+ "left_ring_1_joint": 484,
165
+ "left_ring_2_joint": 24,
166
+ "left_thumb_2_joint": 16,
167
+ "left_thumb_3_joint": 20,
168
+ "right_index_1_joint": 484,
169
+ "right_little_1_joint": 842,
170
+ "right_little_2_joint": 175,
171
+ "right_middle_1_joint": 844,
172
+ "right_ring_1_joint": 836,
173
+ "right_thumb_2_joint": 1922,
174
+ "right_thumb_3_joint": 25,
175
+ "right_thumb_4_joint": 168
176
+ },
177
+ "joint_limit_max_excess_by_joint": {
178
+ "left_index_1_joint": 0.6131885647773743,
179
+ "left_index_2_joint": 0.03145715966820717,
180
+ "left_little_1_joint": 0.7894626259803772,
181
+ "left_little_2_joint": 0.42039117217063904,
182
+ "left_middle_1_joint": 0.8088790774345398,
183
+ "left_middle_2_joint": 0.050838202238082886,
184
+ "left_ring_1_joint": 1.4075347185134888,
185
+ "left_ring_2_joint": 0.3474140167236328,
186
+ "left_thumb_2_joint": 0.2787731885910034,
187
+ "left_thumb_3_joint": 0.579958975315094,
188
+ "left_thumb_4_joint": 0.06642940640449524,
189
+ "right_index_1_joint": 1.0716137886047363,
190
+ "right_index_2_joint": 0.07298866659402847,
191
+ "right_little_1_joint": 1.7565630674362183,
192
+ "right_little_2_joint": 1.098952054977417,
193
+ "right_middle_1_joint": 1.3280658721923828,
194
+ "right_middle_2_joint": 0.045254215598106384,
195
+ "right_ring_1_joint": 1.434998631477356,
196
+ "right_ring_2_joint": 0.034781042486429214,
197
+ "right_thumb_1_joint": 0.20933566987514496,
198
+ "right_thumb_2_joint": 0.5103276371955872,
199
+ "right_thumb_3_joint": 0.6045276522636414,
200
+ "right_thumb_4_joint": 1.3936662673950195
201
+ },
202
+ "coverage": {
203
+ "hard_limit_source": {
204
+ "available": 25,
205
+ "unavailable": 0,
206
+ "unknown": 0
207
+ },
208
+ "joint_role_manifest": {
209
+ "available": 25,
210
+ "unavailable": 0,
211
+ "unknown": 0
212
+ }
213
+ },
214
+ "hard_limit_sources": {
215
+ "runtime.joint_pos_limits": 25
216
+ },
217
+ "soft_limit_sources": {
218
+ "runtime.soft_joint_pos_limits": 25
219
+ },
220
+ "unknown_runtime_joint_names": [],
221
+ "baseline_checks": {
222
+ "post_home": {
223
+ "checked": 25,
224
+ "hard_violation_episodes": 0
225
+ },
226
+ "post_reset": {
227
+ "checked": 25,
228
+ "hard_violation_episodes": 0
229
+ }
230
+ }
231
+ }
232
+ },
233
+ "robot_motion": {
234
+ "joint_acceleration_rms": 51.06955444833674,
235
+ "joint_effort_rms": 36.16924271538786,
236
+ "joint_jerk_rms": 5215.011400975907,
237
+ "joint_velocity_rms": 1.0476902259562038
238
+ },
239
+ "grasp": {
240
+ "kinematic_grasp_stability_index": 0.8598503705834016,
241
+ "mean_kinematic_grasp_stability": 0.4209589510741862,
242
+ "per_object_best_kinematic_grasp_stability": {
243
+ "obj_011_banana_2": 0.4770026187157235,
244
+ "obj_013_apple_3": 0.7955182196413371,
245
+ "obj_013_apple_4": 0.2726134009800061
246
+ },
247
+ "grasp_gsi_diagnostics": {
248
+ "mean_anti_slip_score": 0.8939565156367019,
249
+ "mean_duration_score": 0.16255846370981047,
250
+ "mean_hold_score": 0.14127588616212525,
251
+ "mean_kinematic_stability_score": 0.08999711035345404,
252
+ "mean_legacy_gsi": 0.2957930662917447,
253
+ "mean_lift_score": 0.20866260542767665,
254
+ "mean_motion_stability_score": 0.12701360206464957
255
+ }
256
+ },
257
+ "tool": {
258
+ "tool_selection_accuracy": null,
259
+ "tool_switch_success_rate": null,
260
+ "tool_switch_total": 0,
261
+ "tool_switch_successful": 0
262
+ },
263
+ "runtime": {
264
+ "avg_steps": 1689.76,
265
+ "error_count": 0
266
+ }
267
+ }
268
+ }
mora_v0/per_task_eval/06_fruit_bowl_loading/Inv/native_shards/001/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/06_fruit_bowl_loading/Inv/native_shards/001/summary.json ADDED
@@ -0,0 +1,255 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2k",
5
+ "base_seed": 100600000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 25,
9
+ "successes": 20,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 1920,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.8,
50
+ "success_rate_ci": {
51
+ "mean": 0.8,
52
+ "ci_low": 0.64,
53
+ "ci_high": 0.96
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.91
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 22.604999999999997
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.76,
64
+ "hard_violation_rate": 0.2,
65
+ "drop_rate": 0.16,
66
+ "high_speed_violation_rate": 0.16,
67
+ "safety_violation_step_rate": 0.0002722866633992267
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.8,
74
+ "at_end_success_rate": 0.8,
75
+ "task_family_success_rate": {
76
+ "rigid_pick_place": 0.8
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.87,
81
+ "current_chain_depth_progress_score": 0.88,
82
+ "latched_chain_depth_progress_score": 0.92,
83
+ "avg_current_chain_depth": 2.64,
84
+ "avg_latched_chain_depth": 2.76
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.9448487564091574,
88
+ "expert_time_s": 21.333333333333332,
89
+ "expert_time_step": 1280,
90
+ "avg_steps_to_success": 1356.3,
91
+ "avg_policy_steps_to_success": 452.45,
92
+ "avg_policy_queries_to_success": 23.05
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.0002722866633992267,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_013_apple_3": 1,
98
+ "drop_to_table_obj_013_apple_4": 4,
99
+ "high_speed_obj_013_apple_3": 2,
100
+ "high_speed_obj_013_apple_4": 3
101
+ },
102
+ "robot_constraint_violation_rate": 1.0,
103
+ "finger_joint_limit_saturation_rate": 0.96,
104
+ "joint_limit_active_step_rate": 0.13246746174372379,
105
+ "finger_joint_limit_saturation_step_rate": 0.06760877852202798,
106
+ "max_joint_limit_excess_rad": 1.1197097301483154,
107
+ "robot_constraint_diagnostics": {
108
+ "joint_limit_soft_counts": {
109
+ "left_index_1_joint": 743,
110
+ "left_index_2_joint": 38,
111
+ "left_little_1_joint": 953,
112
+ "left_little_2_joint": 202,
113
+ "left_middle_1_joint": 957,
114
+ "left_middle_2_joint": 80,
115
+ "left_ring_1_joint": 962,
116
+ "left_ring_2_joint": 189,
117
+ "left_thumb_2_joint": 79,
118
+ "right_index_1_joint": 381,
119
+ "right_index_2_joint": 2,
120
+ "right_little_1_joint": 225,
121
+ "right_little_2_joint": 31,
122
+ "right_middle_1_joint": 574,
123
+ "right_middle_2_joint": 2,
124
+ "right_ring_1_joint": 343,
125
+ "right_thumb_2_joint": 2794,
126
+ "right_thumb_4_joint": 28
127
+ },
128
+ "joint_limit_hard_counts": {
129
+ "left_index_1_joint": 43,
130
+ "left_index_2_joint": 50,
131
+ "left_little_1_joint": 23,
132
+ "left_little_2_joint": 67,
133
+ "left_middle_1_joint": 30,
134
+ "left_middle_2_joint": 101,
135
+ "left_ring_1_joint": 23,
136
+ "left_ring_2_joint": 103,
137
+ "left_thumb_2_joint": 3,
138
+ "right_index_1_joint": 22,
139
+ "right_index_2_joint": 9,
140
+ "right_little_1_joint": 16,
141
+ "right_little_2_joint": 2,
142
+ "right_middle_1_joint": 31,
143
+ "right_middle_2_joint": 7,
144
+ "right_ring_1_joint": 24,
145
+ "right_thumb_2_joint": 48,
146
+ "right_thumb_4_joint": 6
147
+ },
148
+ "joint_limit_hard_episode_events": 127,
149
+ "finger_joint_limit_saturation_counts": {
150
+ "left_index_1_joint": 179,
151
+ "left_little_1_joint": 691,
152
+ "left_little_2_joint": 6,
153
+ "left_middle_1_joint": 577,
154
+ "left_ring_1_joint": 661,
155
+ "left_ring_2_joint": 23,
156
+ "left_thumb_2_joint": 27,
157
+ "right_index_1_joint": 58,
158
+ "right_little_1_joint": 174,
159
+ "right_little_2_joint": 16,
160
+ "right_middle_1_joint": 190,
161
+ "right_ring_1_joint": 137,
162
+ "right_thumb_2_joint": 1338,
163
+ "right_thumb_4_joint": 13
164
+ },
165
+ "joint_limit_max_excess_by_joint": {
166
+ "left_index_1_joint": 0.5477548241615295,
167
+ "left_index_2_joint": 0.05519115552306175,
168
+ "left_little_1_joint": 1.1197097301483154,
169
+ "left_little_2_joint": 0.15502040088176727,
170
+ "left_middle_1_joint": 0.9858368635177612,
171
+ "left_middle_2_joint": 0.10088048130273819,
172
+ "left_ring_1_joint": 0.9463478326797485,
173
+ "left_ring_2_joint": 0.30530086159706116,
174
+ "left_thumb_2_joint": 0.19430041313171387,
175
+ "left_thumb_3_joint": 2.025573303399142e-05,
176
+ "left_thumb_4_joint": 0.0009710192680358887,
177
+ "right_index_1_joint": 0.3938329219818115,
178
+ "right_index_2_joint": 0.011164755560457706,
179
+ "right_little_1_joint": 0.7533957362174988,
180
+ "right_little_2_joint": 0.31006598472595215,
181
+ "right_middle_1_joint": 0.5761584043502808,
182
+ "right_middle_2_joint": 0.01680421642959118,
183
+ "right_ring_1_joint": 0.6523428559303284,
184
+ "right_ring_2_joint": 0.004270551726222038,
185
+ "right_thumb_1_joint": 2.721354758250527e-05,
186
+ "right_thumb_2_joint": 0.47596219182014465,
187
+ "right_thumb_4_joint": 0.3904464542865753
188
+ },
189
+ "coverage": {
190
+ "hard_limit_source": {
191
+ "available": 25,
192
+ "unavailable": 0,
193
+ "unknown": 0
194
+ },
195
+ "joint_role_manifest": {
196
+ "available": 25,
197
+ "unavailable": 0,
198
+ "unknown": 0
199
+ }
200
+ },
201
+ "hard_limit_sources": {
202
+ "runtime.joint_pos_limits": 25
203
+ },
204
+ "soft_limit_sources": {
205
+ "runtime.soft_joint_pos_limits": 25
206
+ },
207
+ "unknown_runtime_joint_names": [],
208
+ "baseline_checks": {
209
+ "post_home": {
210
+ "checked": 25,
211
+ "hard_violation_episodes": 0
212
+ },
213
+ "post_reset": {
214
+ "checked": 25,
215
+ "hard_violation_episodes": 0
216
+ }
217
+ }
218
+ }
219
+ },
220
+ "robot_motion": {
221
+ "joint_acceleration_rms": 55.975754265611066,
222
+ "joint_effort_rms": 38.59412443385109,
223
+ "joint_jerk_rms": 5719.717567171954,
224
+ "joint_velocity_rms": 1.138932984897805
225
+ },
226
+ "grasp": {
227
+ "kinematic_grasp_stability_index": 0.9080470112263523,
228
+ "mean_kinematic_grasp_stability": 0.3805195281696569,
229
+ "per_object_best_kinematic_grasp_stability": {
230
+ "obj_011_banana_2": 0.7169296237422979,
231
+ "obj_013_apple_3": 0.8522337630273497,
232
+ "obj_013_apple_4": 0.2924523726192623
233
+ },
234
+ "grasp_gsi_diagnostics": {
235
+ "mean_anti_slip_score": 0.9090674127866221,
236
+ "mean_duration_score": 0.1300901489098694,
237
+ "mean_hold_score": 0.11528858518279317,
238
+ "mean_kinematic_stability_score": 0.08299550195909391,
239
+ "mean_legacy_gsi": 0.2802538537522831,
240
+ "mean_lift_score": 0.19726475918499903,
241
+ "mean_motion_stability_score": 0.1050934838230479
242
+ }
243
+ },
244
+ "tool": {
245
+ "tool_selection_accuracy": null,
246
+ "tool_switch_success_rate": null,
247
+ "tool_switch_total": 0,
248
+ "tool_switch_successful": 0
249
+ },
250
+ "runtime": {
251
+ "avg_steps": 1469.04,
252
+ "error_count": 0
253
+ }
254
+ }
255
+ }
mora_v0/per_task_eval/06_fruit_bowl_loading/Inv/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/06_fruit_bowl_loading/Inv/summary.json ADDED
@@ -0,0 +1,268 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2k",
5
+ "base_seed": 100600000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 50,
9
+ "successes": 31,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 1920,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.62,
50
+ "success_rate_ci": {
51
+ "mean": 0.62,
52
+ "ci_low": 0.48,
53
+ "ci_high": 0.76
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.835
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 22.844086021505362
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.58,
64
+ "hard_violation_rate": 0.36,
65
+ "drop_rate": 0.32,
66
+ "high_speed_violation_rate": 0.24,
67
+ "safety_violation_step_rate": 0.00045586931746232744
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.62,
74
+ "at_end_success_rate": 0.62,
75
+ "task_family_success_rate": {
76
+ "rigid_pick_place": 0.62
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.775,
81
+ "current_chain_depth_progress_score": 0.8133333333333334,
82
+ "latched_chain_depth_progress_score": 0.8666666666666667,
83
+ "avg_current_chain_depth": 2.44,
84
+ "avg_latched_chain_depth": 2.6
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.9364980924748796,
88
+ "expert_time_s": 21.333333333333332,
89
+ "expert_time_step": 1280,
90
+ "avg_steps_to_success": 1370.6451612903227,
91
+ "avg_policy_steps_to_success": 457.2258064516129,
92
+ "avg_policy_queries_to_success": 23.322580645161292
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.00043054324426997597,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_013_apple_3": 8,
98
+ "drop_to_table_obj_013_apple_4": 13,
99
+ "high_speed_obj_013_apple_3": 6,
100
+ "high_speed_obj_013_apple_4": 7
101
+ },
102
+ "robot_constraint_violation_rate": 1.0,
103
+ "finger_joint_limit_saturation_rate": 0.98,
104
+ "joint_limit_active_step_rate": 0.14257312903634292,
105
+ "finger_joint_limit_saturation_step_rate": 0.07801696846903887,
106
+ "max_joint_limit_excess_rad": 1.7565630674362183,
107
+ "robot_constraint_diagnostics": {
108
+ "joint_limit_soft_counts": {
109
+ "left_index_1_joint": 1320,
110
+ "left_index_2_joint": 56,
111
+ "left_little_1_joint": 1563,
112
+ "left_little_2_joint": 355,
113
+ "left_middle_1_joint": 1771,
114
+ "left_middle_2_joint": 166,
115
+ "left_ring_1_joint": 1718,
116
+ "left_ring_2_joint": 288,
117
+ "left_thumb_2_joint": 134,
118
+ "left_thumb_3_joint": 24,
119
+ "left_thumb_4_joint": 3,
120
+ "right_index_1_joint": 1351,
121
+ "right_index_2_joint": 51,
122
+ "right_little_1_joint": 1258,
123
+ "right_little_2_joint": 280,
124
+ "right_middle_1_joint": 1962,
125
+ "right_middle_2_joint": 27,
126
+ "right_ring_1_joint": 1456,
127
+ "right_ring_2_joint": 10,
