wuschelschulz commited on
Commit
55e998f
·
verified ·
1 Parent(s): 8682c0c

Publish completed overnight evaluation report

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. MODEL_CARD.md +7 -0
  2. README.md +18 -0
  3. RESULTS.md +146 -0
  4. champion-metadata.json +34 -0
  5. champion.pt +3 -0
  6. comparison-checkpoints-manifest.json +269 -0
  7. comparison-checkpoints-publication.json +133 -0
  8. config.json +6 -0
  9. final-checkpoint-publication.json +30 -0
  10. final-vs-served.json +720 -0
  11. final-vs-served.jsonl +80 -0
  12. hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json +176 -0
  13. hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl +12 -0
  14. hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json +128 -0
  15. hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.jsonl +6 -0
  16. hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json +176 -0
  17. hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.jsonl +12 -0
  18. hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json +176 -0
  19. hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl +12 -0
  20. hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json +176 -0
  21. hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.jsonl +12 -0
  22. hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json +176 -0
  23. hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl +12 -0
  24. hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json +176 -0
  25. hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.jsonl +12 -0
  26. hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json +176 -0
  27. hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl +12 -0
  28. hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json +176 -0
  29. hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.jsonl +12 -0
  30. league-024737-duel.json +720 -0
  31. league-032337-duel.json +720 -0
  32. league-035932-duel.json +720 -0
  33. league-043512-duel.json +720 -0
  34. mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json +176 -0
  35. mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl +12 -0
  36. mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json +176 -0
  37. mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.jsonl +12 -0
  38. mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json +176 -0
  39. mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl +12 -0
  40. mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json +176 -0
  41. mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.jsonl +12 -0
  42. mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json +176 -0
  43. mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl +12 -0
  44. mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json +176 -0
  45. mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.jsonl +12 -0
  46. mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json +176 -0
  47. mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl +12 -0
  48. mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.json +176 -0
  49. mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.jsonl +12 -0
  50. mlp256-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json +176 -0
MODEL_CARD.md ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ # GIPF Zero checkpoint
2
+
3
+ Custom PyTorch `state_dict` checkpoint for the [GIPF Zero source](https://github.com/wusche1/gipf-zero). Load with `training.model.load_model`.
4
+
5
+ Source SHA-256: `aa533bb8922231a3e0dc53dd2ff947cab6899ef09a9245463c89ac99ea776ef0`.
6
+
7
+ Evaluation evidence: [{"report": "reports/final/20260909T191906Z/final-vs-champion-duel.json", "opponent": "/workspace/gipf/checkpoints/final/20260909T191906Z/champion-before-final-a9b830a90847.pt", "wins": 51, "losses": 29, "cutoffs": 0, "unfinished": 0}, {"report": "reports/final/20260909T191906Z/final-vs-greedy.json", "opponent": "greedy", "wins": 40, "losses": 0, "cutoffs": 0, "unfinished": 0}]. These are engine-baseline results, not a human or expert rating.
README.md ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: pytorch
3
+ tags:
4
+ - reinforcement-learning
5
+ - board-game
6
+ ---
7
+
8
+ # GIPF Zero checkpoint
9
+
10
+ Custom PyTorch `state_dict` checkpoint for the [GIPF Zero source](https://github.com/wusche1/gipf-zero). Load it with `training.model.load_model`.
11
+
12
+ Source training checkpoint SHA-256: `aa533bb8922231a3e0dc53dd2ff947cab6899ef09a9245463c89ac99ea776ef0`.
13
+
14
+ Served `champion.pt` artifact SHA-256: `d5a7e3bda2966a79d22cf34f131976907260e679683d0e2680d935312f41c2f6`.
15
+
16
+ `RESULTS.md` documents the controlled overnight comparison of MLP, square-CNN, hex-CNN, and query-Transformer policies, plus its separately labelled continuation evaluations. These automated results are not a human or expert rating.
17
+
18
+ [Full overnight comparison and results](RESULTS.md).
RESULTS.md ADDED
@@ -0,0 +1,146 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Overnight evaluation — `20260909T235853Z`
2
+
3
+ **Status:** `partial`
4
+
5
+ > This experiment is incomplete or still running. Missing matches are shown as unplayed; no winner is inferred from partial coverage.
6
+
7
+ ## Protocol
8
+
9
+ - Training: isolated **600 seconds per run**, 2 seed(s), native forest search: **True**.
10
+ - Primary matches: 12 paired-colour games per seed and pair, **50 ms per decision** on one CPU thread for each player. Secondary matches use **32 search simulations per decision** on GPU.
11
+ - Whole-match timeouts are 180 seconds (CPU) and 45 seconds (fixed simulations); unfinished games are reported separately.
12
+ - Training seconds are nominal allocations; the elapsed column includes checkpoint and shutdown overhead.
13
+ - Parameter counts are read from each run's initial `start` metric, before training.
14
+ - This is **not a pure architecture ablation**: the compared models use different output heads (flat, query).
15
+ - Results are automated game comparisons; this report makes no human-rating claim.
16
+
17
+ ## Training
18
+
19
+ | Model | Seed | Games | Updates | Seconds | Parameters | State |
20
+ |---|---|---|---|---|---|---|
21
+ | mlp256 | 410101 | 8740 | 8640 | 609.3 | 979371 | complete |
22
+ | mlp256 | 410102 | 8122 | 8032 | 611.9 | 979371 | complete |
23
+ | squarecnn32 | 410101 | 6214 | 6144 | 603.0 | 1163523 | complete |
24
+ | squarecnn32 | 410102 | 8449 | 8320 | 602.7 | 1163523 | complete |
25
+ | hexcnn32 | 410101 | 7877 | 7744 | 602.7 | 1163523 | complete |
26
+ | hexcnn32 | 410102 | 6813 | 6688 | 602.6 | 1163523 | complete |
27
+ | querytransformer256x2 | 410101 | 524 | 480 | 601.2 | 1102124 | complete |
28
+ | querytransformer256x2 | 410102 | 717 | 704 | 601.0 | 1102124 | complete |
29
+ | querytransformer164x5 | 410101 | 582 | 550 | 600.7 | 1115664 | complete |
30
+ | querytransformer164x5 | 410102 | 476 | 448 | 601.7 | 1115664 | complete |
31
+
32
+ ## Pooled equal-CPU ranking
33
+
34
+ Complete primary attempts only; retries use the last complete attempt for each pair/mode/seed. Cutoffs count as games and Wilson intervals use all recorded games.
35
+
36
+ | Model | W | L | C | Played | Unplayed | Win rate | Wilson 95% |
37
+ |---|---|---|---|---|---|---|---|
38
+ | squarecnn32 | 80 | 15 | 1 | 96 | 0 | 83.3% | 74.6%–89.5% |
39
+ | hexcnn32 | 74 | 15 | 7 | 96 | 0 | 77.1% | 67.7%–84.4% |
40
+ | mlp256 | 41 | 40 | 15 | 96 | 0 | 42.7% | 33.3%–52.7% |
41
+ | querytransformer256x2 | 9 | 74 | 13 | 96 | 0 | 9.4% | 5.0%–16.9% |
42
+ | querytransformer164x5 | 7 | 67 | 22 | 96 | 0 | 7.3% | 3.6%–14.3% |
43
+
44
+ ## Equal-CPU pairwise
45
+
46
+ | Pair (left perspective) | Left W | Left L | C | Games | Wilson 95% (left) |
47
+ |---|---|---|---|---|---|
48
+ | hexcnn32 vs mlp256 | 20 | 2 | 2 | 24 | 64.1%–93.3% |
49
+ | hexcnn32 vs querytransformer164x5 | 19 | 0 | 5 | 24 | 59.5%–90.8% |
50
+ | hexcnn32 vs querytransformer256x2 | 24 | 0 | 0 | 24 | 86.2%–100.0% |
51
+ | hexcnn32 vs squarecnn32 | 11 | 13 | 0 | 24 | 27.9%–64.9% |
52
+ | mlp256 vs querytransformer164x5 | 16 | 0 | 8 | 24 | 46.7%–82.0% |
53
+ | mlp256 vs querytransformer256x2 | 19 | 0 | 5 | 24 | 59.5%–90.8% |
54
+ | mlp256 vs squarecnn32 | 4 | 20 | 0 | 24 | 6.7%–35.9% |
55
+ | querytransformer164x5 vs querytransformer256x2 | 7 | 9 | 8 | 24 | 14.9%–49.2% |
56
+ | querytransformer164x5 vs squarecnn32 | 0 | 23 | 1 | 24 | 0.0%–13.8% |
57
+ | querytransformer256x2 vs squarecnn32 | 0 | 24 | 0 | 24 | 0.0%–13.8% |
58
+
59
+ ## Fixed-simulation pairwise
60
+
61
+ Three incomplete fixed-simulation matches reached the original 45-second whole-match cap. Their verified frozen-checkpoint retries used a 180-second cap and replace only those rows below; raw initial and retry reports remain linked for audit. All 20 secondary pair/seed matches are now complete. The primary equal-CPU ranking above is unchanged.
62
+
63
+ | Pair (left perspective) | Left W | Left L | C | Games | Wilson 95% (left) |
64
+ |---|---|---|---|---|---|
65
+ | hexcnn32 vs mlp256 | 18 | 1 | 5 | 24 | 55.1%–88.0% |
66
+ | hexcnn32 vs querytransformer164x5 | 17 | 0 | 7 | 24 | 50.8%–85.1% |
67
+ | hexcnn32 vs querytransformer256x2 | 20 | 2 | 2 | 24 | 64.1%–93.3% |
68
+ | hexcnn32 vs squarecnn32 | 7 | 14 | 3 | 24 | 14.9%–49.2% |
69
+ | mlp256 vs querytransformer164x5 | 14 | 2 | 8 | 24 | 38.8%–75.5% |
70
+ | mlp256 vs querytransformer256x2 | 13 | 2 | 9 | 24 | 35.1%–72.1% |
71
+ | mlp256 vs squarecnn32 | 4 | 15 | 5 | 24 | 6.7%–35.9% |
72
+ | querytransformer164x5 vs querytransformer256x2 | 2 | 14 | 8 | 24 | 2.3%–25.8% |
73
+ | querytransformer164x5 vs squarecnn32 | 1 | 21 | 2 | 24 | 0.7%–20.2% |
74
+ | querytransformer256x2 vs squarecnn32 | 0 | 21 | 3 | 24 | 0.0%–13.8% |
75
+
76
+ ## Extra tie-break attempts
77
+
78
+ Attempt 3 / 80-game tie-breaks are excluded from the primary pools above.
79
+
80
+ | Report | Candidate | Opponent | W | L | C | State |
81
+ |---|---|---|---|---|---|---|
82
+ | squarecnn32-seed410101-vs-hexcnn32-seed410102-equal_cpu_time-seed989001-attempt3.json | squarecnn32 | hexcnn32 | 42 | 38 | 0 | complete |
83
+
84
+ ## Continuation league evaluations
85
+
86
+ These intermediate continuation snapshots are included for audit. They do not by themselves imply promotion; promotion requires its explicit gate.
87
+
88
+ | Report | Evaluation | Candidate games | W | L | C | Unfinished |
89
+ |---|---|---|---|---|---|---|
90
+ | [league-024737-duel.json](league-024737-duel.json) | duel | 21515 | 13 | 65 | 2 | 0 |
91
+ | [league-032337-duel.json](league-032337-duel.json) | duel | 38076 | 31 | 49 | 0 | 0 |
92
+ | [league-035932-duel.json](league-035932-duel.json) | duel | 54044 | 34 | 46 | 0 | 0 |
93
+ | [league-043512-duel.json](league-043512-duel.json) | duel | 69700 | 39 | 41 | 0 | 0 |
94
+
95
+ ## Promotion and publication
96
+
97
+ - Promotion gate: `False`.