128
+ "right_thumb_1_joint": 61,
129
+ "right_thumb_2_joint": 6611,
130
+ "right_thumb_3_joint": 71,
131
+ "right_thumb_4_joint": 264
132
+ },
133
+ "joint_limit_hard_counts": {
134
+ "left_index_1_joint": 74,
135
+ "left_index_2_joint": 79,
136
+ "left_little_1_joint": 42,
137
+ "left_little_2_joint": 127,
138
+ "left_middle_1_joint": 55,
139
+ "left_middle_2_joint": 168,
140
+ "left_ring_1_joint": 45,
141
+ "left_ring_2_joint": 158,
142
+ "left_thumb_2_joint": 6,
143
+ "left_thumb_3_joint": 1,
144
+ "left_thumb_4_joint": 1,
145
+ "right_index_1_joint": 62,
146
+ "right_index_2_joint": 37,
147
+ "right_little_1_joint": 54,
148
+ "right_little_2_joint": 13,
149
+ "right_middle_1_joint": 77,
150
+ "right_middle_2_joint": 31,
151
+ "right_ring_1_joint": 70,
152
+ "right_ring_2_joint": 2,
153
+ "right_thumb_1_joint": 4,
154
+ "right_thumb_2_joint": 94,
155
+ "right_thumb_3_joint": 6,
156
+ "right_thumb_4_joint": 21
157
+ },
158
+ "joint_limit_hard_episode_events": 259,
159
+ "finger_joint_limit_saturation_counts": {
160
+ "left_index_1_joint": 393,
161
+ "left_little_1_joint": 1059,
162
+ "left_little_2_joint": 51,
163
+ "left_middle_1_joint": 1022,
164
+ "left_ring_1_joint": 1145,
165
+ "left_ring_2_joint": 47,
166
+ "left_thumb_2_joint": 43,
167
+ "left_thumb_3_joint": 20,
168
+ "right_index_1_joint": 542,
169
+ "right_little_1_joint": 1016,
170
+ "right_little_2_joint": 191,
171
+ "right_middle_1_joint": 1034,
172
+ "right_ring_1_joint": 973,
173
+ "right_thumb_2_joint": 3260,
174
+ "right_thumb_3_joint": 25,
175
+ "right_thumb_4_joint": 181
176
+ },
177
+ "joint_limit_max_excess_by_joint": {
178
+ "left_index_1_joint": 0.6131885647773743,
179
+ "left_index_2_joint": 0.05519115552306175,
180
+ "left_little_1_joint": 1.1197097301483154,
181
+ "left_little_2_joint": 0.42039117217063904,
182
+ "left_middle_1_joint": 0.9858368635177612,
183
+ "left_middle_2_joint": 0.10088048130273819,
184
+ "left_ring_1_joint": 1.4075347185134888,
185
+ "left_ring_2_joint": 0.3474140167236328,
186
+ "left_thumb_2_joint": 0.2787731885910034,
187
+ "left_thumb_3_joint": 0.579958975315094,
188
+ "left_thumb_4_joint": 0.06642940640449524,
189
+ "right_index_1_joint": 1.0716137886047363,
190
+ "right_index_2_joint": 0.07298866659402847,
191
+ "right_little_1_joint": 1.7565630674362183,
192
+ "right_little_2_joint": 1.098952054977417,
193
+ "right_middle_1_joint": 1.3280658721923828,
194
+ "right_middle_2_joint": 0.045254215598106384,
195
+ "right_ring_1_joint": 1.434998631477356,
196
+ "right_ring_2_joint": 0.034781042486429214,
197
+ "right_thumb_1_joint": 0.20933566987514496,
198
+ "right_thumb_2_joint": 0.5103276371955872,
199
+ "right_thumb_3_joint": 0.6045276522636414,
200
+ "right_thumb_4_joint": 1.3936662673950195
201
+ },
202
+ "coverage": {
203
+ "hard_limit_source": {
204
+ "available": 50,
205
+ "unavailable": 0,
206
+ "unknown": 0
207
+ },
208
+ "joint_role_manifest": {
209
+ "available": 50,
210
+ "unavailable": 0,
211
+ "unknown": 0
212
+ }
213
+ },
214
+ "hard_limit_sources": {
215
+ "runtime.joint_pos_limits": 50
216
+ },
217
+ "soft_limit_sources": {
218
+ "runtime.soft_joint_pos_limits": 50
219
+ },
220
+ "unknown_runtime_joint_names": [],
221
+ "baseline_checks": {
222
+ "post_home": {
223
+ "checked": 50,
224
+ "hard_violation_episodes": 0
225
+ },
226
+ "post_reset": {
227
+ "checked": 50,
228
+ "hard_violation_episodes": 0
229
+ }
230
+ }
231
+ }
232
+ },
233
+ "robot_motion": {
234
+ "joint_acceleration_rms": 53.5226543569739,
235
+ "joint_effort_rms": 37.38168357461948,
236
+ "joint_jerk_rms": 5467.364484073931,
237
+ "joint_velocity_rms": 1.0933116054270042
238
+ },
239
+ "grasp": {
240
+ "kinematic_grasp_stability_index": 0.8839486909048769,
241
+ "mean_kinematic_grasp_stability": 0.40073923962192154,
242
+ "per_object_best_kinematic_grasp_stability": {
243
+ "obj_011_banana_2": 0.5969661212290107,
244
+ "obj_013_apple_3": 0.8238759913343433,
245
+ "obj_013_apple_4": 0.2825328867996342
246
+ },
247
+ "grasp_gsi_diagnostics": {
248
+ "mean_anti_slip_score": 0.901511964211662,
249
+ "mean_duration_score": 0.14632430630983992,
250
+ "mean_hold_score": 0.12828223567245922,
251
+ "mean_kinematic_stability_score": 0.08649630615627397,
252
+ "mean_legacy_gsi": 0.2880234600220139,
253
+ "mean_lift_score": 0.20296368230633782,
254
+ "mean_motion_stability_score": 0.11605354294384873
255
+ }
256
+ },
257
+ "tool": {
258
+ "tool_selection_accuracy": null,
259
+ "tool_switch_success_rate": null,
260
+ "tool_switch_total": 0,
261
+ "tool_switch_successful": 0
262
+ },
263
+ "runtime": {
264
+ "avg_steps": 1579.4,
265
+ "error_count": 0
266
+ }
267
+ }
268
+ }
mora_v0/per_task_eval/06_fruit_bowl_loading/None/native_shards/000/per_episode.jsonl ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {"scene":"Bench2Dex/scenes/06_fruit_bowl_loading.yaml","episode_index":1,"seed":100600000,"success":false,"steps":1920,"metric_schema_version":"core-v1.8-policy-step-query-split","error":null,"episode_seed":100600000,"base_seed":100600000,"seed_policy":"episode_seed = base_seed + episode_index - 1","scene_generalization_sample":{"robot_key":"multi_ur5_rh56dfx_with_flange","appearance":{"background":{"enabled":true,"clean_background":false,"intensity":1500.0,"yaw_deg":180.0,"base_yaw_deg":180.0,"flip_180_applied":false,"physics_enabled":false,"fill_light_intensity":0.0,"fill_light_jitter_ratio":0.0,"asset_kind":"usd_scene","asset_category":"iTHOR","asset_id":"FloorPlan2","asset_split":"seen","asset_pool_size":50,"asset_uri":"dex2bench_dataset/scenes/ithor/FloorPlan2_physics/scene.usda","scene_offset":[0.0,0.0,0.0]},"table_surface":{"enabled":true,"clean_surface":false,"asset_name":"table_surface_wood_wood0361_jpg","asset_uri":"dex2bench_dataset/textures/wood/wood0361.jpg","asset_category":"wood","asset_id":"wood/wood0361.jpg","asset_split":"seen","asset_pool_size":9436},"usd_scene_light":{"enabled":true,"distant":{"intensity":1401.736791365889,"color":[0.7758965034865142,0.9844434134178,0.9761731654571759],"pitch_deg":-24.843286810912247,"yaw_deg":-55.72261825498696,"angle_deg":0.589228407881461},"dome":{"intensity":286.41128980469847,"color":[0.8427530598467396,0.8342586020427717,0.8791819971325703],"exposure":0.017938984950519377}}},"spatial":{"object_pose":{"enabled":true,"position_jitter_cm":2.0,"position_jitter_m":0.02,"yaw_jitter_deg":10.0},"table_height":{"enabled":true,"height_offset_m":-0.046390284662677896,"offset_m_range":[-0.05,0.05]},"camera":{"enabled":true,"asset_split":"seen","profile_name":"train_default_camera","profile_split":"seen","profile_pool_size":1,"cameras":{"cam_chest":{"enabled":true,"position_offset_m":[-0.004705973555456916,0.002804365225703452,-0.0039765706162898885],"target_offset_m":[-0.003909459510789156,0.0011915617787830813,0.0006291158817543843],"distance_offset_m":0.009596471223193154,"offset_xyz_offset_m":[0.0,0.0,0.0],"rpy_offset_deg":[-0.19403878673204078,-0.24781859328055722,-0.009003999449496147],"has_offset":true},"cam_overhead":{"enabled":true,"position_offset_m":[-0.0006778796219223757,0.001341842852081652,-0.0041422883126386],"target_offset_m":[0.0026882105114519475,0.0014343554188501499,0.0007855005199941369],"distance_offset_m":-0.022319755116417123,"offset_xyz_offset_m":[0.0,0.0,0.0],"rpy_offset_deg":[0.5691522716976753,-0.925063079340317,0.12184715709207183],"has_offset":true},"cam_stereo_left":{"enabled":true,"position_offset_m":[-0.0027367302533426495,0.0003483015077487654,-0.001562901893611749],"target_offset_m":[0.004966803369322118,-0.004590387783871354,-0.002689252402592611],"distance_offset_m":0.015109907595066266,"offset_xyz_offset_m":[0.0,0.0,0.0],"rpy_offset_deg":[-0.7631888584760849,-0.11791606130498566,0.6559982939801121],"has_offset":true},"cam_stereo_right":{"enabled":true,"position_offset_m":[-0.0027367302533426495,0.0003483015077487654,-0.001562901893611749],"target_offset_m":[0.004966803369322118,-0.004590387783871354,-0.002689252402592611],"distance_offset_m":0.015109907595066266,"offset_xyz_offset_m":[0.0,0.0,0.0],"rpy_offset_deg":[-0.7631888584760849,-0.11791606130498566,0.6559982939801121],"has_offset":true},"cam_wrist_right":{"enabled":true,"position_offset_m":[0.0,0.0,0.0],"target_offset_m":[0.0,0.0,0.0],"distance_offset_m":0.0,"offset_xyz_offset_m":[0.00011018282137507853,0.004210133187355663,-0.0006686787554521902],"rpy_offset_deg":[0.6389064833324523,-0.9698906466951271,0.40356650770999436],"has_offset":true},"cam_wrist_left":{"enabled":true,"position_offset_m":[0.0,0.0,0.0],"target_offset_m":[0.0,0.0,0.0],"distance_offset_m":0.0,"offset_xyz_offset_m":[-0.004350653199783018,0.004507301764069569,0.0016909315262111967],"rpy_offset_deg":[0.542944372516835,0.777728491740165,0.7390072564504617],"has_offset":true}}}},"clutter":{"tabletop":{"enabled":true,"count":3,"assets":[{"name":"clutter_007_tuna_fish_can","uri":"dex2bench_dataset/Objects/007_tuna_fish_can/usd_decomposition_linux/textured.usd","scale":[0.9461188123817108,0.9461188123817108,0.9461188123817108],"body_type":"dynamic","category":"food","asset_code":"007","spawn_position":[0.8903854929721456,0.33050852387917745,0.7066543257287592],"spawn_yaw_deg":215.2758720211221,"obj_id":"distractor_00_clutter_007_tuna_fish_can"},{"name":"clutter_022_windex_bottle","uri":"dex2bench_dataset/Objects/022_windex_bottle/usd_decomposition_linux/textured.usd","scale":[0.9327128651575448,0.9327128651575448,0.9327128651575448],"body_type":"dynamic","category":"bottle","asset_code":"022","spawn_position":[0.6614923336927134,-0.0030948909966709015,0.7042169114212923],"spawn_yaw_deg":22.458665206127346,"obj_id":"distractor_01_clutter_022_windex_bottle"},{"name":"clutter_037_scissors","uri":"dex2bench_dataset/Objects/037_scissors/usd_decomposition_linux/textured.usd","scale":[1.0647312517434009,1.0647312517434009,1.0647312517434009],"body_type":"dynamic","category":"tool","asset_code":"037","spawn_position":[0.6328013601846563,0.2556872993681196,0.703995148041491],"spawn_yaw_deg":327.29877591491373,"obj_id":"distractor_02_clutter_037_scissors"}]}},"embodiment":{"robot_asset":{"enabled":false,"selection_mode":"default","default_robot_key":"","selected_robot_key":"","allowed_robot_keys":[]}},"resolved_object_placements":{"obj_024_bowl_1":{"position":[-0.5119727004550702,0.2959995563745159,0.7044047153663514],"rpy_deg":[0.0,0.0,-9.829534546305922]},"obj_011_banana_2":{"position":[-0.11112115433849223,0.30585795907079905,0.7040167153514149],"rpy_deg":[0.0,0.0,0.17274370238086334]},"obj_013_apple_3":{"position":[0.25802613730280216,0.30682079377455423,0.7040027153321234],"rpy_deg":[0.0,0.0,3.7072073646765666]},"obj_013_apple_4":{"position":[0.21017630433129814,0.09859509035603417,0.7751057177134016],"rpy_deg":[180.0,0.0,5.7925879459379885]}},"_replay_generalization_mode":"restored","_generalization_split":"seen","_generalization_profile":"none","_anchor_source_index":0},"task_family":"rigid_pick_place","robot_key":"multi_ur5_rh56dfx_with_flange","perturbation_axis":"none","stable_success":false,"ever_instant_success":false,"at_end_success":false,"at_end_success_observed":false,"at_end_success_budget":false,"first_success_step":null,"stable_success_step":null,"first_stable_success_step":null,"steps_to_stable_success":null,"policy_steps_to_stable_success":null,"policy_queries_to_stable_success":null,"time_to_stable_success_s":null,"expert_time_s":21.333333333333332,"expert_time_step":1280,"success_hold_s":0.0,"terminal_success_rate":0.0,"stage_completion_rate":0.25,"normalized_progress_score":0.3333333333333333,"chain_depth_progress_score":0.3333333333333333,"current_stage_completion_rate":0.0,"current_normalized_progress_score":0.0,"current_chain_depth_progress_score":0.0,"current_chain_depth":0,"latched_stage_completion_rate":0.25,"latched_normalized_progress_score":0.3333333333333333,"latched_chain_depth_progress_score":0.3333333333333333,"latched_chain_depth":1,"task_efficiency":null,"kinematic_grasp_stability_index":0.27264785009694187,"mean_kinematic_grasp_stability":0.023088155648577767,"per_object_best_kinematic_grasp_stability":{"obj_011_banana_2":0.0,"obj_013_apple_3":0.14403588289661903,"obj_013_apple_4":0.27264785009694187},"grasp_gsi_diagnostics":{"mean_legacy_gsi":0.2008164867395137,"mean_kinematic_stability_score":0.0007615884675746138,"mean_lift_score":0.05381944444444445,"mean_hold_score":0.01279513888888891,"mean_motion_stability_score":0.010090359834970733,"mean_duration_score":0.17620949074073666,"mean_anti_slip_score":0.7699652777777778},"robot_motion_metrics":{"joint_velocity_rms":0.9087753736542624,"joint_acceleration_rms":46.11043552774689,"joint_jerk_rms":4745.328836182512,"joint_effort_rms":30.576589551014358},"tool_selection_accuracy":null,"tool_switch_success_rate":null,"tool_switch_total":0,"tool_switch_successful":0,"safety_violation_rate":0.0015625000000000263,"safety_violation_rate_time":0.0015625000000000263,"safety_violation_rate_event":0.0015625,"safety_violation_step_rate":0.0015625,"safety_violation_events_per_step":0.0015625,"episode_violation_rate":1.0,"safety_hard_violation":true,"drop_violation":true,"high_speed_violation":true,"excessive_impact_violation":null,"robot_constraint_violation":true,"finger_joint_limit_saturation":true,"impact_metric_available":false,"chain_depth":1,"robot_constraint_diagnostics":{"joint_limit_soft_counts":{"left_little_1_joint":17,"left_