98
+ - `duel_report`: [final-vs-served.json](final-vs-served.json)
99
+ - Hugging Face: [wuschelschulz/gipf-zero](https://huggingface.co/wuschelschulz/gipf-zero)
100
+
101
+ ## Raw duel reports
102
+
103
+ - [hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json](hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json)
104
+ - [hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json](hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json)
105
+ - [hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json](hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json)
106
+ - [hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json](hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json)
107
+ - [hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json](hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json)
108
+ - [hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json](hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json)
109
+ - [hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json](hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json)
110
+ - [hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json](hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json)
111
+ - [hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json](hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json)
112
+ - [mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json](mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json)
113
+ - [mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json](mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json)
114
+ - [mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json](mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json)
115
+ - [mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json](mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json)
116
+ - [mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json](mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json)
117
+ - [mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json](mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json)
118
+ - [mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json](mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json)
119
+ - [mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.json](mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.json)
120
+ - [mlp256-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json](mlp256-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json)
121
+ - [mlp256-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json](mlp256-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json)
122
+ - [mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json](mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json)
123
+ - [mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json](mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json)
124
+ - [mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json](mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json)
125
+ - [mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json](mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json)
126
+ - [mlp256-seed410102-vs-squarecnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json](mlp256-seed410102-vs-squarecnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json)
127
+ - [mlp256-seed410102-vs-squarecnn32-seed410102-equal_simulations-seed1380102-attempt1.json](mlp256-seed410102-vs-squarecnn32-seed410102-equal_simulations-seed1380102-attempt1.json)
128
+ - [querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json](querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json)
129
+ - [querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json](querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json)
130
+ - [querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json](querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json)
131
+ - [querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json](querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json)
132
+ - [querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json](querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json)
133
+ - [querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-retry-complete.json](querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-retry-complete.json)
134
+ - [squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json](squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json)
135
+ - [squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json](squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json)
136
+ - [squarecnn32-seed410101-vs-hexcnn32-seed410102-equal_cpu_time-seed989001-attempt3.json](squarecnn32-seed410101-vs-hexcnn32-seed410102-equal_cpu_time-seed989001-attempt3.json)
137
+ - [squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json](squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json)
138
+ - [squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json](squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json)
139
+ - [squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json](squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json)
140
+ - [squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json](squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json)
141
+ - [squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json](squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json)
142
+ - [squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json](squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json)
143
+ - [squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json](squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json)
144
+ - [squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json](squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json)
145
+ - [squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json](squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json)
146
+ - [squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json](squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json)
champion-metadata.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "GIPF Zero \u2014 43,869 games",
3
+ "kind": "rl",
4
+ "games_trained": 43869,
5
+ "iteration": 727,
6
+ "architecture": {
7
+ "kind": "resnet",
8
+ "width": 32,
9
+ "blocks": 2,
10
+ "head": "flat"
11
+ },
12
+ "source_checkpoint_sha256": "aa533bb8922231a3e0dc53dd2ff947cab6899ef09a9245463c89ac99ea776ef0",
13
+ "updated_utc": "2026-09-09T21:20:52.781670+00:00",
14
+ "evaluations": [
15
+ {
16
+ "report": "reports/final/20260909T191906Z/final-vs-champion-duel.json",
17
+ "opponent": "/workspace/gipf/checkpoints/final/20260909T191906Z/champion-before-final-a9b830a90847.pt",
18
+ "wins": 51,
19
+ "losses": 29,
20
+ "cutoffs": 0,
21
+ "unfinished": 0
22
+ },
23
+ {
24
+ "report": "reports/final/20260909T191906Z/final-vs-greedy.json",
25
+ "opponent": "greedy",
26
+ "wins": 40,
27
+ "losses": 0,
28
+ "cutoffs": 0,
29
+ "unfinished": 0
30
+ }
31
+ ],
32
+ "evaluation_settings": "See linked reports for fixed simulation counts, baseline time budgets, paired opening seeds, and cutoffs.",
33
+ "served_artifact_sha256": "d5a7e3bda2966a79d22cf34f131976907260e679683d0e2680d935312f41c2f6"
34
+ }
champion.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5a7e3bda2966a79d22cf34f131976907260e679683d0e2680d935312f41c2f6
3
+ size 4663985
comparison-checkpoints-manifest.json ADDED
@@ -0,0 +1,269 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema": "gipf-zero-comparison-checkpoints-v1",
3
+ "repository": "wuschelschulz/gipf-zero",
4
+ "tag": "20260909T235853Z",
5
+ "source_github_commit": "f4c83a5140625caba30aea6c045a11d92261f6be",
6
+ "source_github_url": "https://github.com/wusche1/gipf-zero/tree/f4c83a5140625caba30aea6c045a11d92261f6be",
7
+ "protocol": {
8
+ "training_seconds": 600,
9
+ "seeds": [
10
+ 410101,
11
+ 410102
12
+ ],
13
+ "architectures": [
14
+ {
15
+ "name": "mlp256",
16
+ "kind": "mlp",
17
+ "head": "flat",
18
+ "width": 256,
19
+ "blocks": 2
20
+ },
21
+ {
22
+ "name": "squarecnn32",
23
+ "kind": "resnet",
24
+ "head": "flat",
25
+ "width": 32,
26
+ "blocks": 2
27
+ },
28
+ {
29
+ "name": "hexcnn32",
30
+ "kind": "hexresnet",
31
+ "head": "flat",
32
+ "width": 32,
33
+ "blocks": 2
34
+ },
35
+ {
36
+ "name": "querytransformer256x2",
37
+ "kind": "transformer",
38
+ "head": "query",
39
+ "width": 256,
40
+ "blocks": 2
41
+ },
42
+ {
43
+ "name": "querytransformer164x5",
44
+ "kind": "transformer",
45
+ "head": "query",
46
+ "width": 164,
47
+ "blocks": 5
48
+ }
49
+ ],
50
+ "selfplay": {
51
+ "games": 64,
52
+ "games_per_iteration": 32,
53
+ "updates": 32,
54
+ "simulations": 32,
55
+ "batch_size": 256,
56
+ "lr": 0.001,
57
+ "max_ply": 300,
58
+ "replay_size": 100000,
59
+ "device": "cuda",
60
+ "native_forest": true
61
+ },
62
+ "duels": {
63
+ "games": 12,
64
+ "seconds_cpu": 180,
65
+ "seconds_sim": 45,
66
+ "min_cpu_games_per_model": 96,
67
+ "equal_cpu_time": {
68
+ "device": "cpu",
69
+ "threads": 1,
70
+ "batch": 1,
71
+ "simulations": 10000,
72
+ "budget_ms": 50
73
+ },
74
+ "equal_simulations": {
75
+ "device": "cuda",
76
+ "threads": 2,
77
+ "batch": 64,
78
+ "simulations": 32,
79
+ "budget_ms": 0
80
+ }
81
+ }
82
+ },
83
+ "files": [
84
+ {
85
+ "name": "mlp256-seed410101",
86
+ "seed": 410101,
87
+ "frozen_source": "checkpoints/overnight/20260909T235853Z/mlp256-seed410101-9acc8f0f15e5.pt",
88
+ "remote_path": "comparisons/20260909T235853Z/mlp256-seed410101-9acc8f0f15e5.pt",
89
+ "bytes": 11775385,
90
+ "summary_source_sha256": "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f",
91
+ "file_sha256": "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f",
92
+ "config": {
93
+ "kind": "mlp",
94
+ "width": 256,
95
+ "blocks": 2,
96
+ "head": "flat"
97
+ },
98
+ "games": 8740,
99
+ "iteration": 270
100
+ },
101
+ {
102
+ "name": "mlp256-seed410102",
103
+ "seed": 410102,
104
+ "frozen_source": "checkpoints/overnight/20260909T235853Z/mlp256-seed410102-9ca408753dfb.pt",
105
+ "remote_path": "comparisons/20260909T235853Z/mlp256-seed410102-9ca408753dfb.pt",
106
+ "bytes": 11775385,
107
+ "summary_source_sha256": "9ca408753dfbe4c2103016e9694be0e5033ed517459bf8a7cef2cb56164fa9a7",
108
+ "file_sha256": "9ca408753dfbe4c2103016e9694be0e5033ed517459bf8a7cef2cb56164fa9a7",
109
+ "config": {
110
+ "kind": "mlp",
111
+ "width": 256,
112
+ "blocks": 2,
113
+ "head": "flat"
114
+ },
115
+ "games": 8122,
116
+ "iteration": 251
117
+ },
118
+ {
119
+ "name": "squarecnn32-seed410101",
120
+ "seed": 410101,
121
+ "frozen_source": "checkpoints/overnight/20260909T235853Z/squarecnn32-seed410101-e2a8946b1405.pt",
122
+ "remote_path": "comparisons/20260909T235853Z/squarecnn32-seed410101-e2a8946b1405.pt",
123
+ "bytes": 14001446,
124
+ "summary_source_sha256": "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1",
125
+ "file_sha256": "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1",
126
+ "config": {
127
+ "kind": "resnet",
128
+ "width": 32,
129
+ "blocks": 2,
130
+ "head": "flat"
131
+ },
132
+ "games": 6214,
133
+ "iteration": 192
134
+ },
135
+ {
136
+ "name": "squarecnn32-seed410102",
137
+ "seed": 410102,
138
+ "frozen_source": "checkpoints/overnight/20260909T235853Z/squarecnn32-seed410102-d3e8eb4f1e75.pt",
139
+ "remote_path": "comparisons/20260909T235853Z/squarecnn32-seed410102-d3e8eb4f1e75.pt",
140
+ "bytes": 14001446,
141
+ "summary_source_sha256": "d3e8eb4f1e75d18bbeb28bef1b46c6c3f84b1ab6f009e63cc950a88f647c8025",
142
+ "file_sha256": "d3e8eb4f1e75d18bbeb28bef1b46c6c3f84b1ab6f009e63cc950a88f647c8025",
143
+ "config": {
144
+ "kind": "resnet",
145
+ "width": 32,
146
+ "blocks": 2,
147
+ "head": "flat"
148
+ },
149
+ "games": 8449,
150
+ "iteration": 260
151
+ },
152
+ {
153
+ "name": "hexcnn32-seed410101",
154
+ "seed": 410101,
155
+ "frozen_source": "checkpoints/overnight/20260909T235853Z/hexcnn32-seed410101-d6f1c7081cdb.pt",
156
+ "remote_path": "comparisons/20260909T235853Z/hexcnn32-seed410101-d6f1c7081cdb.pt",
157
+ "bytes": 14003159,
158
+ "summary_source_sha256": "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5",
159
+ "file_sha256": "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5",
160
+ "config": {
161
+ "kind": "hexresnet",
162
+ "width": 32,
163
+ "blocks": 2,
164
+ "head": "flat"
165
+ },
166
+ "games": 7877,
167
+ "iteration": 242
168
+ },
169
+ {
170
+ "name": "hexcnn32-seed410102",
171
+ "seed": 410102,
172
+ "frozen_source": "checkpoints/overnight/20260909T235853Z/hexcnn32-seed410102-c7d5bd9e9ee0.pt",
173
+ "remote_path": "comparisons/20260909T235853Z/hexcnn32-seed410102-c7d5bd9e9ee0.pt",
174
+ "bytes": 14003159,
175
+ "summary_source_sha256": "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d",
176
+ "file_sha256": "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d",
177
+ "config": {
178
+ "kind": "hexresnet",
179
+ "width": 32,
180
+ "blocks": 2,
181
+ "head": "flat"
182
+ },
183
+ "games": 6813,
184
+ "iteration": 209
185
+ },
186
+ {
187
+ "name": "querytransformer256x2-seed410101",
188
+ "seed": 410101,
189
+ "frozen_source": "checkpoints/overnight/20260909T235853Z/querytransformer256x2-seed410101-454d54089b4a.pt",
190
+ "remote_path": "comparisons/20260909T235853Z/querytransformer256x2-seed410101-454d54089b4a.pt",
191
+ "bytes": 13274255,
192
+ "summary_source_sha256": "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4",