middle_1_joint":16,"left_ring_1_joint":13,"left_ring_2_joint":2,"right_index_1_joint":76,"right_little_1_joint":49,"right_little_2_joint":33,"right_middle_1_joint":83,"right_ring_1_joint":57,"right_thumb_2_joint":265},"joint_limit_hard_counts":{"left_index_2_joint":1,"left_little_1_joint":1,"left_middle_1_joint":1,"left_ring_1_joint":1,"left_ring_2_joint":2,"right_index_1_joint":3,"right_little_1_joint":1,"right_little_2_joint":1,"right_middle_1_joint":3,"right_ring_1_joint":2,"right_thumb_2_joint":2},"finger_joint_limit_saturation_counts":{"right_index_1_joint":33,"right_little_1_joint":43,"right_little_2_joint":26,"right_middle_1_joint":43,"right_ring_1_joint":44,"right_thumb_2_joint":127},"joint_limit_max_excess_by_joint":{"left_index_2_joint":0.005047173239290714,"left_little_1_joint":0.17394332587718964,"left_little_2_joint":8.015087615831362e-08,"left_middle_1_joint":0.1832997351884842,"left_middle_2_joint":3.5554165833673324e-08,"left_ring_1_joint":0.06158727407455444,"left_ring_2_joint":0.014062504284083843,"left_thumb_2_joint":4.1358098314958625e-06,"left_thumb_3_joint":1.0876885426114313e-05,"left_thumb_4_joint":3.07825894196867e-06,"right_index_1_joint":0.5420998930931091,"right_index_2_joint":1.0806107866301318e-06,"right_little_1_joint":0.802967369556427,"right_little_2_joint":0.40490126609802246,"right_middle_1_joint":0.8122624754905701,"right_middle_2_joint":0.0005087577155791223,"right_ring_1_joint":0.7868996262550354,"right_ring_2_joint":0.00044516599155031145,"right_thumb_1_joint":1.964300281542819e-05,"right_thumb_2_joint":0.1740235686302185,"right_thumb_4_joint":0.001471877098083496},"joint_limit_baseline_checks":[{"phase":"post_reset","checked":true,"hard_violation":false,"hard_limit_source":"runtime.joint_pos_limits","violating_joints":[]},{"phase":"post_home","checked":true,"hard_violation":false,"hard_limit_source":"runtime.joint_pos_limits","violating_joints":[]}],"first_joint_limit_hard_event":{"sim_step":557,"hard_limit_source":"runtime.joint_pos_limits","soft_limit_source":"runtime.soft_joint_pos_limits","shield_applied":false,"violating_joints":[{"joint":"right_ring_1_joint","role":"finger","qpos":-0.008118748664855957,"hard_excess_rad":0.008118748664855957,"hard_tolerance_rad":0.005,"hard_lower":0.0,"hard_upper":1.3442769050598145,"operating_lower":0.0,"operating_upper":1.3442769050598145,"expanded_target":0.5076005458831787,"shielded_target":0.5076005458831787}],"command_context_available":true,"active_joint_names":["shoulder_pan_joint","shoulder_lift_joint","L_arm_shoulder_pan_joint","elbow_joint","L_arm_shoulder_lift_joint","wrist_1_joint","L_arm_elbow_joint","wrist_2_joint","L_arm_wrist_1_joint","wrist_3_joint","L_arm_wrist_2_joint","L_arm_wrist_3_joint","right_index_1_joint","right_little_1_joint","right_middle_1_joint","right_ring_1_joint","right_thumb_1_joint","right_index_2_joint","right_little_2_joint","right_middle_2_joint","right_ring_2_joint","right_thumb_2_joint","left_index_1_joint","left_little_1_joint","left_middle_1_joint","left_ring_1_joint","left_thumb_1_joint","right_thumb_3_joint","left_index_2_joint","left_little_2_joint","left_middle_2_joint","left_ring_2_joint","left_thumb_2_joint","right_thumb_4_joint","left_thumb_3_joint","left_thumb_4_joint"],"raw_active_action":[1.9538546800613403,-1.1197693347930908,-1.7864384651184082,1.8022772073745728,1.623165249824524,-0.059535492211580276,-1.9266407489776611,1.8392990827560425,0.21404603123664856,0.3312782049179077,-1.8961966037750244,-0.0710521712899208,0.27947697043418884,1.0309313535690308,0.2372041940689087,0.5076005458831787,1.1264077425003052,0.16532358527183533,0.9973748922348022,0.15719974040985107,0.4347700774669647,0.3846230208873749,0.014497824013233185,0.22927327454090118,0.022980688139796257,0.027925541624426842,0.31442201137542725,0.4397943913936615,-0.007947391830384731,0.10725340247154236,0.0005640677409246564,-0.0002349543501622975,0.18341654539108276,0.4170401096343994,0.2068338692188263,0.15277059376239777],"full_joint_names":["shoulder_pan_joint","shoulder_lift_joint","L_arm_shoulder_pan_joint","elbow_joint","L_arm_shoulder_lift_joint","wrist_1_joint","L_arm_elbow_joint","wrist_2_joint","L_arm_wrist_1_joint","wrist_3_joint","L_arm_wrist_2_joint","L_arm_wrist_3_joint","right_index_1_joint","right_little_1_joint","right_middle_1_joint","right_ring_1_joint","right_thumb_1_joint","right_index_2_joint","right_little_2_joint","right_middle_2_joint","right_ring_2_joint","right_thumb_2_joint","left_index_1_joint","left_little_1_joint","left_middle_1_joint","left_ring_1_joint","left_thumb_1_joint","right_thumb_3_joint","left_index_2_joint","left_little_2_joint","left_middle_2_joint","left_ring_2_joint","left_thumb_2_joint","right_thumb_4_joint","left_thumb_3_joint","left_thumb_4_joint"],"expanded_full_target":[1.9538546800613403,-1.1197693347930908,-1.7864384651184082,1.8022772073745728,1.623165249824524,-0.059535492211580276,-1.9266407489776611,1.8392990827560425,0.21404603123664856,0.3312782049179077,-1.8961966037750244,-0.0710521712899208,0.27947697043418884,1.0309313535690308,0.2372041940689087,0.5076005458831787,1.1264077425003052,0.16532358527183533,0.9973748922348022,0.15719974040985107,0.4347700774669647,0.3846230208873749,0.014497824013233185,0.22927327454090118,0.022980688139796257,0.027925541624426842,0.31442201137542725,0.4397943913936615,-0.007947391830384731,0.10725340247154236,0.0005640677409246564,-0.0002349543501622975,0.18341654539108276,0.4170401096343994,0.2068338692188263,0.15277059376239777],"shielded_full_target":[1.9538546800613403,-1.1197693347930908,-1.7864384651184082,1.8022772073745728,1.623165249824524,-0.059535492211580276,-1.9266407489776611,1.8392990827560425,0.21404603123664856,0.3312782049179077,-1.8961966037750244,-0.0710521712899208,0.27947697043418884,1.0309313535690308,0.2372041940689087,0.5076005458831787,1.1264077425003052,0.16532358527183533,0.9973748922348022,0.15719974040985107,0.4347700774669647,0.3846230208873749,0.014497824013233185,0.22927327454090118,0.022980688139796257,0.027925541624426842,0.31442201137542725,0.4397943913936615,-0.007947391830384731,0.10725340247154236,0.0005640677409246564,-0.0002349543501622975,0.18341654539108276,0.4170401096343994,0.2068338692188263,0.15277059376239777]},"joint_limit_active_steps":396,"joint_limit_hard_episode_events":7,"joint_limit_in_core_safety":false,"finger_joint_limit_saturation_steps":174,"max_joint_limit_excess_rad":0.8122624754905701,"max_joint_limit_excess_joint":"right_middle_1_joint","joint_role_manifest_available":true,"joint_role_manifest_robot_key":"multi_ur5_rh56dfx_with_flange","joint_limit_hard_source_available":true,"joint_limit_soft_source":"runtime.soft_joint_pos_limits","joint_limit_hard_source":"runtime.joint_pos_limits"},"joint_limit_active_steps":396,"joint_limit_active_step_rate":0.20625,"finger_joint_limit_saturation_steps":174,"finger_joint_limit_saturation_step_rate":0.090625,"max_joint_limit_excess_rad":0.8122624754905701,"max_joint_limit_excess_joint":"right_middle_1_joint","stage_completion":{"move_bowl":true,"place_apple_3":false,"place_apple_4":false,"place_banana":false},"current_stage_completion":{"move_bowl":false,"place_apple_3":false,"place_apple_4":false,"place_banana":false},"stage_first_completion_step":{"move_bowl":217,"place_apple_3":null,"place_apple_4":null,"place_banana":null},"violation_counts":{"drop_to_table_obj_013_apple_4":1,"high_speed_obj_013_apple_3":1,"drop_to_table_obj_013_apple_3":1},"policy_query_count":32,"terminated_reason":"max_steps","timeseries":null,"evaluation_protocol":"reach_and_stop","early_stop":true,"max_episode_steps":1920,"dwell_time_s":0.5}
2
+ {"scene":"Bench2Dex/scenes/06_fruit_bowl_loading.yaml","episode_index":2,"seed":100600001,"success":true,"steps":1463,"metric_schema_version":"core-v1.8-policy-step-query-split","error":null,"episode_seed":100600001,"base_seed":100600000,"seed_policy":"episode_seed = base_seed + episode_index - 1","scene_generalization_sample":{"robot_key":"multi_ur5_rh56dfx_with_flange","appearance":{"background":{"enabled":true,"clean_background":false,"intensity":1500.0,"yaw_deg":180.0,"base_yaw_deg":180.0,"flip_180_applied":false,"physics_enabled":false,"fill_light_intensity":0.0,"fill_light_jitter_ratio":0.0,"asset_kind":"usd_scene","asset_category":"iTHOR","asset_id":"FloorPlan4","asset_split":"seen","asset_pool_size":50,"asset_uri":"dex2bench_dataset/scenes/ithor/FloorPlan4_physics/scene.usda","scene_offset":[-1.96702,1.89762,0.0]},"table_surface":{"enabled":true,"clean_surface":false,"asset_name":"table_surface_carpet_carpet2933_jpg","asset_uri":"dex2bench_dataset/textures/carpet/carpet2933.jpg","asset_category":"carpet","asset_id":"carpet/carpet2933.jpg","asset_split":"seen","asset_pool_size":9436},"usd_scene_light":{"enabled":true,"distant":{"intensity":1182.0333066372884,"color":[0.43757489565493607,0.4341802809736522,0.8742925947949309],"pitch_deg":-7.967221575122796,"yaw_deg":-54.022182645367366,"angle_deg":0.676661762932691},"dome":{"intensity":182.42928187182295,"color":[0.950023045142157,0.9919304544487977,0.9473663914897738],"exposure":-0.15807323572261978}}},"spatial":{"object_pose":{"enabled":true,"position_jitter_cm":2.0,"position_jitter_m":0.02,"yaw_jitter_deg":10.0},"table_height":{"enabled":true,"height_offset_m":0.020188779777316285,"offset_m_range":[-0.05,0.05]},"camera":{"enabled":true,"asset_split":"seen","profile_name":"train_default_camera","profile_split":"seen","profile_pool_size":1,"cameras":{"cam_chest":{"enabled":true,"position_offset_m":[-0.004023163852638587,0.0011485266454746412,-0.0005756509468887797],"target_offset_m":[-0.0038560344839810067,-0.0036082624811809762,-0.004197949417707963],"distance_offset_m":0.02623212232771225,"offset_xyz_offset_m":[0.0,0.0,0.0],"rpy_offset_deg":[0.2565978704086207,-0.37604635542947307,-0.7814989053507113],"has_offset":true},"cam_overhead":{"enabled":true,"position_offset_m":[0.0038425786636525428,-0.0037317926239341225,0.0035350108168602434],"target_offset_m":[0.0038296188853385047,0.0029459935110905206,-0.004038732003367873],"distance_offset_m":-0.022348208506452926,"offset_xyz_offset_m":[0.0,0.0,0.0],"rpy_offset_deg":[0.4358014399566761,0.07850091559946115,0.7517513893858505],"has_offset":true},"cam_stereo_left":{"enabled":true,"position_offset_m":[-0.0002812566074747577,0.002731833337162203,-0.0027296783832966975],"target_offset_m":[-0.0028447293039350952,-0.0033663601524232846,-0.00043218846481178023],"distance_offset_m":-0.024507695256034315,"offset_xyz_offset_m":[0.0,0.0,0.0],"rpy_offset_deg":[-0.3863883371680208,-0.2344946757212909,-0.35588022853181234],"has_offset":true},"cam_stereo_right":{"enabled":true,"position_offset_m":[-0.0002812566074747577,0.002731833337162203,-0.0027296783832966975],"target_offset_m":[-0.0028447293039350952,-0.0033663601524232846,-0.00043218846481178023],"distance_offset_m":-0.024507695256034315,"offset_xyz_offset_m":[0.0,0.0,0.0],"rpy_offset_deg":[-0.3863883371680208,-0.2344946757212909,-0.35588022853181234],"has_offset":true},"cam_wrist_right":{"enabled":true,"position_offset_m":[0.0,0.0,0.0],"target_offset_m":[0.0,0.0,0.0],"distance_offset_m":0.0,"offset_xyz_offset_m":[-0.002753509708623927,-0.0037153883159107403,0.0038861481015847544],"rpy_offset_deg":[-0.30708865970723287,0.1105162436032745,0.44775959708216484],"has_offset":true},"cam_wrist_left":{"enabled":true,"position_offset_m":[0.0,0.0,0.0],"target_offset_m":[0.0,0.0,0.0],"distance_offset_m":0.0,"offset_xyz_offset_m":[0.0018449600113643952,-0.0020775752484223976,-0.0013189834112408284],"rpy_offset_deg":[-0.4250623336047905,-0.46799312753459166,0.5405684311148697],"has_offset":true}}}},"clutter":{"tabletop":{"enabled":true,"count":1,"assets":[{"name":"clutter_002_master_chef_can","uri":"dex2bench_dataset/Objects/002_master_chef_can/usd_decomposition_linux/textured.usd","scale":[0.9053449926020086,0.9053449926020086,0.9053449926020086],"body_type":"dynamic","category":"food","asset_code":"002","spawn_position":[0.4612123747381196,0.3787604533683781,0.7703580792889615],"spawn_yaw_deg":153.3549187128433,"obj_id":"distractor_00_clutter_002_master_chef_can"}]}},"embodiment":{"robot_asset":{"enabled":false,"selection_mode":"default","default_robot_key":"","selected_robot_key":"","allowed_robot_keys":[]}},"resolved_object_placements":{"obj_024_bowl_1":{"position":[-0.49476386371806064,0.3146503862298048,0.7709837798063456],"rpy_deg":[0.0,0.0,7.562078451490816]},"obj_011_banana_2":{"position":[-0.13155349836465635,0.29374706098126396,0.770595779791409],"rpy_deg":[0.0,0.0,4.380988462617026]},"obj_013_apple_3":{"position":[0.2662459969419636,0.28883723081218055,0.7705817797721176],"rpy_deg":[0.0,0.0,-5.307740703292152]},"obj_013_apple_4":{"position":[0.22788862431771426,0.09958193120702422,0.8416847821533958],"rpy_deg":[180.0,0.0,-1.6225473851572012]}},"_replay_generalization_mode":"restored","_generalization_split":"seen","_generalization_profile":"none","_anchor_source_index":1},"task_family":"rigid_pick_place","robot_key":"multi_ur5_rh56dfx_with_flange","perturbation_axis":"none","stable_success":true,"ever_instant_success":true,"at_end_success":true,"at_end_success_observed":true,"at_end_success_budget":null,"first_success_step":1358,"stable_success_step":1462,"first_stable_success_step":1462,"steps_to_stable_success":1463,"policy_steps_to_stable_success":488,"policy_queries_to_stable_success":25,"time_to_stable_success_s":24.38333333333323,"expert_time_s":21.333333333333332,"expert_time_step":1280,"success_hold_s":0.49999999999999994,"terminal_success_rate":1.0,"stage_completion_rate":1.0,"normalized_progress_score":1.0,"chain_depth_progress_score":1.0,"current_stage_completion_rate":1.0,"current_normalized_progress_score":1.0,"current_chain_depth_progress_score":1.0,"current_chain_depth":3,"latched_stage_completion_rate":1.0,"latched_normalized_progress_score":1.0,"latched_chain_depth_progress_score":1.0,"latched_chain_depth":3,"task_efficiency":0.8749145591250891,"kinematic_grasp_stability_index":0.8614319018049568,"mean_kinematic_grasp_stability":0.0816484575856919,"per_object_best_kinematic_grasp_stability":{"obj_011_banana_2":0.8614319018049568,"obj_013_apple_3":0.6295874750902164,"obj_013_apple_4":0.214888745754196},"grasp_gsi_diagnostics":{"mean_legacy_gsi":0.2570205047071631,"mean_kinematic_stability_score":0.008259720931430212,"mean_lift_score":0.18933697881066303,"mean_hold_score":0.10623148781043452,"mean_motion_stability_score":0.09748261562055746,"mean_duration_score":0.09422039948355813,"mean_anti_slip_score":0.8471937419305586},"robot_motion_metrics":{"joint_velocity_rms":1.0698267982392893,"joint_acceleration_rms":45.825846738409815,"joint_jerk_rms":4675.814036687151,"joint_effort_rms":37.32266544911298},"tool_selection_accuracy":null,"tool_switch_success_rate":null,"tool_switch_total":0,"tool_switch_successful":0,"safety_violation_rate":0.0,"safety_violation_rate_time":0.0,"safety_violation_rate_event":0.0,"safety_violation_step_rate":0.0,"safety_violation_events_per_step":0.0,"episode_violation_rate":0.0,"safety_hard_violation":false,"drop_violation":false,"high_speed_violation":false,"excessive_impact_violation":null,"robot_constraint_violation":true,"finger_joint_limit_saturation":true,"impact_metric_available":false,"chain_depth":3,"robot_constraint_diagnostics":{"joint_limit_soft_counts":{"left_index_1_joint":94,"left_little_1_joint":19,"left_little_2_joint":1,"left_middle_1_joint":30,"left_middle_2_joint":5,"left_ring_1_joint":28,"left_ring_2_joint":2,"right_thumb_2_joint":94},"joint_limit_hard_counts":{"left_index_1_joint":1,"left_little_1_joint":1,"left_little_2_joint":4,"left_middle_1_joint":2,"left_middle_2_joint":4,"left_ring_1_joint":1,"left_ring_2_joint":6,"right_thumb_2_joint":1},"finger_joint_limit_saturation_counts":{"left_index_1_joint":73,"left_ring_1_joint":2,"right_thumb_2_joint":39},"joint_limit_max_excess_by_joint":{"left_index_1_joint":0.6117360591888428,"left_index_2_joint":0.002825494622811675,"left_little_1_joint":0.18406687676906586,"left_little_2_joint":0.010833702981472015,"left_middle_1_joint":0.1361093521118164,"left_middle_2_joint":0.01379377581179142,"left_ring_1_joint":0.18561279773712158,"left_ring_2_joint":0.01241863053292036,"left_thumb_2_joint":1.3719751223106869e-05,"left_thumb_3_joint":1.339948175882455e-05,"left_thumb_4_joint":0.000911712646484375,"right_index_2_joint":3.083082447119523e-06,"right_little_1_joint":0.0038580894470214844,"right_little_2_joint":3.482488466488576e-07,"right_middle_1_joint":0.00020110345212742686,"right_middle_2_joint":1.3219719221524429e-06,"right_ring_1_joint":0.0001749225048115477,"right_ring_2_joint":1.5605625094394782e-06,"right_thumb_1_joint":2.051125920843333e-05,"right_thumb_2_joint":0.09694742411375046,"right_thumb_4_joint":0.001255035400390625},"joint_limit_baseline_checks":[{"phase":"post_reset","checked":true,"hard_violation":false,"hard_limit_source":"runtime.joint_pos_limits","violating_joints":[]},{"phase":"post_home","checked":true,"hard_violation":false,"hard_limit_source":"runtime.joint_pos_limits","violating_joints":[]}],"first_joint_limit_hard_event":{"sim_step":135,"hard_limit_source":"runtime.joint_pos_limits","soft_limit_source":"runtime.soft_joint_pos_limits","shield_applied":false,"violating_joints":[{"joint":"left_index_1_joint","role":"finger","qpos":-0.015026772394776344,"hard_excess_rad":0.015026772394776344,"hard_tolerance_rad":0.005,"hard_lower":0.0,"hard_upper":1.3442769050598145,"operating_lower":0.0,"operating_upper":1.3442769050598145,"expanded_target":0.025034639984369278,"shielded_target":0.025034639984369278}],"command_context_available":true,"active_joint_names":["shoulder_pan_joint","shoulder_lift_joint","L_arm_shoulder_pan_joint","elbow_joint","L_arm_shoulder_lift_joint","wrist_1_joint","L_arm_elbow_joint","wrist_2_joint","L_arm_wrist_1_joint","wrist_3_joint","L_arm_wrist_2_joint","L_arm_wrist_3_joint","right_index_1_joint","right_little_1_joint","right_middle_1_joint","right_ring_1_joint","right_thumb_1_joint","right_index_2_joint","right_little_2_joint","right_middle_2_joint","right_ring_2_joint","right_thumb_2_joint","left_index_1_joint","left_little_1_joint","left_middle_1_joint","left_ring_1_joint","left_thumb_1_joint","right_thumb_3_joint","left_index_2_joint","left_little_2_joint","left_middle_2_joint","left_ring_2_joint","left_thumb_2_joint","right_thumb_4_joint","left_thumb_3_joint","left_thumb_4_joint"],"raw_active_action":[1.9539389610290527,-1.7399444580078125,-1.6112391948699951,1.7849712371826172,1.3022633790969849,-0.08321313560009003,-1.7905954122543335,2.0391557216644287,0.16686885058879852,0.24075374007225037,-1.846603512763977,-0.1059429943561554,0.1264888495206833,0.08610692620277405,-0.0019546973053365946,-0.0014643652830272913,0.07904823124408722,-0.0022814483381807804,-0.0027459498960524797,0.007623131386935711,-0.0026881357189267874,0.204097181558609,0.025034639984369278,0.07648053020238876,0.010916220024228096,0.011933216825127602,0.4624929428100586,0.23371517658233643,-0.0033643750939518213,0.000910809962078929,-0.0015711082378402352,-0.007923672907054424,0.18904022872447968,0.2757337987422943,0.21473047137260437,0.1613098829984665],"full_joint_names":["shoulder_pan_joint","shoulder_lift_joint","L_arm_shoulder_pan_joint","elbow_joint","L_arm_shoulder_lift_joint","wrist_1_joint","L_arm_elbow_joint","wrist_2_joint","L_arm_wrist_1_joint","wrist_3_joint","L_arm_wrist_2_joint","L_arm_wrist_3_joint","right_index_1_joint","right_little_1_joint","right_middle_1_joint","right_ring_1_joint","right_thumb_1_joint","right_index_2_joint","right_little_2_joint","right_middle_2_joint","right_ring_2_joint","right_thumb_2_joint","left_index_1_joint","left_little_1_joint","left_middle_1_joint","left_ring_1_joint","left_thumb_1_joint","right_thumb_3_joint","left_index_2_joint","left_little_2_joint","left_middle_2_joint","left_ring_2_joint","left_thumb_2_joint","right_thumb_4_joint","left_thumb_3_joint","left_thumb_4_joint"],"expanded_full_target":[1.9539389610290527,-1.7399444580078125,-1.6112391948699951,1.7849712371826172,1.3022633790969849,-0.08321313560009003,-1.7905954122543335,2.0391557216644287,0.16686885058879852,0.24075374007225037,-1.846603512763977,-0.1059429943561554,0.1264888495206833,0.08610692620277405,-0.0019546973053365946,-0.0014643652830272913,0.07904823124408722,-0.0022814483381807804,-0.0027459498960524797,0.007623131386935711,-0.0026881357189267874,0.204097181558609,0.025034639984369278,0.07648053020238876,0.010916220024228096,0.011933216825127602,0.4624929428100586,0.23371517658233643,-0.0033643750939518213,0.000910809962078929,-0.0015711082378402352,-0.007923672907054424,0.18904022872447968,0.2757337987422943,0.21473047137260437,0.1613098829984665],"shielded_full_target":[1.9539389610290527,-1.7399444580078125,-1.6112391948699951,1.7849712371826172,1.3022633790969849,-0.08321313560009003,-1.7905954122543335,2.0391557216644287,0.16686885058879852,0.24075374007225037,-1.846603512763977,-0.1059429943561554,0.1264888495206833,0.08610692620277405,-0.0019546973053365946,-0.0014643652830272913,0.07904823124408722,-0.0022814483381807804,-0.0027459498960524797,0.007623131386935711,-0.0026881357189267874,0.204097181558609,0.025034639984369278,0.07648053020238876,0.010916220024228096,0.011933216825127602,0.4624929428100586,0.23371517658233643,-0.0033643750939518213,0.000910809962078929,-0.0015711082378402352,-0.007923672907054424,0.18904022872447968,0.2757337987422943,0.21473047137260437,0.1613098829984665]},"joint_limit_active_steps":216,"joint_limit_hard_episode_events":3,"joint_limit_in_core_safety":false,"finger_joint_limit_saturation_steps":114,"max_joint_limit_excess_rad":0.6117360591888428,"max_joint_limit_excess_joint":"left_index_1_joint","joint_role_manifest_available":true,"joint_role_manifest_robot_key":"multi_ur5_rh56dfx_with_flange","joint_limit_hard_source_available":true,"joint_limit_soft_source":"runtime.soft_joint_pos_limits","joint_limit_hard_source":"runtime.joint_pos_limits"},"joint_limit_active_steps":216,"joint_limit_active_step_rate":0.14764183185235816,"finger_joint_limit_saturation_steps":114,"finger_joint_limit_saturation_step_rate":0.07792207792207792,"max_joint_limit_excess_rad":0.6117360591888428,"max_joint_limit_excess_joint":"left_index_1_joint","stage_completion":{"move_bowl":true,"place_apple_3":true,"place_apple_4":true,"place_banana":true},"current_stage_completion":{"move_bowl":true,"place_apple_3":true,"place_apple_4":true,"place_banana":true},"stage_first_completion_step":{"move_bowl":356,"place_apple_3":1022,"place_apple_4":682,"place_banana":1358},"violation_counts":{},"policy_query_count":25,"terminated_reason":"stable_success","timeseries":null,"evaluation_protocol":"reach_and_stop","early_stop":true,"max_episode_steps":1920,"dwell_time_s":0.5}
3
+ {"scene":"Bench2Dex/scenes/06_fruit_bowl_loading.yaml","episode_index":3,"seed":100600002,"success":true,"steps":1429,"metric_schema_version":"core-v1.8-policy-step-query-split","error":null,"episode_seed":100600002,"base_seed":100600000,"seed_policy":"episode_seed = base_seed + episode_index - 1","scene_generalization_sample":{"robot_key":"multi_ur5_rh56dfx_with_flange","appearance":{"background":{"enabled":true,"clean_background":false,"intensity":1500.0,"yaw_deg":180.0,"base_yaw_deg":180.0,"flip_180_applied":false,"physics_enabled":false,"fill_light_intensity":0.0,"fill_light_jitter_ratio":0.0,"asset_kind":"usd_scene","asset_category":"iTHOR","asset_id":"FloorPlan6","asset_split":"seen","asset_pool_size":50,"asset_uri":"dex2bench_dataset/scenes/ithor/FloorPlan6_physics/scene.usda","scene_offset":[-0.32854,1.16282,0.0]},"table_surface":{"enabled":true,"clean_surface":false,"asset_name":"table_surface_stone_stone2274_jpg","asset_uri":"dex2bench_dataset/textures/stone/stone2274.jpg","asset_category":"stone","asset_id":"stone/stone2274.jpg","asset_split":"seen","asset_pool_size":9436},"usd_scene_light":{"enabled":true,"distant":{"intensity":1072.8561860635175,"color":[0.9661066803751989,0.4538292505284091,0.8207950326434174],"pitch_deg":-22.095631753268055,"yaw_deg":-8.285685826925729,"angle_deg":0.6130343029493859},"dome":{"intensity":170.10964991511773,"color":[0.8155009246317254,0.9759041414873626,0.8770243321413326],"exposure":0.1137722284601338}}},"spatial":{"object_pose":{"enabled":true,"position_jitter_cm":2.0,"position_jitter_m":0.02,"yaw_jitter_deg":10.0},"table_height":{"enabled":true,"height_offset_m":0.01193503061694453,"offset_m_range":[-0.05,0.05]},"camera":{"enabled":true,"asset_split":"seen","profile_name":"train_default_camera","profile_split":"seen","profile_pool_size":1,"cameras":{"cam_chest":{"enabled":true,"position_offset_m":[0.003400353661280436,0.001981717133597973,0.004268723749485421],"target_offset_m":[-0.003632034486083933,-0.004381326473124681,0.0012653550854477046],"distance_offset_m":0.01452585086596999,"offset_xyz_offset_m":[0.0,0.0,0.0],"rpy_offset_deg":[0.48025049446631796,-0.5790661548319151,0.09111524540649873],"has_offset":true},"cam_overhead":{"enabled":true,"position_offset_m":[0.0038969336778753106,0.003760103255951371,-0.001479863517464538],"target_offset_m":[0.001785647500803014,0.002312035601601323,0.00405775066380079],"distance_offset_m":0.019594932540250527,"offset_xyz_offset_m":[0.0,0.0,0.0],"rpy_offset_deg":[0.1713817022790911,-0.8625366516726629,-0.08189415774133257],"has_offset":true},"cam_stereo_left":{"enabled":true,"position_offset_m":[0.004982440222158778,0.0017364899742073623,-0.0005698864926310622],"target_offset_m":[0.0006718349095925094,-0.0030629820127613516,0.004570201091167947],"distance_offset_m":0.006976536601625338,"offset_xyz_offset_m":[0.0,0.0,0.0],"rpy_offset_deg":[0.7165883475559225,0.4833387001533511,0.6667817009320043],"has_offset":true},"cam_stereo_right":{"enabled":true,"position_offset_m":[0.004982440222158778,0.0017364899742073623,-0.0005698864926310622],"target_offset_m":[0.0006718349095925094,-0.0030629820127613516,0.004570201091167947],"distance_offset_m":0.006976536601625338,"offset_xyz_offset_m":[0.0,0.0,0.0],"rpy_offset_deg":[0.7165883475559225,0.4833387001533511,0.6667817009320043],"has_offset":true},"cam_wrist_right":{"enabled":true,"position_offset_m":[0.0,0.0,0.0],"target_offset_m":[0.0,0.0,0.0],"distance_offset_m":0.0,"offset_xyz_offset_m":[0.001323473970760036,0.0009100763735466521,0.0008592170171743182],"rpy_offset_deg":[0.6294478500967962,0.023067756249672167,0.20437586171139777],"has_offset":true},"cam_wrist_left":{"enabled":true,"position_offset_m":[0.0,0.0,0.0],"target_offset_m":[0.0,0.0,0.0],"distance_offset_m":0.0,"offset_xyz_offset_m":[0.002798801220590348,0.0012897520397729166,-0.0018426508506331