193
+ "file_sha256": "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4",
194
+ "config": {
195
+ "kind": "transformer",
196
+ "width": 256,
197
+ "blocks": 2,
198
+ "head": "query"
199
+ },
200
+ "games": 524,
201
+ "iteration": 15
202
+ },
203
+ {
204
+ "name": "querytransformer256x2-seed410102",
205
+ "seed": 410102,
206
+ "frozen_source": "checkpoints/overnight/20260909T235853Z/querytransformer256x2-seed410102-a0a1d4c39231.pt",
207
+ "remote_path": "comparisons/20260909T235853Z/querytransformer256x2-seed410102-a0a1d4c39231.pt",
208
+ "bytes": 13274191,
209
+ "summary_source_sha256": "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486",
210
+ "file_sha256": "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486",
211
+ "config": {
212
+ "kind": "transformer",
213
+ "width": 256,
214
+ "blocks": 2,
215
+ "head": "query"
216
+ },
217
+ "games": 717,
218
+ "iteration": 22
219
+ },
220
+ {
221
+ "name": "querytransformer164x5-seed410101",
222
+ "seed": 410101,
223
+ "frozen_source": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt",
224
+ "remote_path": "comparisons/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt",
225
+ "bytes": 13479583,
226
+ "summary_source_sha256": "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99",
227
+ "file_sha256": "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99",
228
+ "config": {
229
+ "kind": "transformer",
230
+ "width": 164,
231
+ "blocks": 5,
232
+ "head": "query"
233
+ },
234
+ "games": 582,
235
+ "iteration": 18
236
+ },
237
+ {
238
+ "name": "querytransformer164x5-seed410102",
239
+ "seed": 410102,
240
+ "frozen_source": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410102-02b9f1909248.pt",
241
+ "remote_path": "comparisons/20260909T235853Z/querytransformer164x5-seed410102-02b9f1909248.pt",
242
+ "bytes": 13479583,
243
+ "summary_source_sha256": "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c",
244
+ "file_sha256": "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c",
245
+ "config": {
246
+ "kind": "transformer",
247
+ "width": 164,
248
+ "blocks": 5,
249
+ "head": "query"
250
+ },
251
+ "games": 476,
252
+ "iteration": 14
253
+ }
254
+ ],
255
+ "total_files": 10,
256
+ "total_bytes": 133067592,
257
+ "source_github_commit_provenance": {
258
+ "kind": "inferred_repository_tree_at_backend_freeze",
259
+ "checkpoint_embedded_git_commit": null,
260
+ "basis": "The checkpoints contain config and RNG/training state but no Git revision. This commit was the latest local commit before summary backend_frozen at 2026-09-09T23:58:53Z; committed blobs match every recorded Python backend hash.",
261
+ "verified_backend_tree_sha256": {
262
+ "training/model.py": "1130a6f7e23524c04790adc702b8ecc7a9d92008bb354a398f377735b379dd71",
263
+ "training/native_search.py": "88beacc193f1fcfae082e46d93b3e1c8ca5edf76bf99c2c6298bd3d6cbe359f0",
264
+ "training/query_transformer.py": "9ff230547d8a0f9fd393fe05414a496724b3c514f76839a8f4c956e941f70385",
265
+ "training/search.py": "54587d2fde6bc62a671d9ff22ca2681f6fff521bbc5ed439c9cdd2b9c1786664",
266
+ "training/train.py": "8cbb2ed0b29d35c5101b6d9882cc2e83dc919277a32816b322a059abf97bd648"
267
+ }
268
+ }
269
+ }
comparison-checkpoints-publication.json ADDED
@@ -0,0 +1,133 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema": "gipf-zero-comparison-publication-receipt-v1",
3
+ "repository": "wuschelschulz/gipf-zero",
4
+ "tag": "20260909T235853Z",
5
+ "repository_commit": "3996ec0178ef203edefeaa052cffa78c11eb2093",
6
+ "source_github_commit": "f4c83a5140625caba30aea6c045a11d92261f6be",
7
+ "source_github_url": "https://github.com/wusche1/gipf-zero/tree/f4c83a5140625caba30aea6c045a11d92261f6be",
8
+ "manifest": "comparison-checkpoints-manifest.json",
9
+ "remote_manifest_path": "comparisons/20260909T235853Z/manifest.json",
10
+ "remote_manifest_sha256": "19504f4cc194e7b527bcfe470a1c38e3103bb4d6decdc9af232e018a8a019a25",
11
+ "files_uploaded": 10,
12
+ "total_bytes": 133067592,
13
+ "network_protocol": {
14
+ "per_process_timeout_seconds": 180,
15
+ "maximum_retries": 2,
16
+ "attempts_per_file": 1
17
+ },
18
+ "remote_lfs_sha256_verification": true,
19
+ "files": [
20
+ {
21
+ "name": "mlp256-seed410101",
22
+ "remote_path": "comparisons/20260909T235853Z/mlp256-seed410101-9acc8f0f15e5.pt",
23
+ "bytes": 11775385,
24
+ "local_sha256": "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f",
25
+ "remote_lfs_sha256": "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f",
26
+ "remote_lfs_size": 11775385,
27
+ "verified": true,
28
+ "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/mlp256-seed410101-9acc8f0f15e5.pt"
29
+ },
30
+ {
31
+ "name": "mlp256-seed410102",
32
+ "remote_path": "comparisons/20260909T235853Z/mlp256-seed410102-9ca408753dfb.pt",
33
+ "bytes": 11775385,
34
+ "local_sha256": "9ca408753dfbe4c2103016e9694be0e5033ed517459bf8a7cef2cb56164fa9a7",
35
+ "remote_lfs_sha256": "9ca408753dfbe4c2103016e9694be0e5033ed517459bf8a7cef2cb56164fa9a7",
36
+ "remote_lfs_size": 11775385,
37
+ "verified": true,
38
+ "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/mlp256-seed410102-9ca408753dfb.pt"
39
+ },
40
+ {
41
+ "name": "squarecnn32-seed410101",
42
+ "remote_path": "comparisons/20260909T235853Z/squarecnn32-seed410101-e2a8946b1405.pt",
43
+ "bytes": 14001446,
44
+ "local_sha256": "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1",
45
+ "remote_lfs_sha256": "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1",
46
+ "remote_lfs_size": 14001446,
47
+ "verified": true,
48
+ "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/squarecnn32-seed410101-e2a8946b1405.pt"
49
+ },
50
+ {
51
+ "name": "squarecnn32-seed410102",
52
+ "remote_path": "comparisons/20260909T235853Z/squarecnn32-seed410102-d3e8eb4f1e75.pt",
53
+ "bytes": 14001446,
54
+ "local_sha256": "d3e8eb4f1e75d18bbeb28bef1b46c6c3f84b1ab6f009e63cc950a88f647c8025",
55
+ "remote_lfs_sha256": "d3e8eb4f1e75d18bbeb28bef1b46c6c3f84b1ab6f009e63cc950a88f647c8025",
56
+ "remote_lfs_size": 14001446,
57
+ "verified": true,
58
+ "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/squarecnn32-seed410102-d3e8eb4f1e75.pt"
59
+ },
60
+ {
61
+ "name": "hexcnn32-seed410101",
62
+ "remote_path": "comparisons/20260909T235853Z/hexcnn32-seed410101-d6f1c7081cdb.pt",
63
+ "bytes": 14003159,
64
+ "local_sha256": "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5",
65
+ "remote_lfs_sha256": "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5",
66
+ "remote_lfs_size": 14003159,
67
+ "verified": true,
68
+ "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/hexcnn32-seed410101-d6f1c7081cdb.pt"
69
+ },
70
+ {
71
+ "name": "hexcnn32-seed410102",
72
+ "remote_path": "comparisons/20260909T235853Z/hexcnn32-seed410102-c7d5bd9e9ee0.pt",
73
+ "bytes": 14003159,
74
+ "local_sha256": "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d",
75
+ "remote_lfs_sha256": "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d",
76
+ "remote_lfs_size": 14003159,
77
+ "verified": true,
78
+ "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/hexcnn32-seed410102-c7d5bd9e9ee0.pt"
79
+ },
80
+ {
81
+ "name": "querytransformer256x2-seed410101",
82
+ "remote_path": "comparisons/20260909T235853Z/querytransformer256x2-seed410101-454d54089b4a.pt",
83
+ "bytes": 13274255,
84
+ "local_sha256": "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4",
85
+ "remote_lfs_sha256": "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4",
86
+ "remote_lfs_size": 13274255,
87
+ "verified": true,
88
+ "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/querytransformer256x2-seed410101-454d54089b4a.pt"
89
+ },
90
+ {
91
+ "name": "querytransformer256x2-seed410102",
92
+ "remote_path": "comparisons/20260909T235853Z/querytransformer256x2-seed410102-a0a1d4c39231.pt",
93
+ "bytes": 13274191,
94
+ "local_sha256": "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486",
95
+ "remote_lfs_sha256": "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486",
96
+ "remote_lfs_size": 13274191,
97
+ "verified": true,
98
+ "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/querytransformer256x2-seed410102-a0a1d4c39231.pt"
99
+ },
100
+ {
101
+ "name": "querytransformer164x5-seed410101",
102
+ "remote_path": "comparisons/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt",
103
+ "bytes": 13479583,
104
+ "local_sha256": "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99",
105
+ "remote_lfs_sha256": "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99",
106
+ "remote_lfs_size": 13479583,
107
+ "verified": true,
108
+ "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt"
109
+ },
110
+ {
111
+ "name": "querytransformer164x5-seed410102",
112
+ "remote_path": "comparisons/20260909T235853Z/querytransformer164x5-seed410102-02b9f1909248.pt",
113
+ "bytes": 13479583,
114
+ "local_sha256": "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c",
115
+ "remote_lfs_sha256": "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c",
116
+ "remote_lfs_size": 13479583,
117
+ "verified": true,
118
+ "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/querytransformer164x5-seed410102-02b9f1909248.pt"
119
+ }
120
+ ],
121
+ "source_github_commit_provenance": {
122
+ "kind": "inferred_repository_tree_at_backend_freeze",
123
+ "checkpoint_embedded_git_commit": null,
124
+ "basis": "The checkpoints contain config and RNG/training state but no Git revision. This commit was the latest local commit before summary backend_frozen at 2026-09-09T23:58:53Z; committed blobs match every recorded Python backend hash.",
125
+ "verified_backend_tree_sha256": {
126
+ "training/model.py": "1130a6f7e23524c04790adc702b8ecc7a9d92008bb354a398f377735b379dd71",
127
+ "training/native_search.py": "88beacc193f1fcfae082e46d93b3e1c8ca5edf76bf99c2c6298bd3d6cbe359f0",
128
+ "training/query_transformer.py": "9ff230547d8a0f9fd393fe05414a496724b3c514f76839a8f4c956e941f70385",
129
+ "training/search.py": "54587d2fde6bc62a671d9ff22ca2681f6fff521bbc5ed439c9cdd2b9c1786664",
130
+ "training/train.py": "8cbb2ed0b29d35c5101b6d9882cc2e83dc919277a32816b322a059abf97bd648"
131
+ }
132
+ }
133
+ }
config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "kind": "resnet",
3
+ "width": 32,
4
+ "blocks": 2,
5
+ "head": "flat"
6
+ }
final-checkpoint-publication.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema": "gipf-zero-final-checkpoint-publication-v1",
3
+ "repository": "wuschelschulz/gipf-zero",
4
+ "tag": "20260909T235853Z",
5
+ "checkpoint": "checkpoints/overnight/20260909T235853Z/selected-final-cb19bc298969.pt",
6
+ "remote_path": "finalists/20260909T235853Z/selected-final-cb19bc298969.pt",
7
+ "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/finalists/20260909T235853Z/selected-final-cb19bc298969.pt",
8
+ "bytes": 14003543,
9
+ "file_sha256": "cb19bc298969043987463b267b909ca38c7ead84bab3c0b46571b6ef30e6d00f",
10
+ "remote_lfs_sha256": "cb19bc298969043987463b267b909ca38c7ead84bab3c0b46571b6ef30e6d00f",
11
+ "remote_lfs_size": 14003543,
12
+ "remote_lfs_sha256_verified": true,
13
+ "config": {
14
+ "kind": "hexresnet",
15
+ "width": 32,
16
+ "blocks": 2,
17
+ "head": "flat"
18
+ },
19
+ "games": 93916,
20
+ "iteration": 2873,
21
+ "evaluation_pending": true,
22
+ "evaluation_report": "reports/overnight/20260909T235853Z/final-vs-served.json",
23
+ "interpretation": "Frozen final training checkpoint only; evaluation and promotion were pending when this receipt was created. It is not claimed as strongest or served champion.",
24
+ "upload_protocol": {
25
+ "per_process_timeout_seconds": 180,
26
+ "maximum_retries": 1,
27
+ "network_upload_attempts": 1
28
+ },
29
+ "repository_commit": "8682c0c683117b697b3f1c36ff63008acf3295b8"
30
+ }
final-vs-served.json ADDED
@@ -0,0 +1,720 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/checkpoints/overnight/20260909T235853Z/selected-final-cb19bc298969.pt",
3
+ "champion": "/workspace/gipf/checkpoints/overnight/20260909T235853Z/served-champion-d5a7e3bda296.pt",
4
+ "hashes": [
5
+ "cb19bc298969043987463b267b909ca38c7ead84bab3c0b46571b6ef30e6d00f",
6
+ "d5a7e3bda2966a79d22cf34f131976907260e679683d0e2680d935312f41c2f6"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 93916,
11
+ "iteration": 2873,
12
+ "config": {
13
+ "kind": "hexresnet",
14
+ "width": 32,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 43869,
21
+ "iteration": 727,
22
+ "config": {
23
+ "kind": "resnet",
24
+ "width": 32,
25
+ "blocks": 2,
26
+ "head": "flat"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 38,
31
+ "losses": 42,
32
+ "cutoffs": 0,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 0.475,
35
+ "wilson_95_decisive": [
36
+ 0.36924733297407575,
37
+ 0.5830435735027006
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 3526,
41
+ "mean": 251.41775382870108,
42
+ "min": 44,
43
+ "max": 10000
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 3523,
47
+ "mean": 243.38575078058471,
48
+ "min": 41,
49
+ "max": 10000
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 3526,
53
+ "mean": 0.05023197256244182,
54
+ "min": 0.02666553482413292,
55
+ "max": 0.06883102655410767
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 3523,
59
+ "mean": 0.05029253078465464,
60
+ "min": 0.02774936333298683,
61
+ "max": 0.10136080533266068
62
+ },
63
+ "elapsed_seconds": 354.87011195346713,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/checkpoints/overnight/20260909T235853Z/selected-final-cb19bc298969.pt",
66
+ "champion": "/workspace/gipf/checkpoints/overnight/20260909T235853Z/served-champion-d5a7e3bda296.pt",
67
+ "games": 80,
68
+ "batch": 1,
69
+ "simulations": 10000,
70
+ "seed": 950001,
71
+ "max_ply": 300,
72
+ "seconds": 600,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/final-vs-served.json",
74
+ "device": "cpu",
75
+ "threads": 1,
76
+ "budget_ms": 50.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 0,
81
+ "candidate_color": 1,
82
+ "winner": -1,
83
+ "outcome": "loss",
84
+ "cutoff_reason": null,
85
+ "ply": 44
86
+ },
87
+ {
88
+ "game": 1,
89
+ "candidate_color": -1,
90
+ "winner": -1,
91
+ "outcome": "win",
92
+ "cutoff_reason": null,
93
+ "ply": 66
94
+ },
95
+ {
96
+ "game": 2,
97
+ "candidate_color": 1,
98
+ "winner": -1,
99
+ "outcome": "loss",
100
+ "cutoff_reason": null,
101
+ "ply": 98
102
+ },
103
+ {
104
+ "game": 3,
105
+ "candidate_color": -1,
106
+ "winner": -1,
107
+ "outcome": "win",
108
+ "cutoff_reason": null,
109
+ "ply": 40
110
+ },
111
+ {
112
+ "game": 4,
113
+ "candidate_color": 1,
114
+ "winner": -1,
115
+ "outcome": "loss",
116
+ "cutoff_reason": null,
117
+ "ply": 122
118
+ },
119
+ {
120
+ "game": 5,
121
+ "candidate_color": -1,
122
+ "winner": 1,
123
+ "outcome": "loss",
124
+ "cutoff_reason": null,
125
+ "ply": 37
126
+ },
127
+ {
128
+ "game": 6,
129
+ "candidate_color": 1,
130
+ "winner": 1,
131
+ "outcome": "win",
132
+ "cutoff_reason": null,
133
+ "ply": 39
134
+ },
135
+ {
136
+ "game": 7,
137
+ "candidate_color": -1,
138
+ "winner": -1,
139
+ "outcome": "win",
140
+ "cutoff_reason": null,
141
+ "ply": 48
142
+ },
143
+ {
144
+ "game": 8,
145
+ "candidate_color": 1,
146
+ "winner": 1,
147
+ "outcome": "win",
148
+ "cutoff_reason": null,
149
+ "ply": 87
150
+ },
151
+ {
152
+ "game": 9,
153
+ "candidate_color": -1,
154
+ "winner": 1,
155
+ "outcome": "loss",
156
+ "cutoff_reason": null,
157
+ "ply": 59
158
+ },
159
+ {
160
+ "game": 10,
161
+ "candidate_color": 1,
162
+ "winner": -1,
163
+ "outcome": "loss",
164
+ "cutoff_reason": null,
165
+ "ply": 108
166
+ },
167
+ {
168
+ "game": 11,
169
+ "candidate_color": -1,
170
+ "winner": 1,
171
+ "outcome": "loss",
172
+ "cutoff_reason": null,
173
+ "ply": 67
174
+ },
175
+ {
176
+ "game": 12,
177
+ "candidate_color": 1,
178
+ "winner": 1,
179
+ "outcome": "win",
180
+ "cutoff_reason": null,
181
+ "ply": 75
182
+ },
183
+ {
184
+ "game": 13,
185
+ "candidate_color": -1,
186
+ "winner": 1,
187
+ "outcome": "loss",
188
+ "cutoff_reason": null,
189
+ "ply": 79
190
+ },
191
+ {
192
+ "game": 14,
193
+ "candidate_color": 1,
194
+ "winner": -1,
195
+ "outcome": "loss",
196
+ "cutoff_reason": null,
197
+ "ply": 66
198
+ },
199
+ {
200
+ "game": 15,
201
+ "candidate_color": -1,
202
+ "winner": 1,
203
+ "outcome": "loss",
204
+ "cutoff_reason": null,
205
+ "ply": 111
206
+ },
207
+ {
208
+ "game": 16,
209
+ "candidate_color": 1,
210
+ "winner": -1,
211
+ "outcome": "loss",
212
+ "cutoff_reason": null,
213
+ "ply": 102
214
+ },
215
+ {
216
+ "game": 17,
217
+ "candidate_color": -1,
218
+ "winner": 1,
219
+ "outcome": "loss",
220
+ "cutoff_reason": null,
221
+ "ply": 49
222
+ },
223
+ {
224
+ "game": 18,
225
+ "candidate_color": 1,
226
+ "winner": -1,
227
+ "outcome": "loss",
228
+ "cutoff_reason": null,
229
+ "ply": 92
230
+ },
231
+ {
232
+ "game": 19,
233
+ "candidate_color": -1,
234
+ "winner": -1,
235
+ "outcome": "win",
236
+ "cutoff_reason": null,
237
+ "ply": 86
238
+ },
239
+ {
240
+ "game": 20,
241
+ "candidate_color": 1,
242
+ "winner": -1,
243
+ "outcome": "loss",
244
+ "cutoff_reason": null,
245
+ "ply": 96
246
+ },
247
+ {
248
+ "game": 21,
249
+ "candidate_color": -1,
250
+ "winner": -1,
251
+ "outcome": "win",
252
+ "cutoff_reason": null,
253
+ "ply": 84
254
+ },
255
+ {
256
+ "game": 22,
257
+ "candidate_color": 1,
258
+ "winner": -1,
259
+ "outcome": "loss",
260
+ "cutoff_reason": null,
261
+ "ply": 118
262
+ },
263
+ {
264
+ "game": 23,
265
+ "candidate_color": -1,
266
+ "winner": 1,
267
+ "outcome": "loss",
268
+ "cutoff_reason": null,
269
+ "ply": 283
270
+ },
271
+ {
272
+ "game": 24,
273
+ "candidate_color": 1,
274
+ "winner": -1,
275
+ "outcome": "loss",
276
+ "cutoff_reason": null,
277
+ "ply": 46
278
+ },
279
+ {
280
+ "game": 25,
281
+ "candidate_color": -1,
282
+ "winner": 1,
283
+ "outcome": "loss",
284
+ "cutoff_reason": null,
285
+ "ply": 79
286
+ },
287
+ {
288
+ "game": 26,
289
+ "candidate_color": 1,
290
+ "winner": -1,
291
+ "outcome": "loss",
292
+ "cutoff_reason": null,
293
+ "ply": 66
294
+ },
295
+ {
296
+ "game": 27,
297
+ "candidate_color": -1,
298
+ "winner": 1,
299
+ "outcome": "loss",
300
+ "cutoff_reason": null,
301
+ "ply": 63
302
+ },
303
+ {
304
+ "game": 28,
305
+ "candidate_color": 1,
306
+ "winner": 1,
307
+ "outcome": "win",
308
+ "cutoff_reason": null,
309
+ "ply": 117
310
+ },
311
+ {
312
+ "game": 29,
313
+ "candidate_color": -1,
314
+ "winner": 1,
315
+ "outcome": "loss",
316
+ "cutoff_reason": null,
317
+ "ply": 81
318
+ },
319
+ {
320
+ "game": 30,
321
+ "candidate_color": 1,
322
+ "winner": 1,
323
+ "outcome": "win",
324
+ "cutoff_reason": null,
325
+ "ply": 87
326
+ },
327
+ {
328
+ "game": 31,
329
+ "candidate_color": -1,
330
+ "winner": 1,
331
+ "outcome": "loss",
332
+ "cutoff_reason": null,
333
+ "ply": 97
334
+ },
335
+ {
336
+ "game": 32,
337
+ "candidate_color": 1,
338
+ "winner": -1,
339
+ "outcome": "loss",
340
+ "cutoff_reason": null,
341
+ "ply": 48
342
+ },
343
+ {
344
+ "game": 33,
345
+ "candidate_color": -1,
346
+ "winner": -1,
347
+ "outcome": "win",
348
+ "cutoff_reason": null,
349
+ "ply": 64
350
+ },
351
+ {
352
+ "game": 34,
353
+ "candidate_color": 1,
354
+ "winner": 1,
355
+ "outcome": "win",
356
+ "cutoff_reason": null,
357
+ "ply": 67
358
+ },
359
+ {
360
+ "game": 35,
361
+ "candidate_color": -1,
362
+ "winner": -1,
363
+ "outcome": "win",
364
+ "cutoff_reason": null,
365
+ "ply": 98
366
+ },
367
+ {
368
+ "game": 36,
369
+ "candidate_color": 1,
370
+ "winner": 1,
371
+ "outcome": "win",
372
+ "cutoff_reason": null,
373
+ "ply": 67
374
+ },
375
+ {
376
+ "game": 37,
377
+ "candidate_color": -1,
378
+ "winner": -1,
379
+ "outcome": "win",
380
+ "cutoff_reason": null,
381
+ "ply": 66
382
+ },
383
+ {
384
+ "game": 38,
385
+ "candidate_color": 1,
386
+ "winner": 1,
387
+ "outcome": "win",
388
+ "cutoff_reason": null,
389
+ "ply": 93
390
+ },
391
+ {
392
+ "game": 39,
393
+ "candidate_color": -1,
394
+ "winner": 1,
395
+ "outcome": "loss",
396
+ "cutoff_reason": null,
397
+ "ply": 73
398
+ },
399
+ {
400
+ "game": 40,
401
+ "candidate_color": 1,
402
+ "winner": 1,
403
+ "outcome": "win",
404
+ "cutoff_reason": null,
405
+ "ply": 77
406
+ },
407
+ {
408
+ "game": 41,
409
+ "candidate_color": -1,
410
+ "winner": 1,
411
+ "outcome": "loss",
412
+ "cutoff_reason": null,
413
+ "ply": 59
414
+ },
415
+ {
416
+ "game": 42,
417
+ "candidate_color": 1,
418
+ "winner": -1,
419
+ "outcome": "loss",
420
+ "cutoff_reason": null,
421
+ "ply": 66
422
+ },
423
+ {
424
+ "game": 43,
425
+ "candidate_color": -1,
426
+ "winner": -1,
427
+ "outcome": "win",
428
+ "cutoff_reason": null,
429
+ "ply": 68
430
+ },
431
+ {
432
+ "game": 44,
433
+ "candidate_color": 1,
434
+ "winner": -1,
435
+ "outcome": "loss",
436
+ "cutoff_reason": null,
437
+ "ply": 42
438
+ },
439
+ {
440
+ "game": 45,
441
+ "candidate_color": -1,
442
+ "winner": -1,
443
+ "outcome": "win",
444
+ "cutoff_reason": null,
445
+ "ply": 80
446
+ },
447
+ {
448
+ "game": 46,
449
+ "candidate_color": 1,
450
+ "winner": 1,
451
+ "outcome": "win",
452
+ "cutoff_reason": null,
453
+ "ply": 65
454
+ },
455
+ {
456
+ "game": 47,
457
+ "candidate_color": -1,
458
+ "winner": 1,
459
+ "outcome": "loss",
460
+ "cutoff_reason": null,
461
+ "ply": 73
462
+ },
463
+ {
464
+ "game": 48,
465
+ "candidate_color": 1,
466
+ "winner": 1,
467
+ "outcome": "win",
468
+ "cutoff_reason": null,
469
+ "ply": 37
470
+ },
471
+ {
472
+ "game": 49,
473
+ "candidate_color": -1,
474
+ "winner": 1,
475
+ "outcome": "loss",
476
+ "cutoff_reason": null,
477
+ "ply": 159
478
+ },
479
+ {
480
+ "game": 50,
481
+ "candidate_color": 1,
482
+ "winner": -1,
483
+ "outcome": "loss",
484
+ "cutoff_reason": null,
485
+ "ply": 76
486
+ },
487
+ {
488
+ "game": 51,
489
+ "candidate_color": -1,
490
+ "winner": -1,
491
+ "outcome": "win",
492
+ "cutoff_reason": null,
493
+ "ply": 62
494
+ },
495
+ {
496
+ "game": 52,
497
+ "candidate_color": 1,
498
+ "winner": -1,
499
+ "outcome": "loss",
500
+ "cutoff_reason": null,
501
+ "ply": 66
502
+ },
503
+ {
504
+ "game": 53,
505
+ "candidate_color": -1,
506
+ "winner": -1,
507
+ "outcome": "win",
508
+ "cutoff_reason": null,
509
+ "ply": 80
510
+ },
511
+ {
512
+ "game": 54,
513
+ "candidate_color": 1,
514
+ "winner": -1,
515
+ "outcome": "loss",
516
+ "cutoff_reason": null,
517
+ "ply": 70
518
+ },
519
+ {
520
+ "game": 55,
521
+ "candidate_color": -1,
522
+ "winner": -1,
523
+ "outcome": "win",
524
+ "cutoff_reason": null,
525
+ "ply": 66
526
+ },
527
+ {
528
+ "game": 56,
529
+ "candidate_color": 1,
530
+ "winner": 1,
531
+ "outcome": "win",
532
+ "cutoff_reason": null,
533
+ "ply": 41
534
+ },
535
+ {
536
+ "game": 57,
537
+ "candidate_color": -1,
538
+ "winner": -1,
539
+ "outcome": "win",
540
+ "cutoff_reason": null,
541
+ "ply": 62
542
+ },
543
+ {
544
+ "game": 58,
545
+ "candidate_color": 1,
546
+ "winner": -1,
547
+ "outcome": "loss",
548
+ "cutoff_reason": null,
549
+ "ply": 96
550
+ },
551
+ {
552
+ "game": 59,
553
+ "candidate_color": -1,
554
+ "winner": -1,
555
+ "outcome": "win",
556
+ "cutoff_reason": null,
557
+ "ply": 72
558
+ },
559
+ {
560
+ "game": 60,
561
+ "candidate_color": 1,
562
+ "winner": 1,
563
+ "outcome": "win",
564
+ "cutoff_reason": null,
565
+ "ply": 79
566
+ },
567
+ {
568
+ "game": 61,
569
+ "candidate_color": -1,
570
+ "winner": 1,
571
+ "outcome": "loss",
572
+ "cutoff_reason": null,
573
+ "ply": 71
574
+ },
575
+ {
576
+ "game": 62,
577
+ "candidate_color": 1,
578
+ "winner": 1,
579
+ "outcome": "win",
580
+ "cutoff_reason": null,
581
+ "ply": 79
582
+ },
583
+ {
584
+ "game": 63,
585
+ "candidate_color": -1,
586
+ "winner": 1,
587
+ "outcome": "loss",
588
+ "cutoff_reason": null,
589
+ "ply": 73
590
+ },
591
+ {
592
+ "game": 64,
593
+ "candidate_color": 1,
594
+ "winner": 1,
595
+ "outcome": "win",
596
+ "cutoff_reason": null,
597
+ "ply": 89
598
+ },
599
+ {
600
+ "game": 65,
601
+ "candidate_color": -1,
602
+ "winner": 1,
603
+ "outcome": "loss",
604
+ "cutoff_reason": null,
605
+ "ply": 35
606
+ },
607
+ {
608
+ "game": 66,
609
+ "candidate_color": 1,
610
+ "winner": 1,
611
+ "outcome": "win",
612
+ "cutoff_reason": null,
613
+ "ply": 59
614
+ },
615
+ {
616
+ "game": 67,
617
+ "candidate_color": -1,
618
+ "winner": 1,
619
+ "outcome": "loss",
620
+ "cutoff_reason": null,
621
+ "ply": 57
622
+ },
623
+ {
624
+ "game": 68,
625
+ "candidate_color": 1,
626
+ "winner": -1,
627
+ "outcome": "loss",
628
+ "cutoff_reason": null,
629
+ "ply": 56
630
+ },
631
+ {
632
+ "game": 69,
633
+ "candidate_color": -1,
634
+ "winner": -1,
635
+ "outcome": "win",
636
+ "cutoff_reason": null,
637
+ "ply": 84
638
+ },
639
+ {
640
+ "game": 70,
641
+ "candidate_color": 1,
642
+ "winner": -1,
643
+ "outcome": "loss",
644
+ "cutoff_reason": null,
645
+ "ply": 84
646
+ },
647
+ {
648
+ "game": 71,
649
+ "candidate_color": -1,
650
+ "winner": -1,
651
+ "outcome": "win",
652
+ "cutoff_reason": null,
653
+ "ply": 88
654
+ },
655
+ {
656
+ "game": 72,
657
+ "candidate_color": 1,
658
+ "winner": 1,
659
+ "outcome": "win",
660
+ "cutoff_reason": null,
661
+ "ply": 49
662
+ },
663
+ {
664
+ "game": 73,
665
+ "candidate_color": -1,
666
+ "winner": -1,
667
+ "outcome": "win",
668
+ "cutoff_reason": null,
669
+ "ply": 82
670
+ },
671
+ {
672
+ "game": 74,
673
+ "candidate_color": 1,
674
+ "winner": 1,
675
+ "outcome": "win",
676
+ "cutoff_reason": null,
677
+ "ply": 89
678
+ },
679
+ {
680
+ "game": 75,
681
+ "candidate_color": -1,
682
+ "winner": -1,
683
+ "outcome": "win",
684
+ "cutoff_reason": null,
685
+ "ply": 80
686
+ },
687
+ {
688
+ "game": 76,
689
+ "candidate_color": 1,
690
+ "winner": -1,
691
+ "outcome": "loss",
692
+ "cutoff_reason": null,
693
+ "ply": 38
694
+ },
695
+ {
696
+ "game": 77,
697
+ "candidate_color": -1,
698
+ "winner": 1,
699
+ "outcome": "loss",
700
+ "cutoff_reason": null,
701
+ "ply": 115
702
+ },
703
+ {
704
+ "game": 78,
705
+ "candidate_color": 1,
706
+ "winner": 1,
707
+ "outcome": "win",
708
+ "cutoff_reason": null,
709
+ "ply": 79
710
+ },
711
+ {
712
+ "game": 79,
713
+ "candidate_color": -1,
714
+ "winner": 1,
715
+ "outcome": "loss",
716
+ "cutoff_reason": null,
717
+ "ply": 73
718
+ }
719
+ ]
720
+ }
final-vs-served.jsonl ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"game": 0, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 44}
2
+ {"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 66}
3
+ {"game": 2, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 98}
4
+ {"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 40}
5
+ {"game": 4, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 122}
6
+ {"game": 5, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 37}
7
+ {"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 39}
8
+ {"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 48}
9
+ {"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 87}
10
+ {"game": 9, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 59}
11
+ {"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 108}
12
+ {"game": 11, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 67}
13