528],"rpy_offset_deg":[-0.9892360242527425,0.16236313413285042,-0.2827967982658455],"has_offset":true}}}},"clutter":{"tabletop":{"enabled":true,"count":1,"assets":[{"name":"clutter_063b_marbles","uri":"dex2bench_dataset/Objects/063-b_marbles/google_16k/usd_decomposition_linux/textured.usd","scale":[0.9710277418342569,0.9710277418342569,0.9710277418342569],"body_type":"dynamic","category":"toy","asset_code":"063","spawn_position":[-0.1520973436839751,0.44181308581136974,0.7622156576233596],"spawn_yaw_deg":143.43661121805224,"obj_id":"distractor_00_clutter_063b_marbles"}]}},"embodiment":{"robot_asset":{"enabled":false,"selection_mode":"default","default_robot_key":"","selected_robot_key":"","allowed_robot_keys":[]}},"resolved_object_placements":{"obj_024_bowl_1":{"position":[-0.4922540029300627,0.318805309356915,0.7627300306459739],"rpy_deg":[0.0,0.0,-3.808953087405338]},"obj_011_banana_2":{"position":[-0.12959631093159232,0.3002819279323551,0.7623420306310373],"rpy_deg":[0.0,0.0,3.5811296517998734]},"obj_013_apple_3":{"position":[0.2744092730740134,0.3127231692220158,0.7623280306117459],"rpy_deg":[0.0,0.0,-4.348324145545314]},"obj_013_apple_4":{"position":[0.22748070807979712,0.10272135086828041,0.8334310329930241],"rpy_deg":[180.0,0.0,6.6376724079010785]}},"_replay_generalization_mode":"restored","_generalization_split":"seen","_generalization_profile":"none","_anchor_source_index":2},"task_family":"rigid_pick_place","robot_key":"multi_ur5_rh56dfx_with_flange","perturbation_axis":"none","stable_success":true,"ever_instant_success":true,"at_end_success":true,"at_end_success_observed":true,"at_end_success_budget":null,"first_success_step":1325,"stable_success_step":1428,"first_stable_success_step":1428,"steps_to_stable_success":1429,"policy_steps_to_stable_success":477,"policy_queries_to_stable_success":24,"time_to_stable_success_s":23.816666666666595,"expert_time_s":21.333333333333332,"expert_time_step":1280,"success_hold_s":0.49999999999999994,"terminal_success_rate":1.0,"stage_completion_rate":1.0,"normalized_progress_score":1.0,"chain_depth_progress_score":1.0,"current_stage_completion_rate":1.0,"current_normalized_progress_score":1.0,"current_chain_depth_progress_score":1.0,"current_chain_depth":3,"latched_stage_completion_rate":1.0,"latched_normalized_progress_score":1.0,"latched_chain_depth_progress_score":1.0,"latched_chain_depth":3,"task_efficiency":0.8957312806158179,"kinematic_grasp_stability_index":0.9826459258842869,"mean_kinematic_grasp_stability":0.4989791909942594,"per_object_best_kinematic_grasp_stability":{"obj_011_banana_2":0.7591197183273605,"obj_013_apple_3":0.9826459258842869,"obj_013_apple_4":0.3166575696037757},"grasp_gsi_diagnostics":{"mean_legacy_gsi":0.2543534854753201,"mean_kinematic_stability_score":0.08648041495421852,"mean_lift_score":0.18964310706787962,"mean_hold_score":0.10730114299043557,"mean_motion_stability_score":0.1036381379600186,"mean_duration_score":0.10999144701034089,"mean_anti_slip_score":0.8108234196407746},"robot_motion_metrics":{"joint_velocity_rms":1.149220416087748,"joint_acceleration_rms":49.20481266845619,"joint_jerk_rms":5045.028377578865,"joint_effort_rms":38.24805708681475},"tool_selection_accuracy":null,"tool_switch_success_rate":null,"tool_switch_total":0,"tool_switch_successful":0,"safety_violation_rate":0.0,"safety_violation_rate_time":0.0,"safety_violation_rate_event":0.0,"safety_violation_step_rate":0.0,"safety_violation_events_per_step":0.0,"episode_violation_rate":0.0,"safety_hard_violation":false,"drop_violation":false,"high_speed_violation":false,"excessive_impact_violation":null,"robot_constraint_violation":true,"finger_joint_limit_saturation":true,"impact_metric_available":false,"chain_depth":3,"robot_constraint_diagnostics":{"joint_limit_soft_counts":{"left_little_1_joint":36,"left_middle_1_joint":26,"left_ring_1_joint":34,"right_thumb_2_joint":92},"joint_limit_hard_counts":{"left_index_1_joint":1,"left_little_1_joint":1,"left_middle_1_joint":1,"left_middle_2_joint":2,"left_ring_1_joint":1,"right_thumb_2_joint":1},"finger_joint_limit_saturation_counts":{"left_little_1_joint":25,"left_middle_1_joint":1,"left_ring_1_joint":21},"joint_limit_max_excess_by_joint":{"left_index_1_joint":0.007024293765425682,"left_index_2_joint":3.1900270869300584e-08,"left_little_1_joint":0.5526291131973267,"left_little_2_joint":1.7471315061357018e-07,"left_middle_1_joint":0.24838416278362274,"left_middle_2_joint":0.008601075038313866,"left_ring_1_joint":0.5249602794647217,"left_ring_2_joint":3.392162568616186e-07,"left_thumb_2_joint":5.781252184533514e-06,"left_thumb_3_joint":1.4241742974263616e-05,"left_thumb_4_joint":0.0009059607982635498,"right_index_2_joint":1.280012725146662e-06,"right_little_2_joint":2.860175527530373e-07,"right_middle_1_joint":6.266761192819104e-05,"right_ring_1_joint":7.962180825415999e-05,"right_ring_2_joint":1.1570742799449363e-06,"right_thumb_1_joint":2.253305137855932e-05,"right_thumb_2_joint":0.08636168390512466,"right_thumb_4_joint":0.001330256462097168},"joint_limit_baseline_checks":[{"phase":"post_reset","checked":true,"hard_violation":false,"hard_limit_source":"runtime.joint_pos_limits","violating_joints":[]},{"phase":"post_home","checked":true,"hard_violation":false,"hard_limit_source":"runtime.joint_pos_limits","violating_joints":[]}],"first_joint_limit_hard_event":{"sim_step":910,"hard_limit_source":"runtime.joint_pos_limits","soft_limit_source":"runtime.soft_joint_pos_limits","shield_applied":false,"violating_joints":[{"joint":"right_thumb_2_joint","role":"finger","qpos":-0.014222213067114353,"hard_excess_rad":0.014222213067114353,"hard_tolerance_rad":0.005,"hard_lower":0.0,"hard_upper":0.5234770178794861,"operating_lower":0.0,"operating_upper":0.5234770178794861,"expanded_target":0.36467018723487854,"shielded_target":0.36467018723487854}],"command_context_available":true,"active_joint_names":["shoulder_pan_joint","shoulder_lift_joint","L_arm_shoulder_pan_joint","elbow_joint","L_arm_shoulder_lift_joint","wrist_1_joint","L_arm_elbow_joint","wrist_2_joint","L_arm_wrist_1_joint","wrist_3_joint","L_arm_wrist_2_joint","L_arm_wrist_3_joint","right_index_1_joint","right_little_1_joint","right_middle_1_joint","right_ring_1_joint","right_thumb_1_joint","right_index_2_joint","right_little_2_joint","right_middle_2_joint","right_ring_2_joint","right_thumb_2_joint","left_index_1_joint","left_little_1_joint","left_middle_1_joint","left_ring_1_joint","left_thumb_1_joint","right_thumb_3_joint","left_index_2_joint","left_little_2_joint","left_middle_2_joint","left_ring_2_joint","left_thumb_2_joint","right_thumb_4_joint","left_thumb_3_joint","left_thumb_4_joint"],"raw_active_action":[1.815077304840088,-0.9935176372528076,-1.805777907371521,1.384558081626892,1.944242238998413,-0.10795307159423828,-2.0867745876312256,1.877427577972412,0.01750536821782589,0.20221853256225586,-2.0802342891693115,-0.04203509911894798,0.6237634420394897,1.2325361967086792,0.7199846506118774,1.0467873811721802,1.2213164567947388,0.5478605628013611,1.2154310941696167,0.6505714058876038,0.9904810786247253,0.36467018723487854,0.020383678376674652,0.22329333424568176,0.01925363391637802,0.028315190225839615,0.1756991744041443,0.4195297658443451,-0.001227974658831954,0.09627914428710938,-0.003588135587051511,0.00042816365021280944,0.09239709377288818,0.4148246943950653,0.1043529361486435,0.07801569998264313],"full_joint_names":["shoulder_pan_joint","shoulder_lift_joint","L_arm_shoulder_pan_joint","elbow_joint","L_arm_shoulder_lift_joint","wrist_1_joint","L_arm_elbow_joint","wrist_2_joint","L_arm_wrist_1_joint","wrist_3_joint","L_arm_wrist_2_joint","L_arm_wrist_3_joint","right_index_1_joint","right_little_1_joint","right_middle_1_joint","right_ring_1_joint","right_thumb_1_joint","right_index_2_joint","right_little_2_joint","right_middle_2_joint","right_ring_2_joint","right_thumb_2_joint","left_index_1_joint","left_little_1_joint","left_middle_1_joint","left_ring_1_joint","left_thumb_1_joint","right_thumb_3_joint","left_index_2_joint","left_little_2_joint","left_middle_2_joint","left_ring_2_joint","left_thumb_2_joint","right_thumb_4_joint","left_thumb_3_joint","left_thumb_4_joint"],"expanded_full_target":[1.815077304840088,-0.9935176372528076,-1.805777907371521,1.384558081626892,1.944242238998413,-0.10795307159423828,-2.0867745876312256,1.877427577972412,0.01750536821782589,0.20221853256225586,-2.0802342891693115,-0.04203509911894798,0.6237634420394897,1.2325361967086792,0.7199846506118774,1.0467873811721802,1.2213164567947388,0.5478605628013611,1.2154310941696167,0.6505714058876038,0.9904810786247253,0.36467018723487854,0.020383678376674652,0.22329333424568176,0.01925363391637802,0.028315190225839615,0.1756991744041443,0.4195297658443451,-0.001227974658831954,0.09627914428710938,-0.003588135587051511,0.00042816365021280944,0.09239709377288818,0.4148246943950653,0.1043529361486435,0.07801569998264313],"shielded_full_target":[1.815077304840088,-0.9935176372528076,-1.805777907371521,1.384558081626892,1.944242238998413,-0.10795307159423828,-2.0867745876312256,1.877427577972412,0.01750536821782589,0.20221853256225586,-2.0802342891693115,-0.04203509911894798,0.6237634420394897,1.2325361967086792,0.7199846506118774,1.0467873811721802,1.2213164567947388,0.5478605628013611,1.2154310941696167,0.6505714058876038,0.9904810786247253,0.36467018723487854,0.020383678376674652,0.22329333424568176,0.01925363391637802,0.028315190225839615,0.1756991744041443,0.4195297658443451,-0.001227974658831954,0.09627914428710938,-0.003588135587051511,0.00042816365021280944,0.09239709377288818,0.4148246943950653,0.1043529361486435,0.07801569998264313]},"joint_limit_active_steps":128,"joint_limit_hard_episode_events":2,"joint_limit_in_core_safety":false,"finger_joint_limit_saturation_steps":25,"max_joint_limit_excess_rad":0.5526291131973267,"max_joint_limit_excess_joint":"left_little_1_joint","joint_role_manifest_available":true,"joint_role_manifest_robot_key":"multi_ur5_rh56dfx_with_flange","joint_limit_hard_source_available":true,"joint_limit_soft_source":"runtime.soft_joint_pos_limits","joint_limit_hard_source":"runtime.joint_pos_limits"},"joint_limit_active_steps":128,"joint_limit_active_step_rate":0.08957312806158152,"finger_joint_limit_saturation_steps":25,"finger_joint_limit_saturation_step_rate":0.01749475157452764,"max_joint_limit_excess_rad":0.5526291131973267,"max_joint_limit_excess_joint":"left_little_1_joint","stage_completion":{"move_bowl":true,"place_apple_3":true,"place_apple_4":true,"place_banana":true},"current_stage_completion":{"move_bowl":true,"place_apple_3":true,"place_apple_4":true,"place_banana":true},"stage_first_completion_step":{"move_bowl":324,"place_apple_3":1023,"place_apple_4":662,"place_banana":1325},"violation_counts":{},"policy_query_count":24,"terminated_reason":"stable_success","timeseries":null,"evaluation_protocol":"reach_and_stop","early_stop":true,"max_episode_steps":1920,"dwell_time_s":0.5}
4
+ {"scene":"Bench2Dex/scenes/06_fruit_bowl_loading.yaml","episode_index":4,"seed":100600003,"success":true,"steps":1367,"metric_schema_version":"core-v1.8-policy-step-query-split","error":null,"episode_seed":100600003,"base_seed":100600000,"seed_policy":"episode_seed = base_seed + episode_index - 1","scene_generalization_sample":{"robot_key":"multi_ur5_rh56dfx_with_flange","appearance":{"background":{"enabled":true,"clean_background":false,"intensity":1500.0,"yaw_deg":90.0,"base_yaw_deg":90.0,"flip_180_applied":false,"physics_enabled":false,"fill_light_intensity":0.0,"fill_light_jitter_ratio":0.0,"asset_kind":"usd_scene","asset_category":"iTHOR","asset_id":"FloorPlan9","asset_split":"seen","asset_pool_size":50,"asset_uri":"dex2bench_dataset/scenes/ithor/FloorPlan9_physics/scene.usda","scene_offset":[0.0,1.21873,0.0]},"table_surface":{"enabled":true,"clean_surface":false,"asset_name":"table_surface_flooring_flooring0169_jpg","asset_uri":"dex2bench_dataset/textures/flooring/flooring0169.jpg","asset_category":"flooring","asset_id":"flooring/flooring0169.jpg","asset_split":"seen","asset_pool_size":9436},"usd_scene_light":{"enabled":true,"distant":{"intensity":1289.8306277417275,"color":[0.4718179796513552,0.5744265088931003,0.5218560933573312],"pitch_deg":-24.312408349432904,"yaw_deg":75.76898246276082,"angle_deg":0.544050633739936},"dome":{"intensity":321.16977929597124,"color":[0.9593367517287849,0.8890279945694377,0.6700445455138871],"exposure":-0.46979858007630404}}},"spatial":{"object_pose":{"enabled":true,"position_jitter_cm":2.0,"position_jitter_m":0.02,"yaw_jitter_deg":10.0},"table_height":{"enabled":true,"height_offset_m":0.013386865795101732,"offset_m_range":[-0.05,0.05]},"camera":{"enabled":true,"asset_split":"seen","profile_name":"train_default_camera","profile_split":"seen","profile_pool_size":1,"cameras":{"cam_chest":{"enabled":true,"position_offset_m":[0.004931052555408005,0.002662783926972774,-0.003921256952861749],"target_offset_m":[0.0016375155033947408,-0.0020092302716603907,-0.0017629155171739507],"distance_offset_m":0.015536785135044744,"offset_xyz_offset_m":[0.0,0.0,0.0],"rpy_offset_deg":[-0.8918332567534619,0.9061387013108448,-0.612373487279473],"has_offset":true},"cam_overhead":{"enabled":true,"position_offset_m":[-0.004096651762740337,0.002150970198491203,-0.0036722075194285967],"target_offset_m":[0.0024075293631977603,-0.004269419230990157,0.0035548143709852277],"distance_offset_m":0.010010734889985128,"offset_xyz_offset_m":[0.0,0.0,0.0],"rpy_offset_deg":[0.4840328690193223,-0.16394745756406315,-0.6015653133360594],"has_offset":true},"cam_stereo_left":{"enabled":true,"position_offset_m":[-0.0027170108265941783,-0.004870951440573613,-0.003207127428672049],"target_offset_m":[-0.0012181297129606706,0.0038804575205631025,-0.002106266568970633],"distance_offset_m":-0.011313592831618254,"offset_xyz_offset_m":[0.0,0.0,0.0],"rpy_offset_deg":[0.8540358599597515,-0.028620229182253265,-0.512240618972231],"has_offset":true},"cam_stereo_right":{"enabled":true,"position_offset_m":[-0.0027170108265941783,-0.004870951440573613,-0.003207127428672049],"target_offset_m":[-0.0012181297129606706,0.0038804575205631025,-0.002106266568970633],"distance_offset_m":-0.011313592831618254,"offset_xyz_offset_m":[0.0,0.0,0.0],"rpy_offset_deg":[0.8540358599597515,-0.028620229182253265,-0.512240618972231],"has_offset":true},"cam_wrist_right":{"enabled":true,"position_offset_m":[0.0,0.0,0.0],"target_offset_m":[0.0,0.0,0.0],"distance_offset_m":0.0,"offset_xyz_offset_m":[0.0018960642812012839,0.0024478160565804133,-0.00045315395103729186],"rpy_offset_deg":[0.033140325802673454,-0.756978737621071,-0.3753180588357816],"has_offset":true},"cam_wrist_left":{"enabled":true,"position_offset_m":[0.0,0.0,0.0],"target_offset_m":[0.0,0.0,0.0],"distance_offset_m":0.0,"offset_xyz_offset_m":[-0.0033020530897772916,0.0015878983872834084,-0.0007477083400042692],"rpy_offset_deg":[0.3608605171289676,-0.9726825922462836,-0.31871251703772496],"has_offset":true}}}},"clutter":{"tabletop":{"enabled":true,"count":5,"assets":[{"name":"clutter_043_phillips_screwdriver","uri":"dex2bench_dataset/Objects/043_phillips_screwdriver/usd_decomposition_linux/textured.usd","scale":[0.8587373068910474,0.8587373068910474,0.8587373068910474],"body_type":"dynamic","category":"tool","asset_code":"043","spawn_position":[-0.8919363129301692,0.05519570333167231,0.7636951524864234],"spawn_yaw_deg":192.2200426856642,"obj_id":"distractor_00_clutter_043_phillips_screwdriver"},{"name":"clutter_052_extra_large_clamp","uri":"dex2bench_dataset/Objects/052_extra_large_clamp/usd_decomposition_linux/textured.usd","scale":[1.0496620492259,1.0496620492259,1.0496620492259],"body_type":"dynamic","category":"tool","asset_code":"052","spawn_position":[0.7869695255838483,0.24693824771685796,0.7638056809646927],"spawn_yaw_deg":1.883249832259426,"obj_id":"distractor_01_clutter_052_extra_large_clamp"},{"name":"clutter_035_power_drill","uri":"dex2bench_dataset/Objects/035_power_drill/usd_decomposition_linux/textured.usd","scale":[1.139137484482672,1.139137484482672,1.139137484482672],"body_type":"dynamic","category":"tool","asset_code":"035","placement_failed":true},{"name":"clutter_010_potted_meat_can","uri":"dex2bench_dataset/Objects/010_potted_meat_can/usd_decomposition_linux/textured.usd","scale":[0.9824520760572878,0.9824520760572878,0.9824520760572878],"body_type":"dynamic","category":"food","asset_code":"010","spawn_position":[-0.6923068508972088,0.08123552885393998,0.7665189230731265],"spawn_yaw_deg":321.8285903885525,"obj_id":"distractor_03_clutter_010_potted_meat_can"},{"name":"clutter_031_spoon","uri":"dex2bench_dataset/Objects/031_spoon/usd_decomposition_linux/textured.usd","scale":[1.1149856258528976,1.1149856258528976,1.1149856258528976],"body_type":"dynamic","category":"kitchen","asset_code":"031","placement_failed":true}]}},"embodiment":{"robot_asset":{"enabled":false,"selection_mode":"default","default_robot_key":"","selected_robot_key":"","allowed_robot_keys":[]}},"resolved_object_placements":{"obj_024_bowl_1":{"position":[-0.5138436144138258,0.3082777159459296,0.7641818658241311],"rpy_deg":[0.0,0.0,-9.184229858584985]},"obj_011_banana_2":{"position":[-0.1123676927756811,0.30286999347348803,0.7637938658091945],"rpy_deg":[0.0,0.0,5.472698912746798]},"obj_013_apple_3":{"position":[0.2611440737443205,0.3167152183878622,0.7637798657899031],"rpy_deg":[0.0,0.0,9.4947708894303]},"obj_013_apple_4":{"position":[0.1978282738960201,0.0967360752930076,0.8348828681711813],"rpy_deg":[180.0,0.0,-2.2295115233401557]}},"_replay_generalization_mode":"restored","_generalization_split":"seen","_generalization_profile":"none","_anchor_source_index":3},"task_family":"rigid_pick_place","robot_key":"multi_ur5_rh56dfx_with_flange","perturbation_axis":"none","stable_success":true,"ever_instant_success":true,"at_end_success":true,"at_end_success_observed":true,"at_end_success_budget":null,"first_success_step":1280,"stable_success_step":1366,"first_stable_success_step":1366,"steps_to_stable_success":1367,"policy_steps_to_stable_success":456,"policy_queries_to_stable_success":23,"time_to_stable_success_s":22.78333333333332,"expert_time_s":21.333333333333332,"expert_time_step":1280,"success_hold_s":0.49999999999999994,"terminal_success_rate":1.0,"stage_completion_rate":1.0,"normalized_progress_score":1.0,"chain_depth_progress_score":1.0,"current_stage_completion_rate":1.0,"current_normalized_progress_score":1.0,"current_chain_depth_progress_score":1.0,"current_chain_depth":3,"latched_stage_completion_rate":1.0,"latched_normalized_progress_score":1.0,"latched_chain_depth_progress_score":1.0,"latched_chain_depth":3,"task_efficiency":0.9363569861009514,"kinematic_grasp_stability_index":0.9950039179499223,"mean_kinematic_grasp_stability":0.4362429625816659,"per_object_best_kinematic_grasp_stability":{"obj_011_banana_2":0.7760803662405804,"obj_013_apple_3":0.9950039179499223,"obj_013_apple_4":0.07477893583112963},"grasp_gsi_diagnostics":{"mean_legacy_gsi":0.2854555558806595,"mean_kinematic_stability_score":0.08361057512538146,"mean_lift_score":0.1916605705925384,"mean_hold_score":0.10596196049743893,"mean_motion_stability_score":0.09923182171314131,"mean_duration_score":0.08734455010972934,"mean_anti_slip_score":0.9965861984881721},"robot_motion_metrics":{"joint_velocity_rms":1.2340103298565313,"joint_acceleration_rms":64.88688924841554,"joint_jerk_rms":6673.157770743822,"joint_effort_rms":40.537281194858075},"tool_selection_accuracy":null,"tool_switch_success_rate":null,"tool_switch_total":0,"tool_switch_successful":0,"safety_violation_rate":0.0,"safety_violation_rate_time":0.0,"safety_violation_rate_event":0.0,"safety_violation_step_rate":0.0,"safety_violation_events_per_step":0.0,"episode_violation_rate":0.0,"safety_hard_violation":false,"drop_violation":false,"high_speed_violation":false,"excessive_impact_violation":null,"robot_constraint_violation":true,"finger_joint_limit_saturation":true,"impact_metric_available":false,"chain_depth":3,"robot_constraint_diagnostics":{"joint_limit_soft_counts":{"left_index_1_joint":22,"left_little_1_joint":49,"left_little_2_joint":1,"left_middle_1_joint":30,"left_middle_2_joint":5,"left_ring_1_joint":50,"right_index_1_joint":65,"right_middle_1_joint":109,"right_ring_1_joint":162,"right_thumb_2_joint":16},"joint_limit_hard_counts":{"left_index_1_joint":1,"left_little_1_joint":1,"left_little_2_joint":6,"left_middle_1_joint":1,"left_middle_2_joint":7,"left_ring_1_joint":1,"right_index_1_joint":2,"right_middle_1_joint":2,"right_ring_1_joint":1,"right_thumb_2_joint":1},"finger_joint_limit_saturation_counts":{"left_little_1_joint":26,"left_ring_1_joint":43,"right_index_1_joint":18,"right_middle_1_joint":61,"right_ring_1_joint":120},"joint_limit_max_excess_by_joint":{"left_index_1_joint":0.14456355571746826,"left_index_2_joint":3.216705124486907e-08,"left_little_1_joint":0.25582605600357056,"left_little_2_joint":0.010279158130288124,"left_middle_1_joint":0.17057687044143677,"left_middle_2_joint":0.017673863098025322,"left_ring_1_joint":0.594866931438446,"left_ring_2_joint":3.632331697644986e-07,"left_thumb_3_joint":1.1054180504288524e-05,"left_thumb_4_joint":0.0008418560028076172,"right_index_1_joint":0.2386331856250763,"right_index_2_joint":1.507136857981095e-06,"right_little_1_joint":0.0017497539520263672,"right_little_2_joint":1.8652782785011368e-07,"right_middle_1_joint":0.4249444305896759,"right_middle_2_joint":2.293190118507482e-05,"right_ring_1_joint":0.41728025674819946,"right_ring_2_joint":8.893977792467922e-06,"right_thumb_2_joint":0.013744711875915527,"right_thumb_4_joint":0.0013561248779296875},"joint_limit_baseline_checks":[{"phase":"post_reset","checked":true,"hard_violation":false,"hard_limit_source":"runtime.joint_pos_limits","violating_joints":[]},{"phase":"post_home","checked":true,"hard_violation":false,"hard_limit_source":"runtime.joint_pos_limits","violating_joints":[]}],"first_joint_limit_hard_event":{"sim_step":780,"hard_limit_source":"runtime.joint_pos_limits","soft_limit_source":"runtime.soft_joint_pos_limits","shield_applied":false,"violating_joints":[{"joint":"right_middle_1_joint","role":"finger","qpos":-0.02514691837131977,"hard_excess_rad":0.02514691837131977,"hard_tolerance_rad":0.005,"hard_lower":0.0,"hard_upper":1.3442769050598145,"operating_lower":0.0,"operating_upper":1.3442769050598145,"expanded_target":-0.004089833702892065,"shielded_target":-0.004089833702892065}],"command_context_available":true,"active_joint_names":["shoulder_pan_joint","shoulder_lift_joint","L_arm_shoulder_pan_joint","elbow_joint","L_arm_shoulder_lift_joint","wrist_1_joint","L_arm_elbow_joint","wrist_2_joint","L_arm_wrist_1_joint","wrist_3_joint","L_arm_wrist_2_joint","L_arm_wrist_3_joint","right_index_1_joint","right_little_1_joint","right_middle_1_joint","right_ring_1_joint","right_thumb_1_joint","right_index_2_joint","right_little_2_joint","right_middle_2_joint","right_ring_2_joint","right_thumb_2_joint","left_index_1_joint","left_little_1_joint","left_middle_1_joint","left_ring_1_joint","left_thumb_1_joint","right_thumb_3_joint","left_index_2_joint","left_little_2_joint","left_middle_2_joint","left_ring_2_joint","left_thumb_2_joint","right_thumb_4_joint","left_thumb_3_joint","left_thumb_4_joint"],"raw_active_action":[1.8170868158340454,-1.022891879081726,-1.735153079032898,1.4777790307998657,1.8394111394882202,-0.14088307321071625,-2.0145044326782227,1.8405927419662476,0.14282122254371643,0.18355216085910797,-1.87846839427948,-0.023397186771035194,0.007290948648005724,0.20820003747940063,-0.004089833702892065,-0.0015998263843357563,0.27747291326522827,-0.005069158971309662,0.09607481211423874,-0.009440092369914055,-0.0029216900002211332,0.15479020774364471,0.02344173565506935,0.17767909169197083,0.01746380515396595,0.028598980978131294,0.2236274927854538,0.1779910773038864,0.00016380361921619624,0.07499966025352478,0.0022214623168110847,0.0032458947971463203,0.160631000995636,0.23494581878185272,0.1834627240896225,0.1375773549079895],"full_joint_names":["shoulder_pan_joint","shoulder_lift_joint","L_arm_shoulder_pan_joint","elbow_joint","L_arm_shoulder_lift_joint","wrist_1_joint","L_arm_elbow_joint","wrist_2_joint","L_arm_wrist_1_joint","wrist_3_joint","L_arm_wrist_2_joint","L_arm_wrist_3_joint","right_index_1_joint","right_little_1_joint","right_middle_1_joint","right_ring_1_joint","right_thumb_1_joint","right_index_2_joint","right_little_2_joint","right_middle_2_joint","right_ring_2_joint","right_thumb_2_joint","left_index_1_joint","left_little_1_joint","left_middle_1_joint","left_ring_1_joint","left_thumb_1_joint","right_thumb_3_joint","left_index_2_joint","left_little_2_joint","left_middle_2_joint","left_ring_2_joint","left_thumb_2_joint","right_thumb_4_joint","left_thumb_3_joint","left_thumb_4_joint"],"expanded_full_target":[1.8170868158340454,-1.022891879081726,-1.735153079032898,1.4777790307998657,1.8394111394882202,-0.14088307321071625,-2.0145044326782227,1.8405927419662476,0.14282122254371643,0.18355216085910797,-1.87846839427948,-0.023397186771035194,0.007290948648005724,0.20820003747940063,-0.004089833702892065,-0.0015998263843357563,0.27747291326522827,-0.005069158971309662,0.09607481211423874,-0.009440092369914055,-0.0029216900002211332,0.15479020774364471,0.02344173565506935,0.17767909169197083,0.01746380515396595,0.028598980978131294,0.2236274927854538,0.1779910773038864,0.00016380361921619624,0.07499966025352478,0.0022214623168110847,0.0032458947971463203,0.160631000995636,0.23494581878185272,0.1834627240896225,0.1375773549079895],"shielded_full_target":[1.8170868158340454,-1.022891879081726,-1.735153079032898,1.4777790307998657,1.8394111394882202,-0.14088307321071625,-2.0145044326782227,1.8405927419662476,0.14282122254371643,0.18355216085910797,-1.87846839427948,-0.023397186771035194,0.007290948648005724,0.20820003747940063,-0.004089833702892065,-0.0015998263843357563,0.27747291326522827,-0.005069158971309662,0.09607481211423874,-0.009440092369914055,-0.0029216900002211332,0.15479020774364471,0.02344173565506935,0.17767909169197083,0.01746380515396595,0.028598980978131294,0.2236274927854538,0.1779910773038864,0.00016380361921619624,0.07499966025352478,0.0022214623168110847,0.0032458947971463203,0.160631000995636,0.23494581878185272,0.1834627240896225,0.1375773549079895]},"joint_limit_active_steps":225,"joint_limit_hard_episode_events":2,"joint_limit_in_core_safety":false,"finger_joint_limit_saturation_steps":170,"max_joint_limit_excess_rad":0.594866931438446,"max_joint_limit_excess_joint":"left_ring_1_joint","joint_role_manifest_available":true,"joint_role_manifest_robot_key":"multi_ur5_rh56dfx_with_flange","joint_limit_hard_source_available":true,"joint_limit_soft_source":"runtime.soft_joint_pos_limits","joint_limit_hard_source":"runtime.joint_pos_limits"},"joint_limit_active_steps":225,"joint_limit_active_step_rate":0.1645940014630578,"finger_joint_limit_saturation_steps":170,"finger_joint_limit_saturation_step_rate":0.12435991221653256,"max_joint_limit_excess_rad":0.594866931438446,"max_joint_limit_excess_joint":"left_ring_1_joint","stage_completion":{"move_bowl":true,"place_apple_3":true,"place_apple_4":true,"place_banana":true},"current_stage_completion":{"move_bowl":true,"place_apple_3":true,"place_apple_4":true,"place_banana":true},"stage_first_completion_step":{"move_bowl":302,"place_apple_3":962,"place_apple_4":600,"place_banana":1280},"violation_counts":{},"policy_query_count":23,"terminated_reason":"stable_success","timeseries":null,"evaluation_protocol":"reach_and_stop","early_stop":true,"max_episode_steps":1920,"dwell_time_s":0.5}