+ {"game": 12, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 75}
14
+ {"game": 13, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 79}
15
+ {"game": 14, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 66}
16
+ {"game": 15, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 111}
17
+ {"game": 16, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 102}
18
+ {"game": 17, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 49}
19
+ {"game": 18, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 92}
20
+ {"game": 19, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 86}
21
+ {"game": 20, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 96}
22
+ {"game": 21, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 84}
23
+ {"game": 22, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 118}
24
+ {"game": 23, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 283}
25
+ {"game": 24, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 46}
26
+ {"game": 25, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 79}
27
+ {"game": 26, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 66}
28
+ {"game": 27, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 63}
29
+ {"game": 28, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 117}
30
+ {"game": 29, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 81}
31
+ {"game": 30, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 87}
32
+ {"game": 31, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 97}
33
+ {"game": 32, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 48}
34
+ {"game": 33, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 64}
35
+ {"game": 34, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 67}
36
+ {"game": 35, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 98}
37
+ {"game": 36, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 67}
38
+ {"game": 37, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 66}
39
+ {"game": 38, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 93}
40
+ {"game": 39, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 73}
41
+ {"game": 40, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 77}
42
+ {"game": 41, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 59}
43
+ {"game": 42, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 66}
44
+ {"game": 43, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 68}
45
+ {"game": 44, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 42}
46
+ {"game": 45, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 80}
47
+ {"game": 46, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 65}
48
+ {"game": 47, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 73}
49
+ {"game": 48, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 37}
50
+ {"game": 49, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 159}
51
+ {"game": 50, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 76}
52
+ {"game": 51, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 62}
53
+ {"game": 52, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 66}
54
+ {"game": 53, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 80}
55
+ {"game": 54, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 70}
56
+ {"game": 55, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 66}
57
+ {"game": 56, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41}
58
+ {"game": 57, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 62}
59
+ {"game": 58, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 96}
60
+ {"game": 59, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 72}
61
+ {"game": 60, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 79}
62
+ {"game": 61, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 71}
63
+ {"game": 62, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 79}
64
+ {"game": 63, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 73}
65
+ {"game": 64, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 89}
66
+ {"game": 65, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 35}
67
+ {"game": 66, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 59}
68
+ {"game": 67, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 57}
69
+ {"game": 68, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 56}
70
+ {"game": 69, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 84}
71
+ {"game": 70, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 84}
72
+ {"game": 71, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 88}
73
+ {"game": 72, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 49}
74
+ {"game": 73, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 82}
75
+ {"game": 74, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 89}
76
+ {"game": 75, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 80}
77
+ {"game": 76, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 38}
78
+ {"game": 77, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 115}
79
+ {"game": 78, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 79}
80
+ {"game": 79, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 73}
hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt",
3
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt",
4
+ "hashes": [
5
+ "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5",
6
+ "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 7877,
11
+ "iteration": 242,
12
+ "config": {
13
+ "kind": "hexresnet",
14
+ "width": 32,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 582,
21
+ "iteration": 18,
22
+ "config": {
23
+ "kind": "transformer",
24
+ "width": 164,
25
+ "blocks": 5,
26
+ "head": "query"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 8,
31
+ "losses": 0,
32
+ "cutoffs": 4,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 1.0,
35
+ "wilson_95_decisive": [
36
+ 0.6755924351161198,
37
+ 1.0
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 925,
41
+ "mean": 197.71243243243242,
42
+ "min": 49,
43
+ "max": 10000
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 911,
47
+ "mean": 16.644346871569702,
48
+ "min": 6,
49
+ "max": 21
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 925,
53
+ "mean": 0.05026920535274454,
54
+ "min": 0.030910689383745193,
55
+ "max": 0.059041790664196014
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 911,
59
+ "mean": 0.0512864647921825,
60
+ "min": 0.05002184584736824,
61
+ "max": 0.06940874457359314
62
+ },
63
+ "elapsed_seconds": 93.33740724623203,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt",
66
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt",
67
+ "games": 12,
68
+ "batch": 1,
69
+ "simulations": 10000,
70
+ "seed": 1380101,
71
+ "max_ply": 300,
72
+ "seconds": 180,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json",
74
+ "device": "cpu",
75
+ "threads": 1,
76
+ "budget_ms": 50.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 0,
81
+ "candidate_color": 1,
82
+ "winner": 1,
83
+ "outcome": "win",
84
+ "cutoff_reason": null,
85
+ "ply": 71
86
+ },
87
+ {
88
+ "game": 1,
89
+ "candidate_color": -1,
90
+ "winner": -1,
91
+ "outcome": "win",
92
+ "cutoff_reason": null,
93
+ "ply": 50
94
+ },
95
+ {
96
+ "game": 2,
97
+ "candidate_color": 1,
98
+ "winner": 1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 19
102
+ },
103
+ {
104
+ "game": 3,
105
+ "candidate_color": -1,
106
+ "winner": 0,
107
+ "outcome": "cutoff",
108
+ "cutoff_reason": "insertion",
109
+ "ply": 300
110
+ },
111
+ {
112
+ "game": 4,
113
+ "candidate_color": 1,
114
+ "winner": 0,
115
+ "outcome": "cutoff",
116
+ "cutoff_reason": "insertion",
117
+ "ply": 300
118
+ },
119
+ {
120
+ "game": 5,
121
+ "candidate_color": -1,
122
+ "winner": -1,
123
+ "outcome": "win",
124
+ "cutoff_reason": null,
125
+ "ply": 20
126
+ },
127
+ {
128
+ "game": 6,
129
+ "candidate_color": 1,
130
+ "winner": 0,
131
+ "outcome": "cutoff",
132
+ "cutoff_reason": "insertion",
133
+ "ply": 300
134
+ },
135
+ {
136
+ "game": 7,
137
+ "candidate_color": -1,
138
+ "winner": 0,
139
+ "outcome": "cutoff",
140
+ "cutoff_reason": "insertion",
141
+ "ply": 300
142
+ },
143
+ {
144
+ "game": 8,
145
+ "candidate_color": 1,
146
+ "winner": 1,
147
+ "outcome": "win",
148
+ "cutoff_reason": null,
149
+ "ply": 37
150
+ },
151
+ {
152
+ "game": 9,
153
+ "candidate_color": -1,
154
+ "winner": -1,
155
+ "outcome": "win",
156
+ "cutoff_reason": null,
157
+ "ply": 54
158
+ },
159
+ {
160
+ "game": 10,
161
+ "candidate_color": 1,
162
+ "winner": 1,
163
+ "outcome": "win",
164
+ "cutoff_reason": null,
165
+ "ply": 45
166
+ },
167
+ {
168
+ "game": 11,
169
+ "candidate_color": -1,
170
+ "winner": -1,
171
+ "outcome": "win",
172
+ "cutoff_reason": null,
173
+ "ply": 34
174
+ }
175
+ ]
176
+ }
hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 71}
2
+ {"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 50}
3
+ {"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 19}
4
+ {"game": 3, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
5
+ {"game": 4, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
6
+ {"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 20}
7
+ {"game": 6, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
8
+ {"game": 7, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
9
+ {"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 37}
10
+ {"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 54}
11
+ {"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 45}
12
+ {"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 34}
hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json ADDED
@@ -0,0 +1,128 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt",
3
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt",
4
+ "hashes": [
5
+ "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5",
6
+ "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 7877,
11
+ "iteration": 242,
12
+ "config": {
13
+ "kind": "hexresnet",
14
+ "width": 32,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 582,
21
+ "iteration": 18,
22
+ "config": {
23
+ "kind": "transformer",
24
+ "width": 164,
25
+ "blocks": 5,
26
+ "head": "query"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 6,
31
+ "losses": 0,
32
+ "cutoffs": 0,
33
+ "unfinished": 6,
34
+ "win_rate_decisive": 1.0,
35
+ "wilson_95_decisive": [
36
+ 0.6096657120978348,
37
+ 1.0
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 1184,
41
+ "mean": 32.0,
42
+ "min": 32,
43
+ "max": 32
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 1172,
47
+ "mean": 31.907849829351537,
48
+ "min": 5,
49
+ "max": 32
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 344,
53
+ "mean": 0.03558920219887134,
54
+ "min": 0.026320748031139374,
55
+ "max": 0.3319869302213192
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 344,
59
+ "mean": 0.09499198118181423,
60
+ "min": 0.020720209926366806,
61
+ "max": 0.18930847942829132
62
+ },
63
+ "elapsed_seconds": 45.010730154812336,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt",
66
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt",
67
+ "games": 12,
68
+ "batch": 64,
69
+ "simulations": 32,
70
+ "seed": 1380101,
71
+ "max_ply": 300,
72
+ "seconds": 45,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json",
74
+ "device": "cuda",
75
+ "threads": 2,
76
+ "budget_ms": 0.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 11,
81
+ "candidate_color": -1,
82
+ "winner": -1,
83
+ "outcome": "win",
84
+ "cutoff_reason": null,
85
+ "ply": 28
86
+ },
87
+ {
88
+ "game": 3,
89
+ "candidate_color": -1,
90
+ "winner": -1,
91
+ "outcome": "win",
92
+ "cutoff_reason": null,
93
+ "ply": 32
94
+ },
95
+ {
96
+ "game": 8,
97
+ "candidate_color": 1,
98
+ "winner": 1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 41
102
+ },
103
+ {
104
+ "game": 0,
105
+ "candidate_color": 1,
106
+ "winner": 1,
107
+ "outcome": "win",
108
+ "cutoff_reason": null,
109
+ "ply": 47
110
+ },
111
+ {
112
+ "game": 9,
113
+ "candidate_color": -1,
114
+ "winner": -1,
115
+ "outcome": "win",
116
+ "cutoff_reason": null,
117
+ "ply": 50
118
+ },
119
+ {
120
+ "game": 10,
121
+ "candidate_color": 1,
122
+ "winner": 1,
123
+ "outcome": "win",
124
+ "cutoff_reason": null,
125
+ "ply": 71
126
+ }
127
+ ]
128
+ }
hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.jsonl ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 28}
2
+ {"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 32}
3
+ {"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41}
4
+ {"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 47}
5
+ {"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 50}
6
+ {"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 71}
hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "checkpoints/overnight/20260909T235853Z/hexcnn32-seed410101-d6f1c7081cdb.pt",
3
+ "champion": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt",
4
+ "hashes": [
5
+ "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5",
6
+ "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 7877,
11
+ "iteration": 242,
12
+ "config": {
13
+ "kind": "hexresnet",
14
+ "width": 32,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 582,
21
+ "iteration": 18,
22
+ "config": {
23
+ "kind": "transformer",
24
+ "width": 164,
25
+ "blocks": 5,
26
+ "head": "query"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 6,
31
+ "losses": 0,
32
+ "cutoffs": 6,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 1.0,
35
+ "wilson_95_decisive": [
36
+ 0.6096657120978348,
37
+ 1.0
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 1259,