mora_v0/per_task_eval/06_fruit_bowl_loading/None/native_shards/000/summary.json ADDED
@@ -0,0 +1,241 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2k",
5
+ "base_seed": 100600000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 4,
9
+ "successes": 3,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 1920,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.75,
50
+ "success_rate_ci": {
51
+ "mean": 0.75,
52
+ "ci_low": 0.25,
53
+ "ci_high": 1.0
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.8125
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 23.661111111111047
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.75,
64
+ "hard_violation_rate": 0.25,
65
+ "drop_rate": 0.25,
66
+ "high_speed_violation_rate": 0.25,
67
+ "safety_violation_step_rate": 0.0004855154555753358
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.75,
74
+ "at_end_success_rate": 0.75,
75
+ "task_family_success_rate": {
76
+ "rigid_pick_place": 0.75
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.75,
81
+ "current_chain_depth_progress_score": 0.75,
82
+ "latched_chain_depth_progress_score": 0.8333333333333334,
83
+ "avg_current_chain_depth": 2.25,
84
+ "avg_latched_chain_depth": 2.5
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.9023342752806195,
88
+ "expert_time_s": 21.333333333333332,
89
+ "expert_time_step": 1280,
90
+ "avg_steps_to_success": 1419.6666666666667,
91
+ "avg_policy_steps_to_success": 473.6666666666667,
92
+ "avg_policy_queries_to_success": 24
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.0004855154555753358,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_013_apple_3": 1,
98
+ "drop_to_table_obj_013_apple_4": 1,
99
+ "high_speed_obj_013_apple_3": 1
100
+ },
101
+ "robot_constraint_violation_rate": 1.0,
102
+ "finger_joint_limit_saturation_rate": 1.0,
103
+ "joint_limit_active_step_rate": 0.15617413821006634,
104
+ "finger_joint_limit_saturation_step_rate": 0.07816798834762907,
105
+ "max_joint_limit_excess_rad": 0.8122624754905701,
106
+ "robot_constraint_diagnostics": {
107
+ "joint_limit_soft_counts": {
108
+ "left_index_1_joint": 116,
109
+ "left_little_1_joint": 121,
110
+ "left_little_2_joint": 2,
111
+ "left_middle_1_joint": 102,
112
+ "left_middle_2_joint": 10,
113
+ "left_ring_1_joint": 125,
114
+ "left_ring_2_joint": 4,
115
+ "right_index_1_joint": 141,
116
+ "right_little_1_joint": 49,
117
+ "right_little_2_joint": 33,
118
+ "right_middle_1_joint": 192,
119
+ "right_ring_1_joint": 219,
120
+ "right_thumb_2_joint": 467
121
+ },
122
+ "joint_limit_hard_counts": {
123
+ "left_index_1_joint": 3,
124
+ "left_index_2_joint": 1,
125
+ "left_little_1_joint": 4,
126
+ "left_little_2_joint": 10,
127
+ "left_middle_1_joint": 5,
128
+ "left_middle_2_joint": 13,
129
+ "left_ring_1_joint": 4,
130
+ "left_ring_2_joint": 8,
131
+ "right_index_1_joint": 5,
132
+ "right_little_1_joint": 1,
133
+ "right_little_2_joint": 1,
134
+ "right_middle_1_joint": 5,
135
+ "right_ring_1_joint": 3,
136
+ "right_thumb_2_joint": 5
137
+ },
138
+ "joint_limit_hard_episode_events": 14,
139
+ "finger_joint_limit_saturation_counts": {
140
+ "left_index_1_joint": 73,
141
+ "left_little_1_joint": 51,
142
+ "left_middle_1_joint": 1,
143
+ "left_ring_1_joint": 66,
144
+ "right_index_1_joint": 51,
145
+ "right_little_1_joint": 43,
146
+ "right_little_2_joint": 26,
147
+ "right_middle_1_joint": 104,
148
+ "right_ring_1_joint": 164,
149
+ "right_thumb_2_joint": 166
150
+ },
151
+ "joint_limit_max_excess_by_joint": {
152
+ "left_index_1_joint": 0.6117360591888428,
153
+ "left_index_2_joint": 0.005047173239290714,
154
+ "left_little_1_joint": 0.5526291131973267,
155
+ "left_little_2_joint": 0.010833702981472015,
156
+ "left_middle_1_joint": 0.24838416278362274,
157
+ "left_middle_2_joint": 0.017673863098025322,
158
+ "left_ring_1_joint": 0.594866931438446,
159
+ "left_ring_2_joint": 0.014062504284083843,
160
+ "left_thumb_2_joint": 1.3719751223106869e-05,
161
+ "left_thumb_3_joint": 1.4241742974263616e-05,
162
+ "left_thumb_4_joint": 0.000911712646484375,
163
+ "right_index_1_joint": 0.5420998930931091,
164
+ "right_index_2_joint": 3.083082447119523e-06,
165
+ "right_little_1_joint": 0.802967369556427,
166
+ "right_little_2_joint": 0.40490126609802246,
167
+ "right_middle_1_joint": 0.8122624754905701,
168
+ "right_middle_2_joint": 0.0005087577155791223,
169
+ "right_ring_1_joint": 0.7868996262550354,
170
+ "right_ring_2_joint": 0.00044516599155031145,
171
+ "right_thumb_1_joint": 2.253305137855932e-05,
172
+ "right_thumb_2_joint": 0.1740235686302185,
173
+ "right_thumb_4_joint": 0.001471877098083496
174
+ },
175
+ "coverage": {
176
+ "hard_limit_source": {
177
+ "available": 4,
178
+ "unavailable": 0,
179
+ "unknown": 0
180
+ },
181
+ "joint_role_manifest": {
182
+ "available": 4,
183
+ "unavailable": 0,
184
+ "unknown": 0
185
+ }
186
+ },
187
+ "hard_limit_sources": {
188
+ "runtime.joint_pos_limits": 4
189
+ },
190
+ "soft_limit_sources": {
191
+ "runtime.soft_joint_pos_limits": 4
192
+ },
193
+ "unknown_runtime_joint_names": [],
194
+ "baseline_checks": {
195
+ "post_home": {
196
+ "checked": 4,
197
+ "hard_violation_episodes": 0
198
+ },
199
+ "post_reset": {
200
+ "checked": 4,
201
+ "hard_violation_episodes": 0
202
+ }
203
+ }
204
+ }
205
+ },
206
+ "robot_motion": {
207
+ "joint_acceleration_rms": 51.506996045757106,
208
+ "joint_effort_rms": 36.671148320450044,
209
+ "joint_jerk_rms": 5284.832255298087,
210
+ "joint_velocity_rms": 1.0904582294594578
211
+ },
212
+ "grasp": {
213
+ "kinematic_grasp_stability_index": 0.777932398934027,
214
+ "mean_kinematic_grasp_stability": 0.25998969170254876,
215
+ "per_object_best_kinematic_grasp_stability": {
216
+ "obj_011_banana_2": 0.5991579965932244,
217
+ "obj_013_apple_3": 0.6878183004552612,
218
+ "obj_013_apple_4": 0.2197432753215108
219
+ },
220
+ "grasp_gsi_diagnostics": {
221
+ "mean_anti_slip_score": 0.8561421594593208,
222
+ "mean_duration_score": 0.11694147183609126,
223
+ "mean_hold_score": 0.08307243254679948,
224
+ "mean_kinematic_stability_score": 0.0447780748696512,
225
+ "mean_legacy_gsi": 0.2494115082006641,
226
+ "mean_lift_score": 0.15611502522888138,
227
+ "mean_motion_stability_score": 0.07761073378217202
228
+ }
229
+ },
230
+ "tool": {
231
+ "tool_selection_accuracy": null,
232
+ "tool_switch_success_rate": null,
233
+ "tool_switch_total": 0,
234
+ "tool_switch_successful": 0
235
+ },
236
+ "runtime": {
237
+ "avg_steps": 1544.75,
238
+ "error_count": 0
239
+ }
240
+ }
241
+ }
mora_v0/per_task_eval/06_fruit_bowl_loading/None/native_shards/001/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/06_fruit_bowl_loading/None/native_shards/001/summary.json ADDED
@@ -0,0 +1,265 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2k",
5
+ "base_seed": 100600000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 25,
9
+ "successes": 19,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 1920,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.76,
50
+ "success_rate_ci": {
51
+ "mean": 0.76,
52
+ "ci_low": 0.56,
53
+ "ci_high": 0.92
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.87
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 23.315789473684166
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.56,
64
+ "hard_violation_rate": 0.4,
65
+ "drop_rate": 0.24,
66
+ "high_speed_violation_rate": 0.28,
67
+ "safety_violation_step_rate": 0.0005249343832020997
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.76,
74
+ "at_end_success_rate": 0.76,
75
+ "task_family_success_rate": {
76
+ "rigid_pick_place": 0.76
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.85,
81
+ "current_chain_depth_progress_score": 0.8666666666666667,
82
+ "latched_chain_depth_progress_score": 0.88,
83
+ "avg_current_chain_depth": 2.6,
84
+ "avg_latched_chain_depth": 2.64
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.9162671379183472,
88
+ "expert_time_s": 21.333333333333332,
89
+ "expert_time_step": 1280,
90
+ "avg_steps_to_success": 1398.9473684210527,
91
+ "avg_policy_steps_to_success": 466.6842105263158,
92
+ "avg_policy_queries_to_success": 23.789473684210527
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.0004986876640419947,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_011_banana_2": 2,
98
+ "drop_to_table_obj_013_apple_3": 4,
99
+ "drop_to_table_obj_013_apple_4": 6,
100
+ "high_speed_obj_011_banana_2": 1,
101
+ "high_speed_obj_013_apple_3": 5,
102
+ "high_speed_obj_013_apple_4": 1
103
+ },
104
+ "robot_constraint_violation_rate": 1.0,
105
+ "finger_joint_limit_saturation_rate": 0.92,
106
+ "joint_limit_active_step_rate": 0.1505774278215223,
107
+ "finger_joint_limit_saturation_step_rate": 0.0826246719160105,
108
+ "max_joint_limit_excess_rad": 1.3681563138961792,
109
+ "robot_constraint_diagnostics": {
110
+ "joint_limit_soft_counts": {
111
+ "left_index_1_joint": 543,
112
+ "left_index_2_joint": 25,
113
+ "left_little_1_joint": 632,
114
+ "left_little_2_joint": 123,
115
+ "left_middle_1_joint": 734,
116
+ "left_middle_2_joint": 93,
117
+ "left_ring_1_joint": 747,
118
+ "left_ring_2_joint": 118,
119
+ "left_thumb_3_joint": 3,
120
+ "left_thumb_4_joint": 32,
121
+ "right_index_1_joint": 597,
122
+ "right_little_1_joint": 638,
123
+ "right_little_2_joint": 53,
124
+ "right_middle_1_joint": 978,
125
+ "right_middle_2_joint": 24,
126
+ "right_ring_1_joint": 823,
127
+ "right_ring_2_joint": 28,
128
+ "right_thumb_1_joint": 65,
129
+ "right_thumb_2_joint": 3581,
130
+ "right_thumb_3_joint": 44,
131
+ "right_thumb_4_joint": 149
132
+ },
133
+ "joint_limit_hard_counts": {
134
+ "left_index_1_joint": 27,
135
+ "left_index_2_joint": 29,
136
+ "left_little_1_joint": 22,
137
+ "left_little_2_joint": 33,
138
+ "left_middle_1_joint": 23,
139
+ "left_middle_2_joint": 47,
140
+ "left_ring_1_joint": 22,
141
+ "left_ring_2_joint": 67,
142
+ "left_thumb_3_joint": 1,
143
+ "left_thumb_4_joint": 5,
144
+ "right_index_1_joint": 21,
145
+ "right_index_2_joint": 1,
146
+ "right_little_1_joint": 22,
147
+ "right_little_2_joint": 7,
148
+ "right_middle_1_joint": 28,
149
+ "right_middle_2_joint": 27,
150
+ "right_ring_1_joint": 32,
151
+ "right_ring_2_joint": 20,
152
+ "right_thumb_1_joint": 4,
153
+ "right_thumb_2_joint": 42,
154
+ "right_thumb_3_joint": 4,
155
+ "right_thumb_4_joint": 18
156
+ },
157
+ "joint_limit_hard_episode_events": 108,
158
+ "finger_joint_limit_saturation_counts": {
159
+ "left_index_1_joint": 191,
160
+ "left_little_1_joint": 388,
161
+ "left_little_2_joint": 34,
162
+ "left_middle_1_joint": 375,
163
+ "left_ring_1_joint": 446,
164
+ "right_index_1_joint": 423,
165
+ "right_little_1_joint": 524,
166
+ "right_little_2_joint": 5,
167
+ "right_middle_1_joint": 578,
168
+ "right_ring_1_joint": 567,
169
+ "right_thumb_1_joint": 6,
170
+ "right_thumb_2_joint": 1764,
171
+ "right_thumb_3_joint": 18,
172
+ "right_thumb_4_joint": 108
173
+ },
174
+ "joint_limit_max_excess_by_joint": {
175
+ "left_index_1_joint": 0.7259299755096436,
176
+ "left_index_2_joint": 0.043371912091970444,
177
+ "left_little_1_joint": 0.803394615650177,
178
+ "left_little_2_joint": 0.29885968565940857,
179
+ "left_middle_1_joint": 0.9780590534210205,
180
+ "left_middle_2_joint": 0.06611323356628418,
181
+ "left_ring_1_joint": 0.9009199142456055,
182
+ "left_ring_2_joint": 0.09001022577285767,
183
+ "left_thumb_2_joint": 0.000598454789724201,
184
+ "left_thumb_3_joint": 0.038882434368133545,
185
+ "left_thumb_4_joint": 0.08876076340675354,
186
+ "right_index_1_joint": 1.084540843963623,
187
+ "right_index_2_joint": 0.006405135150998831,
188
+ "right_little_1_joint": 1.348089337348938,
189
+ "right_little_2_joint": 0.30017805099487305,
190
+ "right_middle_1_joint": 1.2186177968978882,
191
+ "right_middle_2_joint": 0.07856173068284988,
192
+ "right_ring_1_joint": 1.059390902519226,
193
+ "right_ring_2_joint": 0.050104543566703796,
194
+ "right_thumb_1_joint": 0.2992635667324066,
195
+ "right_thumb_2_joint": 0.5160137414932251,
196
+ "right_thumb_3_joint": 0.5287180542945862,
197
+ "right_thumb_4_joint": 1.3681563138961792
198
+ },
199
+ "coverage": {
200
+ "hard_limit_source": {
201
+ "available": 25,
202
+ "unavailable": 0,
203
+ "unknown": 0
204
+ },
205
+ "joint_role_manifest": {
206
+ "available": 25,
207
+ "unavailable": 0,
208
+ "unknown": 0
209
+ }
210
+ },
211
+ "hard_limit_sources": {
212
+ "runtime.joint_pos_limits": 25
213
+ },
214
+ "soft_limit_sources": {
215
+ "runtime.soft_joint_pos_limits": 25
216
+ },
217
+ "unknown_runtime_joint_names": [],