41
+ "mean": 32.0,
42
+ "min": 32,
43
+ "max": 32
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 1247,
47
+ "mean": 32.0,
48
+ "min": 32,
49
+ "max": 32
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 369,
53
+ "mean": 0.033492098312552385,
54
+ "min": 0.0261453315615654,
55
+ "max": 0.28923046216368675
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 370,
59
+ "mean": 0.08941787915253961,
60
+ "min": 0.05584017187356949,
61
+ "max": 0.17711329832673073
62
+ },
63
+ "elapsed_seconds": 45.535510551184416,
64
+ "config": {
65
+ "candidate": "checkpoints/overnight/20260909T235853Z/hexcnn32-seed410101-d6f1c7081cdb.pt",
66
+ "champion": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt",
67
+ "games": 12,
68
+ "batch": 64,
69
+ "simulations": 32,
70
+ "seed": 1380101,
71
+ "max_ply": 300,
72
+ "seconds": 180,
73
+ "output": "reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json",
74
+ "device": "cuda",
75
+ "threads": 1,
76
+ "budget_ms": 0.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 11,
81
+ "candidate_color": -1,
82
+ "winner": -1,
83
+ "outcome": "win",
84
+ "cutoff_reason": null,
85
+ "ply": 28
86
+ },
87
+ {
88
+ "game": 3,
89
+ "candidate_color": -1,
90
+ "winner": -1,
91
+ "outcome": "win",
92
+ "cutoff_reason": null,
93
+ "ply": 32
94
+ },
95
+ {
96
+ "game": 8,
97
+ "candidate_color": 1,
98
+ "winner": 1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 41
102
+ },
103
+ {
104
+ "game": 0,
105
+ "candidate_color": 1,
106
+ "winner": 1,
107
+ "outcome": "win",
108
+ "cutoff_reason": null,
109
+ "ply": 47
110
+ },
111
+ {
112
+ "game": 9,
113
+ "candidate_color": -1,
114
+ "winner": -1,
115
+ "outcome": "win",
116
+ "cutoff_reason": null,
117
+ "ply": 50
118
+ },
119
+ {
120
+ "game": 10,
121
+ "candidate_color": 1,
122
+ "winner": 1,
123
+ "outcome": "win",
124
+ "cutoff_reason": null,
125
+ "ply": 71
126
+ },
127
+ {
128
+ "game": 4,
129
+ "candidate_color": 1,
130
+ "winner": 0,
131
+ "outcome": "cutoff",
132
+ "cutoff_reason": "insertion",
133
+ "ply": 300
134
+ },
135
+ {
136
+ "game": 1,
137
+ "candidate_color": -1,
138
+ "winner": 0,
139
+ "outcome": "cutoff",
140
+ "cutoff_reason": "insertion",
141
+ "ply": 300
142
+ },
143
+ {
144
+ "game": 2,
145
+ "candidate_color": 1,
146
+ "winner": 0,
147
+ "outcome": "cutoff",
148
+ "cutoff_reason": "insertion",
149
+ "ply": 300
150
+ },
151
+ {
152
+ "game": 5,
153
+ "candidate_color": -1,
154
+ "winner": 0,
155
+ "outcome": "cutoff",
156
+ "cutoff_reason": "insertion",
157
+ "ply": 300
158
+ },
159
+ {
160
+ "game": 7,
161
+ "candidate_color": -1,
162
+ "winner": 0,
163
+ "outcome": "cutoff",
164
+ "cutoff_reason": "insertion",
165
+ "ply": 300
166
+ },
167
+ {
168
+ "game": 6,
169
+ "candidate_color": 1,
170
+ "winner": 0,
171
+ "outcome": "cutoff",
172
+ "cutoff_reason": "insertion",
173
+ "ply": 300
174
+ }
175
+ ]
176
+ }
hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.jsonl ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 28}
2
+ {"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 32}
3
+ {"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41}
4
+ {"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 47}
5
+ {"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 50}
6
+ {"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 71}
7
+ {"game": 4, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
8
+ {"game": 1, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
9
+ {"game": 2, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
10
+ {"game": 5, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
11
+ {"game": 7, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
12
+ {"game": 6, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt",
3
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt",
4
+ "hashes": [
5
+ "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5",
6
+ "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 7877,
11
+ "iteration": 242,
12
+ "config": {
13
+ "kind": "hexresnet",
14
+ "width": 32,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 524,
21
+ "iteration": 15,
22
+ "config": {
23
+ "kind": "transformer",
24
+ "width": 256,
25
+ "blocks": 2,
26
+ "head": "query"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 12,
31
+ "losses": 0,
32
+ "cutoffs": 0,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 1.0,
35
+ "wilson_95_decisive": [
36
+ 0.7575059933447593,
37
+ 1.0
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 213,
41
+ "mean": 1017.1032863849765,
42
+ "min": 61,
43
+ "max": 10000
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 196,
47
+ "mean": 22.714285714285715,
48
+ "min": 14,
49
+ "max": 27
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 213,
53
+ "mean": 0.049378569102343256,
54
+ "min": 0.029349185526371002,
55
+ "max": 0.05088867247104645
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 196,
59
+ "mean": 0.051197532659434546,
60
+ "min": 0.05004933848977089,
61
+ "max": 0.055476345121860504
62
+ },
63
+ "elapsed_seconds": 20.587458580732346,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt",
66
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt",
67
+ "games": 12,
68
+ "batch": 1,
69
+ "simulations": 10000,
70
+ "seed": 1380101,
71
+ "max_ply": 300,
72
+ "seconds": 180,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json",
74
+ "device": "cpu",
75
+ "threads": 1,
76
+ "budget_ms": 50.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 0,
81
+ "candidate_color": 1,
82
+ "winner": 1,
83
+ "outcome": "win",
84
+ "cutoff_reason": null,
85
+ "ply": 45
86
+ },
87
+ {
88
+ "game": 1,
89
+ "candidate_color": -1,
90
+ "winner": -1,
91
+ "outcome": "win",
92
+ "cutoff_reason": null,
93
+ "ply": 20
94
+ },
95
+ {
96
+ "game": 2,
97
+ "candidate_color": 1,
98
+ "winner": 1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 33
102
+ },
103
+ {
104
+ "game": 3,
105
+ "candidate_color": -1,
106
+ "winner": -1,
107
+ "outcome": "win",
108
+ "cutoff_reason": null,
109
+ "ply": 22
110
+ },
111
+ {
112
+ "game": 4,
113
+ "candidate_color": 1,
114
+ "winner": 1,
115
+ "outcome": "win",
116
+ "cutoff_reason": null,
117
+ "ply": 55
118
+ },
119
+ {
120
+ "game": 5,
121
+ "candidate_color": -1,
122
+ "winner": -1,
123
+ "outcome": "win",
124
+ "cutoff_reason": null,
125
+ "ply": 20
126
+ },
127
+ {
128
+ "game": 6,
129
+ "candidate_color": 1,
130
+ "winner": 1,
131
+ "outcome": "win",
132
+ "cutoff_reason": null,
133
+ "ply": 23
134
+ },
135
+ {
136
+ "game": 7,
137
+ "candidate_color": -1,
138
+ "winner": -1,
139
+ "outcome": "win",
140
+ "cutoff_reason": null,
141
+ "ply": 40
142
+ },
143
+ {
144
+ "game": 8,
145
+ "candidate_color": 1,
146
+ "winner": 1,
147
+ "outcome": "win",
148
+ "cutoff_reason": null,
149
+ "ply": 49
150
+ },
151
+ {
152
+ "game": 9,
153
+ "candidate_color": -1,
154
+ "winner": -1,
155
+ "outcome": "win",
156
+ "cutoff_reason": null,
157
+ "ply": 28
158
+ },
159
+ {
160
+ "game": 10,
161
+ "candidate_color": 1,
162
+ "winner": 1,
163
+ "outcome": "win",
164
+ "cutoff_reason": null,
165
+ "ply": 23
166
+ },
167
+ {
168
+ "game": 11,
169
+ "candidate_color": -1,
170
+ "winner": -1,
171
+ "outcome": "win",
172
+ "cutoff_reason": null,
173
+ "ply": 28
174
+ }
175
+ ]
176
+ }
hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 45}
2
+ {"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 20}
3
+ {"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 33}
4
+ {"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 22}
5
+ {"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 55}
6
+ {"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 20}
7
+ {"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 23}
8
+ {"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 40}
9
+ {"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 49}
10
+ {"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 28}
11
+ {"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 23}
12
+ {"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 28}
hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt",
3
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt",
4
+ "hashes": [
5
+ "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5",
6
+ "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 7877,
11
+ "iteration": 242,
12
+ "config": {
13
+ "kind": "hexresnet",
14
+ "width": 32,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 524,
21
+ "iteration": 15,
22
+ "config": {
23
+ "kind": "transformer",
24
+ "width": 256,
25
+ "blocks": 2,
26
+ "head": "query"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 9,
31
+ "losses": 2,
32
+ "cutoffs": 1,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 0.8181818181818182,
35
+ "wilson_95_decisive": [
36
+ 0.523019438039156,
37
+ 0.9486323102539151
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 359,
41
+ "mean": 32.0,
42
+ "min": 32,
43
+ "max": 32
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 352,
47
+ "mean": 32.0,
48
+ "min": 32,
49
+ "max": 32
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 210,
53
+ "mean": 0.035037198857892124,
54
+ "min": 0.002302609384059906,
55
+ "max": 0.31664491817355156
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 210,
59
+ "mean": 0.051772340440324374,
60
+ "min": 0.034844595938920975,
61
+ "max": 0.17626779153943062
62
+ },
63
+ "elapsed_seconds": 18.266604859381914,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt",
66
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt",
67
+ "games": 12,
68
+ "batch": 64,
69
+ "simulations": 32,
70
+ "seed": 1380101,
71
+ "max_ply": 300,
72
+ "seconds": 45,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json",
74
+ "device": "cuda",
75
+ "threads": 2,
76
+ "budget_ms": 0.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 9,
81
+ "candidate_color": -1,
82
+ "winner": -1,
83
+ "outcome": "win",
84
+ "cutoff_reason": null,
85
+ "ply": 18
86
+ },
87
+ {
88
+ "game": 1,
89
+ "candidate_color": -1,
90
+ "winner": -1,
91
+ "outcome": "win",
92
+ "cutoff_reason": null,
93
+ "ply": 20
94
+ },
95
+ {
96
+ "game": 11,
97
+ "candidate_color": -1,
98
+ "winner": -1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 20
102
+ },
103
+ {
104
+ "game": 0,
105
+ "candidate_color": 1,
106
+ "winner": -1,
107
+ "outcome": "loss",
108
+ "cutoff_reason": null,
109
+ "ply": 24
110
+ },
111
+ {
112
+ "game": 7,
113
+ "candidate_color": -1,
114
+ "winner": -1,
115
+ "outcome": "win",
116
+ "cutoff_reason": null,
117
+ "ply": 22
118
+ },
119
+ {
120
+ "game": 5,
121
+ "candidate_color": -1,
122
+ "winner": -1,
123
+ "outcome": "win",
124
+ "cutoff_reason": null,
125
+ "ply": 28
126
+ },
127
+ {
128
+ "game": 10,
129
+ "candidate_color": 1,
130
+ "winner": -1,
131
+ "outcome": "loss",
132
+ "cutoff_reason": null,
133
+ "ply": 32
134
+ },
135
+ {
136
+ "game": 2,
137
+ "candidate_color": 1,
138
+ "winner": 1,
139
+ "outcome": "win",
140
+ "cutoff_reason": null,
141
+ "ply": 33
142
+ },
143
+ {
144
+ "game": 3,
145
+ "candidate_color": -1,
146
+ "winner": -1,
147
+ "outcome": "win",
148
+ "cutoff_reason": null,
149
+ "ply": 36
150
+ },
151
+ {
152
+ "game": 6,
153
+ "candidate_color": 1,
154
+ "winner": 1,
155
+ "outcome": "win",
156
+ "cutoff_reason": null,
157
+ "ply": 47
158
+ },
159
+ {
160
+ "game": 4,
161
+ "candidate_color": 1,
162
+ "winner": 1,
163
+ "outcome": "win",
164
+ "cutoff_reason": null,
165
+ "ply": 51
166
+ },
167
+ {
168
+ "game": 8,
169
+ "candidate_color": 1,
170
+ "winner": 0,
171
+ "outcome": "cutoff",
172
+ "cutoff_reason": "insertion",
173
+ "ply": 300
174
+ }
175
+ ]
176
+ }
hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.jsonl ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 18}
2
+ {"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 20}
3
+ {"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 20}
4
+ {"game": 0, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 24}
5
+ {"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 22}
6
+ {"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 28}
7
+ {"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 32}
8
+ {"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 33}
9
+ {"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 36}
10
+ {"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 47}
11
+ {"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 51}
12
+ {"game": 8, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt",
3
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt",
4
+ "hashes": [
5
+ "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d",
6
+ "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 6813,
11
+ "iteration": 209,
12
+ "config": {
13
+ "kind": "hexresnet",
14
+ "width": 32,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 476,
21
+ "iteration": 14,
22
+ "config": {
23
+ "kind": "transformer",
24
+ "width": 164,
25
+ "blocks": 5,
26
+ "head": "query"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 11,
31
+ "losses": 0,
32
+ "cutoffs": 1,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 1.0,
35
+ "wilson_95_decisive": [
36
+ 0.7411670330319684,
37
+ 1.0
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 368,
41
+ "mean": 533.054347826087,
42
+ "min": 58,
43
+ "max": 10000
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 354,
47
+ "mean": 16.86723163841808,
48
+ "min": 8,
49
+ "max": 21
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 368,
53
+ "mean": 0.04981631146865371,
54
+ "min": 0.029736820608377457,