218
+ "baseline_checks": {
219
+ "post_home": {
220
+ "checked": 25,
221
+ "hard_violation_episodes": 0
222
+ },
223
+ "post_reset": {
224
+ "checked": 25,
225
+ "hard_violation_episodes": 0
226
+ }
227
+ }
228
+ }
229
+ },
230
+ "robot_motion": {
231
+ "joint_acceleration_rms": 55.4466848016811,
232
+ "joint_effort_rms": 38.66477960359484,
233
+ "joint_jerk_rms": 5699.719921975496,
234
+ "joint_velocity_rms": 1.1319011800424932
235
+ },
236
+ "grasp": {
237
+ "kinematic_grasp_stability_index": 0.8725148810506497,
238
+ "mean_kinematic_grasp_stability": 0.30106191316853287,
239
+ "per_object_best_kinematic_grasp_stability": {
240
+ "obj_011_banana_2": 0.6488679572786213,
241
+ "obj_013_apple_3": 0.7595894775091738,
242
+ "obj_013_apple_4": 0.23822332109887515
243
+ },
244
+ "grasp_gsi_diagnostics": {
245
+ "mean_anti_slip_score": 0.9212429211308789,
246
+ "mean_duration_score": 0.10449475361261912,
247
+ "mean_hold_score": 0.09854895221487692,
248
+ "mean_kinematic_stability_score": 0.060358373288199756,
249
+ "mean_legacy_gsi": 0.2693817804421844,
250
+ "mean_lift_score": 0.18009597524831225,
251
+ "mean_motion_stability_score": 0.09228592384535429
252
+ }
253
+ },
254
+ "tool": {
255
+ "tool_selection_accuracy": null,
256
+ "tool_switch_success_rate": null,
257
+ "tool_switch_total": 0,
258
+ "tool_switch_successful": 0
259
+ },
260
+ "runtime": {
261
+ "avg_steps": 1524,
262
+ "error_count": 0
263
+ }
264
+ }
265
+ }
mora_v0/per_task_eval/06_fruit_bowl_loading/None/native_shards/002/per_episode.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
mora_v0/per_task_eval/06_fruit_bowl_loading/None/native_shards/002/summary.json ADDED
@@ -0,0 +1,261 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metric_schema_version": "core-v1.8-policy-step-query-split",
3
+ "metadata": {
4
+ "policy_name": "bench2dex_q1_2k",
5
+ "base_seed": 100600000,
6
+ "seed_policy": "episode_seed = base_seed + episode_index - 1",
7
+ "randomization_recorded": true,
8
+ "n_episodes": 21,
9
+ "successes": 14,
10
+ "error_count": 0,
11
+ "evaluation_protocol": "reach_and_stop",
12
+ "early_stop": true,
13
+ "success_definition": "terminal.raw_condition holds for dwell_time_s",
14
+ "measurement_window": "episode_start_to_stable_success_or_failure_budget",
15
+ "max_episode_steps": 1920,
16
+ "dwell_time_s": 0.5,
17
+ "field_semantics": {
18
+ "success_rate": "primary protocol success rate: stable_success for reach_and_stop; terminal at_end_success_budget for fixed_horizon",
19
+ "avg_time_to_success_s": "main efficiency metric: mean time_to_stable_success_s over reached-success episodes",
20
+ "task_efficiency": "benchmark-specific expert-normalized speed ratio: expert_time_s / time_to_stable_success_s over reached-success episodes",
21
+ "avg_steps_to_success": "diagnostic mean steps_to_stable_success over reached-success episodes, or None if no episode succeeds",
22
+ "avg_policy_steps_to_success": "diagnostic mean policy-rate control/action steps through stable success over reached-success episodes",
23
+ "avg_policy_queries_to_success": "diagnostic mean actual model inference queries through stable success over reached-success episodes; None when the first-success query count was not observed",
24
+ "safe_success_rate": "episode success rate with no task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
25
+ "hard_violation_rate": "episode rate with any task-level hard safety violation (drop or tracked-object high speed; joint limits are diagnostic by default)",
26
+ "drop_rate": "episode rate with drop_off_table or drop_to_table_after_lift violation",
27
+ "high_speed_violation_rate": "episode rate with any tracked object exceeding the configured speed threshold",
28
+ "excessive_impact_rate": "deprecated contact-force impact metric; no longer populated by core safety",
29
+ "robot_constraint_violation_rate": "diagnostic episode rate with any asset hard-joint-limit observation; excluded from safe_success_rate unless include_in_core_safety is explicitly enabled",
30
+ "finger_joint_limit_saturation_rate": "episode rate with dexterous finger joints saturated beyond finger joint-limit thresholds; diagnostic by default and not included in safe_success_rate",
31
+ "joint_limit_active_step_rate": "diagnostic mean fraction of steps with any joint exceeding the soft diagnostic margin",
32
+ "finger_joint_limit_saturation_step_rate": "diagnostic mean fraction of steps with finger joint-limit saturation",
33
+ "max_joint_limit_excess_rad": "diagnostic maximum operating/soft-limit excess observed in radians; hard-limit source and role validation are in robot_constraint_diagnostics",
34
+ "safety_violation_rate_time": "compatibility diagnostic; with fixed physics_dt this is equivalent to safety_violation_step_rate and is omitted from grouped summary output",
35
+ "safety_violation_step_rate": "core safety metric: fraction of executed steps with task-level safety violations; joint-limit observations are excluded by default",
36
+ "safety_violation_rate_event": "deprecated compatibility alias of safety_violation_events_per_step; omitted from grouped summary output",
37
+ "safety_violation_events_per_step": "diagnostic event density; may exceed 1 when multiple violations happen in one step",
38
+ "episode_violation_rate": "compatibility alias of hard_violation_rate; omitted from grouped core output",
39
+ "latched_stage_completion_rate": "main progress metric; ever-reached stage completion rate at episode end",
40
+ "current_stage_completion_rate": "diagnostic terminal-state stage completion rate at episode end",
41
+ "current_normalized_progress_score": "chain-depth diagnostic; normalized progress score at episode end",
42
+ "current_chain_depth_progress_score": "alias of current_normalized_progress_score for ordered dependency-depth diagnostics",
43
+ "latched_chain_depth_progress_score": "alias of latched_normalized_progress_score for ever-reached dependency-depth diagnostics",
44
+ "robustness": "success rates grouped by benchmark protocol category: cov, inv, and inv+cov"
45
+ }
46
+ },
47
+ "core_metrics": {
48
+ "completion": {
49
+ "success_rate": 0.6666666666666666,
50
+ "success_rate_ci": {
51
+ "mean": 0.6666666666666666,
52
+ "ci_low": 0.47619047619047616,
53
+ "ci_high": 0.8571428571428571
54
+ }
55
+ },
56
+ "progress": {
57
+ "latched_stage_completion_rate": 0.9047619047619048
58
+ },
59
+ "efficiency": {
60
+ "avg_time_to_success_s": 21.81309523809528
61
+ },
62
+ "safety": {
63
+ "safe_success_rate": 0.38095238095238093,
64
+ "hard_violation_rate": 0.5238095238095238,
65
+ "drop_rate": 0.2857142857142857,
66
+ "high_speed_violation_rate": 0.23809523809523808,
67
+ "safety_violation_step_rate": 0.00040928123917765956
68
+ },
69
+ "robustness": {}
70
+ },
71
+ "diagnostic_metrics": {
72
+ "completion": {
73
+ "ever_instant_success_rate": 0.7142857142857143,
74
+ "at_end_success_rate": 0.6666666666666666,
75
+ "task_family_success_rate": {
76
+ "rigid_pick_place": 0.6666666666666666
77
+ }
78
+ },
79
+ "progress": {
80
+ "current_stage_completion_rate": 0.8571428571428571,
81
+ "current_chain_depth_progress_score": 0.8888888888888888,
82
+ "latched_chain_depth_progress_score": 0.9047619047619048,
83
+ "avg_current_chain_depth": 2.6666666666666665,
84
+ "avg_latched_chain_depth": 2.7142857142857144
85
+ },
86
+ "efficiency": {
87
+ "task_efficiency": 0.9790692584654413,
88
+ "expert_time_s": 21.333333333333332,
89
+ "expert_time_step": 1280,
90
+ "avg_steps_to_success": 1308.7857142857142,
91
+ "avg_policy_steps_to_success": 436.64285714285717,
92
+ "avg_policy_queries_to_success": 22.214285714285715
93
+ },
94
+ "safety": {
95
+ "safety_violation_events_per_step": 0.0003777980669332242,
96
+ "violation_counts": {
97
+ "drop_to_table_obj_013_apple_3": 3,
98
+ "drop_to_table_obj_013_apple_4": 4,
99
+ "high_speed_obj_013_apple_4": 5
100
+ },
101
+ "robot_constraint_violation_rate": 1.0,
102
+ "finger_joint_limit_saturation_rate": 0.9523809523809523,
103
+ "joint_limit_active_step_rate": 0.14318546736769197,
104
+ "finger_joint_limit_saturation_step_rate": 0.065012750684759,
105
+ "max_joint_limit_excess_rad": 1.8464717864990234,
106
+ "robot_constraint_diagnostics": {
107
+ "joint_limit_soft_counts": {
108
+ "left_index_1_joint": 676,
109
+ "left_index_2_joint": 10,
110
+ "left_little_1_joint": 767,
111
+ "left_little_2_joint": 168,
112
+ "left_middle_1_joint": 874,
113
+ "left_middle_2_joint": 55,
114
+ "left_ring_1_joint": 888,
115
+ "left_ring_2_joint": 138,
116
+ "left_thumb_4_joint": 8,
117
+ "right_index_1_joint": 243,
118
+ "right_index_2_joint": 15,
119
+ "right_little_1_joint": 154,
120
+ "right_little_2_joint": 25,
121
+ "right_middle_1_joint": 376,
122
+ "right_middle_2_joint": 1,
123
+ "right_ring_1_joint": 215,
124
+ "right_ring_2_joint": 5,
125
+ "right_thumb_1_joint": 29,
126
+ "right_thumb_2_joint": 3106,
127
+ "right_thumb_3_joint": 4,
128
+ "right_thumb_4_joint": 74
129
+ },
130
+ "joint_limit_hard_counts": {
131
+ "left_index_1_joint": 35,
132
+ "left_index_2_joint": 19,
133
+ "left_little_1_joint": 23,
134
+ "left_little_2_joint": 46,
135
+ "left_middle_1_joint": 27,
136
+ "left_middle_2_joint": 51,
137
+ "left_ring_1_joint": 24,
138
+ "left_ring_2_joint": 62,
139
+ "left_thumb_4_joint": 4,
140
+ "right_index_1_joint": 17,
141
+ "right_index_2_joint": 3,
142
+ "right_little_1_joint": 20,
143
+ "right_little_2_joint": 1,
144
+ "right_middle_1_joint": 19,
145
+ "right_middle_2_joint": 7,
146
+ "right_ring_1_joint": 8,
147
+ "right_ring_2_joint": 2,
148
+ "right_thumb_1_joint": 1,
149
+ "right_thumb_2_joint": 44,
150
+ "right_thumb_3_joint": 1,
151
+ "right_thumb_4_joint": 2
152
+ },
153
+ "joint_limit_hard_episode_events": 106,
154
+ "finger_joint_limit_saturation_counts": {
155
+ "left_index_1_joint": 308,
156
+ "left_little_1_joint": 573,
157
+ "left_little_2_joint": 33,
158
+ "left_middle_1_joint": 570,
159
+ "left_ring_1_joint": 591,
160
+ "right_index_1_joint": 91,
161
+ "right_little_1_joint": 98,
162
+ "right_little_2_joint": 21,
163
+ "right_middle_1_joint": 129,
164
+ "right_ring_1_joint": 140,
165
+ "right_thumb_1_joint": 11,
166
+ "right_thumb_2_joint": 1164,
167
+ "right_thumb_4_joint": 55
168
+ },
169
+ "joint_limit_max_excess_by_joint": {
170
+ "left_index_1_joint": 1.0561188459396362,
171
+ "left_index_2_joint": 0.03003278374671936,
172
+ "left_little_1_joint": 1.2152470350265503,
173
+ "left_little_2_joint": 0.35844624042510986,
174
+ "left_middle_1_joint": 0.9178369045257568,
175
+ "left_middle_2_joint": 0.06374622136354446,
176
+ "left_ring_1_joint": 1.0948232412338257,
177
+ "left_ring_2_joint": 0.09489177167415619,
178
+ "left_thumb_1_joint": 1.3960625437903218e-05,
179
+ "left_thumb_2_joint": 0.0001357053843094036,
180
+ "left_thumb_3_joint": 0.0014423278626054525,
181
+ "left_thumb_4_joint": 0.05945345759391785,
182
+ "right_index_1_joint": 1.3628153800964355,
183
+ "right_index_2_joint": 0.04186670482158661,
184
+ "right_little_1_joint": 1.5627667903900146,
185
+ "right_little_2_joint": 0.46307969093322754,
186
+ "right_middle_1_joint": 1.5407770872116089,
187
+ "right_middle_2_joint": 0.014859394170343876,
188
+ "right_ring_1_joint": 1.5077695846557617,
189
+ "right_ring_2_joint": 0.01759668067097664,
190
+ "right_thumb_1_joint": 0.20241443812847137,
191
+ "right_thumb_2_joint": 0.3299984037876129,
192
+ "right_thumb_3_joint": 0.058279942721128464,
193
+ "right_thumb_4_joint": 1.8464717864990234
194
+ },
195
+ "coverage": {
196
+ "hard_limit_source": {
197
+ "available": 21,
198
+ "unavailable": 0,
199
+ "unknown": 0
200
+ },
201
+ "joint_role_manifest": {
202
+ "available": 21,
203
+ "unavailable": 0,
204
+ "unknown": 0
205
+ }
206
+ },
207
+ "hard_limit_sources": {
208
+ "runtime.joint_pos_limits": 21
209
+ },
210
+ "soft_limit_sources": {
211
+ "runtime.soft_joint_pos_limits": 21
212
+ },
213
+ "unknown_runtime_joint_names": [],
214
+ "baseline_checks": {
215
+ "post_home": {
216
+ "checked": 21,
217
+ "hard_violation_episodes": 0
218
+ },
219
+ "post_reset": {
220
+ "checked": 21,
221
+ "hard_violation_episodes": 0
222
+ }
223
+ }
224
+ }
225
+ },
226
+ "robot_motion": {
227
+ "joint_acceleration_rms": 50.883508255463994,
228
+ "joint_effort_rms": 37.06935578713564,
229
+ "joint_jerk_rms": 5191.61639514104,
230
+ "joint_velocity_rms": 1.0760794781969723
231
+ },
232
+ "grasp": {
233
+ "kinematic_grasp_stability_index": 0.9206533898175904,
234
+ "mean_kinematic_grasp_stability": 0.4187949098047191,
235
+ "per_object_best_kinematic_grasp_stability": {
236
+ "obj_011_banana_2": 0.6480668871561036,
237
+ "obj_013_apple_3": 0.8587079676162609,
238
+ "obj_013_apple_4": 0.23955071898877642
239
+ },
240
+ "grasp_gsi_diagnostics": {
241
+ "mean_anti_slip_score": 0.9175531276402206,
242
+ "mean_duration_score": 0.14958515758230523,
243
+ "mean_hold_score": 0.1418215730442538,
244
+ "mean_kinematic_stability_score": 0.0937217756978944,
245
+ "mean_legacy_gsi": 0.30311913534637996,
246
+ "mean_lift_score": 0.22309270697136802,
247
+ "mean_motion_stability_score": 0.13509446734065214
248
+ }
249
+ },
250
+ "tool": {
251
+ "tool_selection_accuracy": null,
252
+ "tool_switch_success_rate": null,
253
+ "tool_switch_total": 0,
254
+ "tool_switch_successful": 0
255
+ },
256
+ "runtime": {
257
+ "avg_steps": 1512.5238095238096,
258
+ "error_count": 0
259
+ }
260
+ }
261
+ }