55
+ "max": 0.05099663510918617
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 354,
59
+ "mean": 0.05114389503490453,
60
+ "min": 0.050041306763887405,
61
+ "max": 0.06353601440787315
62
+ },
63
+ "elapsed_seconds": 36.48766961693764,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt",
66
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt",
67
+ "games": 12,
68
+ "batch": 1,
69
+ "simulations": 10000,
70
+ "seed": 1380102,
71
+ "max_ply": 300,
72
+ "seconds": 180,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json",
74
+ "device": "cpu",
75
+ "threads": 1,
76
+ "budget_ms": 50.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 0,
81
+ "candidate_color": 1,
82
+ "winner": 1,
83
+ "outcome": "win",
84
+ "cutoff_reason": null,
85
+ "ply": 39
86
+ },
87
+ {
88
+ "game": 1,
89
+ "candidate_color": -1,
90
+ "winner": -1,
91
+ "outcome": "win",
92
+ "cutoff_reason": null,
93
+ "ply": 30
94
+ },
95
+ {
96
+ "game": 2,
97
+ "candidate_color": 1,
98
+ "winner": 1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 59
102
+ },
103
+ {
104
+ "game": 3,
105
+ "candidate_color": -1,
106
+ "winner": -1,
107
+ "outcome": "win",
108
+ "cutoff_reason": null,
109
+ "ply": 16
110
+ },
111
+ {
112
+ "game": 4,
113
+ "candidate_color": 1,
114
+ "winner": 1,
115
+ "outcome": "win",
116
+ "cutoff_reason": null,
117
+ "ply": 29
118
+ },
119
+ {
120
+ "game": 5,
121
+ "candidate_color": -1,
122
+ "winner": 0,
123
+ "outcome": "cutoff",
124
+ "cutoff_reason": "insertion",
125
+ "ply": 300
126
+ },
127
+ {
128
+ "game": 6,
129
+ "candidate_color": 1,
130
+ "winner": 1,
131
+ "outcome": "win",
132
+ "cutoff_reason": null,
133
+ "ply": 25
134
+ },
135
+ {
136
+ "game": 7,
137
+ "candidate_color": -1,
138
+ "winner": -1,
139
+ "outcome": "win",
140
+ "cutoff_reason": null,
141
+ "ply": 34
142
+ },
143
+ {
144
+ "game": 8,
145
+ "candidate_color": 1,
146
+ "winner": 1,
147
+ "outcome": "win",
148
+ "cutoff_reason": null,
149
+ "ply": 29
150
+ },
151
+ {
152
+ "game": 9,
153
+ "candidate_color": -1,
154
+ "winner": -1,
155
+ "outcome": "win",
156
+ "cutoff_reason": null,
157
+ "ply": 24
158
+ },
159
+ {
160
+ "game": 10,
161
+ "candidate_color": 1,
162
+ "winner": 1,
163
+ "outcome": "win",
164
+ "cutoff_reason": null,
165
+ "ply": 25
166
+ },
167
+ {
168
+ "game": 11,
169
+ "candidate_color": -1,
170
+ "winner": -1,
171
+ "outcome": "win",
172
+ "cutoff_reason": null,
173
+ "ply": 34
174
+ }
175
+ ]
176
+ }
hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 39}
2
+ {"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 30}
3
+ {"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 59}
4
+ {"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 16}
5
+ {"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 29}
6
+ {"game": 5, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
7
+ {"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25}
8
+ {"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 34}
9
+ {"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 29}
10
+ {"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 24}
11
+ {"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25}
12
+ {"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 34}
hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt",
3
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt",
4
+ "hashes": [
5
+ "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d",
6
+ "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 6813,
11
+ "iteration": 209,
12
+ "config": {
13
+ "kind": "hexresnet",
14
+ "width": 32,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 476,
21
+ "iteration": 14,
22
+ "config": {
23
+ "kind": "transformer",
24
+ "width": 164,
25
+ "blocks": 5,
26
+ "head": "query"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 11,
31
+ "losses": 0,
32
+ "cutoffs": 1,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 1.0,
35
+ "wilson_95_decisive": [
36
+ 0.7411670330319684,
37
+ 1.0
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 367,
41
+ "mean": 32.0,
42
+ "min": 32,
43
+ "max": 32
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 353,
47
+ "mean": 32.0,
48
+ "min": 32,
49
+ "max": 32
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 208,
53
+ "mean": 0.03744002062684068,
54
+ "min": 0.0012018196284770966,
55
+ "max": 0.3364960700273514
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 207,
59
+ "mean": 0.08625425635904506,
60
+ "min": 0.06370929256081581,
61
+ "max": 0.20670939981937408
62
+ },
63
+ "elapsed_seconds": 25.690135687589645,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt",
66
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt",
67
+ "games": 12,
68
+ "batch": 64,
69
+ "simulations": 32,
70
+ "seed": 1380102,
71
+ "max_ply": 300,
72
+ "seconds": 45,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json",
74
+ "device": "cuda",
75
+ "threads": 2,
76
+ "budget_ms": 0.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 4,
81
+ "candidate_color": 1,
82
+ "winner": 1,
83
+ "outcome": "win",
84
+ "cutoff_reason": null,
85
+ "ply": 19
86
+ },
87
+ {
88
+ "game": 11,
89
+ "candidate_color": -1,
90
+ "winner": -1,
91
+ "outcome": "win",
92
+ "cutoff_reason": null,
93
+ "ply": 20
94
+ },
95
+ {
96
+ "game": 9,
97
+ "candidate_color": -1,
98
+ "winner": -1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 22
102
+ },
103
+ {
104
+ "game": 10,
105
+ "candidate_color": 1,
106
+ "winner": 1,
107
+ "outcome": "win",
108
+ "cutoff_reason": null,
109
+ "ply": 27
110
+ },
111
+ {
112
+ "game": 2,
113
+ "candidate_color": 1,
114
+ "winner": 1,
115
+ "outcome": "win",
116
+ "cutoff_reason": null,
117
+ "ply": 29
118
+ },
119
+ {
120
+ "game": 1,
121
+ "candidate_color": -1,
122
+ "winner": -1,
123
+ "outcome": "win",
124
+ "cutoff_reason": null,
125
+ "ply": 30
126
+ },
127
+ {
128
+ "game": 3,
129
+ "candidate_color": -1,
130
+ "winner": -1,
131
+ "outcome": "win",
132
+ "cutoff_reason": null,
133
+ "ply": 34
134
+ },
135
+ {
136
+ "game": 7,
137
+ "candidate_color": -1,
138
+ "winner": -1,
139
+ "outcome": "win",
140
+ "cutoff_reason": null,
141
+ "ply": 34
142
+ },
143
+ {
144
+ "game": 8,
145
+ "candidate_color": 1,
146
+ "winner": 1,
147
+ "outcome": "win",
148
+ "cutoff_reason": null,
149
+ "ply": 35
150
+ },
151
+ {
152
+ "game": 0,
153
+ "candidate_color": 1,
154
+ "winner": 1,
155
+ "outcome": "win",
156
+ "cutoff_reason": null,
157
+ "ply": 37
158
+ },
159
+ {
160
+ "game": 6,
161
+ "candidate_color": 1,
162
+ "winner": 1,
163
+ "outcome": "win",
164
+ "cutoff_reason": null,
165
+ "ply": 49
166
+ },
167
+ {
168
+ "game": 5,
169
+ "candidate_color": -1,
170
+ "winner": 0,
171
+ "outcome": "cutoff",
172
+ "cutoff_reason": "insertion",
173
+ "ply": 300
174
+ }
175
+ ]
176
+ }
hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.jsonl ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 19}
2
+ {"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 20}
3
+ {"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 22}
4
+ {"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 27}
5
+ {"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 29}
6
+ {"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 30}
7
+ {"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 34}
8
+ {"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 34}
9
+ {"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 35}
10
+ {"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 37}
11
+ {"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 49}
12
+ {"game": 5, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt",
3
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt",
4
+ "hashes": [
5
+ "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d",
6
+ "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 6813,
11
+ "iteration": 209,
12
+ "config": {
13
+ "kind": "hexresnet",
14
+ "width": 32,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 717,
21
+ "iteration": 22,
22
+ "config": {
23
+ "kind": "transformer",
24
+ "width": 256,
25
+ "blocks": 2,
26
+ "head": "query"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 12,
31
+ "losses": 0,
32
+ "cutoffs": 0,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 1.0,
35
+ "wilson_95_decisive": [
36
+ 0.7575059933447593,
37
+ 1.0
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 117,
41
+ "mean": 1548.6581196581196,
42
+ "min": 64,
43
+ "max": 10000
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 110,
47
+ "mean": 21.654545454545456,
48
+ "min": 15,
49
+ "max": 25
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 117,
53
+ "mean": 0.049162707172143154,
54
+ "min": 0.031186256557703018,
55
+ "max": 0.05291455611586571
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 110,
59
+ "mean": 0.05114460059187629,
60
+ "min": 0.05002905800938606,
61
+ "max": 0.05573977157473564
62
+ },
63
+ "elapsed_seconds": 11.402687206864357,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt",
66
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt",
67
+ "games": 12,
68
+ "batch": 1,
69
+ "simulations": 10000,
70
+ "seed": 1380102,
71
+ "max_ply": 300,
72
+ "seconds": 180,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json",
74
+ "device": "cpu",
75
+ "threads": 1,
76
+ "budget_ms": 50.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 0,
81
+ "candidate_color": 1,
82
+ "winner": 1,
83
+ "outcome": "win",
84
+ "cutoff_reason": null,
85
+ "ply": 21
86
+ },
87
+ {
88
+ "game": 1,
89
+ "candidate_color": -1,
90
+ "winner": -1,
91
+ "outcome": "win",
92
+ "cutoff_reason": null,
93
+ "ply": 16
94
+ },
95
+ {
96
+ "game": 2,
97
+ "candidate_color": 1,
98
+ "winner": 1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 29
102
+ },
103
+ {
104
+ "game": 3,
105
+ "candidate_color": -1,
106
+ "winner": -1,
107
+ "outcome": "win",
108
+ "cutoff_reason": null,
109
+ "ply": 30
110
+ },
111
+ {
112
+ "game": 4,
113
+ "candidate_color": 1,
114
+ "winner": 1,
115
+ "outcome": "win",
116
+ "cutoff_reason": null,
117
+ "ply": 21
118
+ },
119
+ {
120
+ "game": 5,
121
+ "candidate_color": -1,
122
+ "winner": -1,
123
+ "outcome": "win",
124
+ "cutoff_reason": null,
125
+ "ply": 16
126
+ },
127
+ {
128
+ "game": 6,
129
+ "candidate_color": 1,
130
+ "winner": 1,
131
+ "outcome": "win",
132
+ "cutoff_reason": null,
133
+ "ply": 13
134
+ },
135
+ {
136
+ "game": 7,
137
+ "candidate_color": -1,
138
+ "winner": -1,
139
+ "outcome": "win",
140
+ "cutoff_reason": null,
141
+ "ply": 18
142
+ },
143
+ {
144
+ "game": 8,
145
+ "candidate_color": 1,
146
+ "winner": 1,
147
+ "outcome": "win",
148
+ "cutoff_reason": null,
149
+ "ply": 17
150
+ },
151
+ {
152
+ "game": 9,
153
+ "candidate_color": -1,
154
+ "winner": -1,
155
+ "outcome": "win",
156
+ "cutoff_reason": null,
157
+ "ply": 24
158
+ },
159
+ {
160
+ "game": 10,
161
+ "candidate_color": 1,
162
+ "winner": 1,
163
+ "outcome": "win",
164
+ "cutoff_reason": null,
165
+ "ply": 17
166
+ },
167
+ {
168
+ "game": 11,
169
+ "candidate_color": -1,
170
+ "winner": -1,
171
+ "outcome": "win",
172
+ "cutoff_reason": null,
173
+ "ply": 16
174
+ }
175
+ ]
176
+ }
hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 21}
2
+ {"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 16}
3
+ {"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 29}
4
+ {"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 30}
5
+ {"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 21}
6
+ {"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 16}
7
+ {"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 13}
8
+ {"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 18}
9
+ {"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 17}
10
+ {"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 24}
11
+ {"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 17}
12
+ {"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 16}
hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt",
3
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt",
4
+ "hashes": [
5
+ "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d",
6
+ "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 6813,
11
+ "iteration": 209,
12
+ "config": {
13
+ "kind": "hexresnet",
14
+ "width": 32,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 717,
21
+ "iteration": 22,
22
+ "config": {
23
+ "kind": "transformer",
24
+ "width": 256,
25
+ "blocks": 2,
26
+ "head": "query"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 11,
31
+ "losses": 0,
32
+ "cutoffs": 1,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 1.0,
35
+ "wilson_95_decisive": [
36
+ 0.7411670330319684,
37
+ 1.0
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 339,
41
+ "mean": 32.0,
42
+ "min": 32,
43
+ "max": 32
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 324,
47
+ "mean": 32.0,
48
+ "min": 32,
49
+ "max": 32
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 205,
53
+ "mean": 0.03397062422662246,
54
+ "min": 0.001332070678472519,
55
+ "max": 0.30108968541026115
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 203,
59
+ "mean": 0.05458392078066107,
60
+ "min": 0.03652486577630043,
61
+ "max": 0.17201025411486626
62
+ },
63
+ "elapsed_seconds": 18.07961431518197,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt",
66
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt",
67
+ "games": 12,
68
+ "batch": 64,
69
+ "simulations": 32,
70
+ "seed": 1380102,
71
+ "max_ply": 300,
72
+ "seconds": 45,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json",
74
+ "device": "cuda",
75
+ "threads": 2,
76
+ "budget_ms": 0.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 8,
81
+ "candidate_color": 1,
82
+ "winner": 1,
83
+ "outcome": "win",
84
+ "cutoff_reason": null,
85
+ "ply": 15
86
+ },
87
+ {
88
+ "game": 2,
89
+ "candidate_color": 1,
90
+ "winner": 1,
91
+ "outcome": "win",
92
+ "cutoff_reason": null,
93
+ "ply": 17
94
+ },
95
+ {
96
+ "game": 6,
97
+ "candidate_color": 1,
98
+ "winner": 1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 17
102
+ },
103
+ {
104
+ "game": 1,
105
+ "candidate_color": -1,
106
+ "winner": -1,
107
+ "outcome": "win",
108
+ "cutoff_reason": null,
109
+ "ply": 22
110
+ },
111
+ {
112
+ "game": 3,
113
+ "candidate_color": -1,
114
+ "winner": -1,
115
+ "outcome": "win",
116
+ "cutoff_reason": null,
117
+ "ply": 22
118
+ },
119
+ {
120
+ "game": 5,
121
+ "candidate_color": -1,
122
+ "winner": -1,
123
+ "outcome": "win",
124
+ "cutoff_reason": null,
125
+ "ply": 24
126
+ },
127
+ {
128
+ "game": 4,
129
+ "candidate_color": 1,
130
+ "winner": 1,
131
+ "outcome": "win",
132
+ "cutoff_reason": null,
133
+ "ply": 25
134
+ },
135
+ {
136
+ "game": 10,
137
+ "candidate_color": 1,
138
+ "winner": 1,
139
+ "outcome": "win",
140
+ "cutoff_reason": null,
141
+ "ply": 31
142
+ },
143
+ {
144
+ "game": 0,
145
+ "candidate_color": 1,
146
+ "winner": 1,
147
+ "outcome": "win",
148
+ "cutoff_reason": null,
149
+ "ply": 33
150
+ },
151
+ {
152
+ "game": 9,
153
+ "candidate_color": -1,
154
+ "winner": -1,
155
+ "outcome": "win",
156
+ "cutoff_reason": null,
157
+ "ply": 34
158
+ },
159
+ {
160
+ "game": 7,
161
+ "candidate_color": -1,
162
+ "winner": -1,
163
+ "outcome": "win",
164
+ "cutoff_reason": null,
165
+ "ply": 50
166
+ },
167
+ {
168
+ "game": 11,
169
+ "candidate_color": -1,
170
+ "winner": 0,
171
+ "outcome": "cutoff",
172
+ "cutoff_reason": "insertion",
173
+ "ply": 300
174
+ }
175
+ ]
176
+ }
hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.jsonl ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 15}
2
+ {"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 17}
3
+ {"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 17}
4
+ {"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 22}
5
+ {"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 22}
6
+ {"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 24}
7
+ {"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25}
8
+ {"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 31}
9
+ {"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 33}
10
+ {"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 34}
11
+ {"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 50}
12
+ {"game": 11, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
league-024737-duel.json ADDED
@@ -0,0 +1,720 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/checkpoints/league/candidate-024737.pt",
3
+ "champion": "/workspace/gipf/checkpoints/league/champion-024737.pt",
4
+ "hashes": [
5
+ "2afe8b00ed3ef5b5052f28e2ede2853ed79ff76e1b843492e4d1e2d83059adbb",
6
+ "d5a7e3bda2966a79d22cf34f131976907260e679683d0e2680d935312f41c2f6"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 21515,
11
+ "iteration": 658,
12
+ "config": {
13
+ "kind": "hexresnet",
14
+ "width": 32,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 43869,
21
+ "iteration": 727,
22
+ "config": {
23
+ "kind": "resnet",
24
+ "width": 32,
25
+ "blocks": 2,
26
+ "head": "flat"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 13,
31
+ "losses": 65,
32
+ "cutoffs": 2,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 0.16666666666666666,
35
+ "wilson_95_decisive": [
36
+ 0.1000694541672165,
37
+ 0.26455575230757966
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 3471,
41
+ "mean": 168.01267646211465,
42
+ "min": 14,
43
+ "max": 10000
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 3545,
47
+ "mean": 289.8758815232722,
48
+ "min": 42,
49
+ "max": 10000
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 3471,
53
+ "mean": 0.050376033062156635,
54
+ "min": 0.025996044278144836,
55
+ "max": 0.07281234115362167
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 3545,
59
+ "mean": 0.05013121535163504,
60
+ "min": 0.02710983157157898,
61
+ "max": 0.06725717708468437
62
+ },
63
+ "elapsed_seconds": 353.1475488767028,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/checkpoints/league/candidate-024737.pt",
66
+ "champion": "/workspace/gipf/checkpoints/league/champion-024737.pt",
67
+ "games": 80,
68
+ "batch": 1,
69
+ "simulations": 10000,
70
+ "seed": 324737,
71
+ "max_ply": 300,
72
+ "seconds": 1200,
73
+ "output": "/workspace/gipf/reports/league-024737-duel.json",
74
+ "device": "cpu",
75
+ "threads": 1,
76
+ "budget_ms": 50.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 0,
81
+ "candidate_color": 1,
82
+ "winner": -1,
83
+ "outcome": "loss",
84
+ "cutoff_reason": null,
85
+ "ply": 78
86
+ },
87
+ {
88
+ "game": 1,
89
+ "candidate_color": -1,
90
+ "winner": 1,
91
+ "outcome": "loss",
92
+ "cutoff_reason": null,
93
+ "ply": 75
94
+ },
95
+ {
96
+ "game": 2,
97
+ "candidate_color": 1,
98
+ "winner": -1,
99
+ "outcome": "loss",
100
+ "cutoff_reason": null,
101
+ "ply": 50
102
+ },
103
+ {
104
+ "game": 3,
105
+ "candidate_color": -1,
106
+ "winner": 1,
107
+ "outcome": "loss",
108
+ "cutoff_reason": null,
109
+ "ply": 79
110
+ },
111
+ {
112
+ "game": 4,
113
+ "candidate_color": 1,
114
+ "winner": -1,
115
+ "outcome": "loss",
116
+ "cutoff_reason": null,
117
+ "ply": 74
118
+ },
119
+ {
120
+ "game": 5,
121
+ "candidate_color": -1,
122
+ "winner": 1,
123
+ "outcome": "loss",
124
+ "cutoff_reason": null,
125
+ "ply": 91
126
+ },
127
+ {
128
+ "game": 6,
129
+ "candidate_color": 1,
130
+ "winner": -1,
131
+ "outcome": "loss",
132
+ "cutoff_reason": null,
133
+ "ply": 48
134
+ },
135
+ {
136
+ "game": 7,
137
+ "candidate_color": -1,
138
+ "winner": 1,
139
+ "outcome": "loss",
140
+ "cutoff_reason": null,
141
+ "ply": 77
142
+ },
143
+ {
144
+ "game": 8,
145
+ "candidate_color": 1,
146
+ "winner": 1,
147
+ "outcome": "win",
148
+ "cutoff_reason": null,
149
+ "ply": 39
150
+ },
151
+ {
152
+ "game": 9,
153
+ "candidate_color": -1,
154
+ "winner": 1,
155
+ "outcome": "loss",
156
+ "cutoff_reason": null,
157
+ "ply": 63
158
+ },
159
+ {
160
+ "game": 10,
161
+ "candidate_color": 1,
162
+ "winner": 0,
163
+ "outcome": "cutoff",
164
+ "cutoff_reason": "insertion",
165
+ "ply": 300
166
+ },
167
+ {
168
+ "game": 11,
169
+ "candidate_color": -1,
170
+ "winner": 1,
171
+ "outcome": "loss",
172
+ "cutoff_reason": null,
173
+ "ply": 89
174
+ },
175
+ {
176
+ "game": 12,
177
+ "candidate_color": 1,
178
+ "winner": -1,
179
+ "outcome": "loss",
180
+ "cutoff_reason": null,
181
+ "ply": 82
182
+ },
183
+ {
184
+ "game": 13,
185
+ "candidate_color": -1,
186
+ "winner": 1,
187
+ "outcome": "loss",
188
+ "cutoff_reason": null,
189
+ "ply": 63
190
+ },
191
+ {
192
+ "game": 14,
193
+ "candidate_color": 1,
194
+ "winner": 1,
195
+ "outcome": "win",
196
+ "cutoff_reason": null,
197
+ "ply": 85
198
+ },
199
+ {
200
+ "game": 15,
201
+ "candidate_color": -1,
202
+ "winner": 1,
203
+ "outcome": "loss",
204
+ "cutoff_reason": null,
205
+ "ply": 69
206
+ },
207
+ {
208
+ "game": 16,
209
+ "candidate_color": 1,
210
+ "winner": -1,
211
+ "outcome": "loss",
212
+ "cutoff_reason": null,
213
+ "ply": 64
214
+ },
215
+ {
216
+ "game": 17,
217
+ "candidate_color": -1,
218
+ "winner": 1,
219
+ "outcome": "loss",
220
+ "cutoff_reason": null,
221
+ "ply": 85
222
+ },
223
+ {
224
+ "game": 18,
225
+ "candidate_color": 1,
226
+ "winner": -1,
227
+ "outcome": "loss",
228
+ "cutoff_reason": null,
229
+ "ply": 74
230
+ },
231
+ {
232
+ "game": 19,
233
+ "candidate_color": -1,
234
+ "winner": 1,
235
+ "outcome": "loss",
236
+ "cutoff_reason": null,
237
+ "ply": 51
238
+ },
239
+ {
240
+ "game": 20,
241
+ "candidate_color": 1,
242
+ "winner": -1,
243
+ "outcome": "loss",
244
+ "cutoff_reason": null,
245
+ "ply": 94
246
+ },
247
+ {
248
+ "game": 21,
249
+ "candidate_color": -1,
250
+ "winner": 1,
251
+ "outcome": "loss",
252
+ "cutoff_reason": null,
253
+ "ply": 55
254
+ },
255
+ {
256
+ "game": 22,
257
+ "candidate_color": 1,
258
+ "winner": 1,
259
+ "outcome": "win",
260
+ "cutoff_reason": null,
261
+ "ply": 97
262
+ },
263
+ {
264
+ "game": 23,
265
+ "candidate_color": -1,
266
+ "winner": 1,
267
+ "outcome": "loss",
268
+ "cutoff_reason": null,
269
+ "ply": 75
270
+ },
271
+ {
272
+ "game": 24,
273
+ "candidate_color": 1,
274
+ "winner": -1,
275
+ "outcome": "loss",
276
+ "cutoff_reason": null,
277
+ "ply": 84
278
+ },
279
+ {
280
+ "game": 25,
281
+ "candidate_color": -1,
282
+ "winner": 1,
283
+ "outcome": "loss",
284
+ "cutoff_reason": null,
285
+ "ply": 59
286
+ },
287
+ {
288
+ "game": 26,
289
+ "candidate_color": 1,
290
+ "winner": 1,
291
+ "outcome": "win",
292
+ "cutoff_reason": null,
293
+ "ply": 77
294
+ },
295
+ {
296
+ "game": 27,
297
+ "candidate_color": -1,
298
+ "winner": 1,
299
+ "outcome": "loss",
300
+ "cutoff_reason": null,
301
+ "ply": 47
302
+ },
303
+ {
304
+ "game": 28,
305
+ "candidate_color": 1,
306
+ "winner": -1,
307
+ "outcome": "loss",
308
+ "cutoff_reason": null,
309
+ "ply": 106
310
+ },
311
+ {
312
+ "game": 29,
313
+ "candidate_color": -1,
314
+ "winner": 1,
315
+ "outcome": "loss",
316
+ "cutoff_reason": null,
317
+ "ply": 61
318
+ },
319
+ {
320
+ "game": 30,
321
+ "candidate_color": 1,
322
+ "winner": -1,
323
+ "outcome": "loss",
324
+ "cutoff_reason": null,
325
+ "ply": 42
326
+ },
327
+ {
328
+ "game": 31,
329
+ "candidate_color": -1,
330
+ "winner": 1,
331
+ "outcome": "loss",
332
+ "cutoff_reason": null,
333
+ "ply": 65
334
+ },
335
+ {
336
+ "game": 32,
337
+ "candidate_color": 1,
338
+ "winner": -1,
339
+ "outcome": "loss",
340
+ "cutoff_reason": null,
341
+ "ply": 38
342
+ },
343
+ {
344
+ "game": 33,
345
+ "candidate_color": -1,
346
+ "winner": 1,
347
+ "outcome": "loss",
348
+ "cutoff_reason": null,
349
+ "ply": 73
350
+ },
351
+ {
352
+ "game": 34,
353
+ "candidate_color": 1,
354
+ "winner": 1,
355
+ "outcome": "win",
356
+ "cutoff_reason": null,
357
+ "ply": 21
358
+ },
359
+ {
360
+ "game": 35,
361
+ "candidate_color": -1,
362
+ "winner": 1,
363
+ "outcome": "loss",
364
+ "cutoff_reason": null,
365
+ "ply": 63
366
+ },
367
+ {
368
+ "game": 36,
369
+ "candidate_color": 1,
370
+ "winner": -1,
371
+ "outcome": "loss",
372
+ "cutoff_reason": null,
373
+ "ply": 164
374
+ },
375
+ {
376
+ "game": 37,
377
+ "candidate_color": -1,
378
+ "winner": 1,
379
+ "outcome": "loss",
380
+ "cutoff_reason": null,
381
+ "ply": 89
382
+ },
383
+ {
384
+ "game": 38,
385
+ "candidate_color": 1,
386
+ "winner": -1,
387
+ "outcome": "loss",
388
+ "cutoff_reason": null,
389
+ "ply": 70
390
+ },
391
+ {
392
+ "game": 39,
393
+ "candidate_color": -1,
394
+ "winner": 1,
395
+ "outcome": "loss",
396
+ "cutoff_reason": null,
397
+ "ply": 113
398
+ },
399
+ {
400
+ "game": 40,
401
+ "candidate_color": 1,
402
+ "winner": 1,
403
+ "outcome": "win",
404
+ "cutoff_reason": null,
405
+ "ply": 95
406
+ },
407
+ {
408
+ "game": 41,
409
+ "candidate_color": -1,
410
+ "winner": -1,
411
+ "outcome": "win",
412
+ "cutoff_reason": null,
413
+ "ply": 78
414
+ },
415
+ {
416
+ "game": 42,
417
+ "candidate_color": 1,
418
+ "winner": 1,
419
+ "outcome": "win",
420
+ "cutoff_reason": null,
421
+ "ply": 73
422
+ },
423
+ {
424
+ "game": 43,
425
+ "candidate_color": -1,
426
+ "winner": 1,
427
+ "outcome": "loss",
428
+ "cutoff_reason": null,
429
+ "ply": 59
430
+ },
431
+ {
432
+ "game": 44,
433
+ "candidate_color": 1,
434
+ "winner": -1,
435
+ "outcome": "loss",
436
+ "cutoff_reason": null,
437
+ "ply": 66
438
+ },
439
+ {
440
+ "game": 45,
441
+ "candidate_color": -1,
442
+ "winner": 1,
443
+ "outcome": "loss",
444
+ "cutoff_reason": null,
445
+ "ply": 93
446
+ },
447
+ {
448
+ "game": 46,
449
+ "candidate_color": 1,
450
+ "winner": -1,
451
+ "outcome": "loss",
452
+ "cutoff_reason": null,
453
+ "ply": 74
454
+ },
455
+ {
456
+ "game": 47,
457
+ "candidate_color": -1,
458
+ "winner": 1,
459
+ "outcome": "loss",
460
+ "cutoff_reason": null,
461
+ "ply": 63
462
+ },
463
+ {
464
+ "game": 48,
465
+ "candidate_color": 1,
466
+ "winner": 1,
467
+ "outcome": "win",
468
+ "cutoff_reason": null,
469
+ "ply": 53
470
+ },
471
+ {
472
+ "game": 49,
473
+ "candidate_color": -1,
474
+ "winner": 1,
475
+ "outcome": "loss",
476
+ "cutoff_reason": null,
477
+ "ply": 105
478
+ },
479
+ {
480
+ "game": 50,
481
+ "candidate_color": 1,
482
+ "winner": -1,
483
+ "outcome": "loss",
484
+ "cutoff_reason": null,
485
+ "ply": 86
486
+ },
487
+ {
488
+ "game": 51,
489
+ "candidate_color": -1,
490
+ "winner": 1,
491
+ "outcome": "loss",
492
+ "cutoff_reason": null,
493
+ "ply": 41
494
+ },
495
+ {
496
+ "game": 52,
497
+ "candidate_color": 1,
498
+ "winner": -1,
499
+ "outcome": "loss",
500
+ "cutoff_reason": null,
501
+ "ply": 42
502
+ },
503
+ {
504
+ "game": 53,
505
+ "candidate_color": -1,
506
+ "winner": -1,
507
+ "outcome": "win",
508
+ "cutoff_reason": null,
509
+ "ply": 100
510
+ },
511
+ {
512
+ "game": 54,
513
+ "candidate_color": 1,
514
+ "winner": -1,
515
+ "outcome": "loss",
516
+ "cutoff_reason": null,
517
+ "ply": 58
518
+ },
519
+ {
520
+ "game": 55,
521
+ "candidate_color": -1,
522
+ "winner": 1,
523
+ "outcome": "loss",
524
+ "cutoff_reason": null,
525
+ "ply": 43
526
+ },
527
+ {
528
+ "game": 56,
529
+ "candidate_color": 1,
530
+ "winner": 0,
531
+ "outcome": "cutoff",
532
+ "cutoff_reason": "insertion",
533
+ "ply": 300
534
+ },
535
+ {
536
+ "game": 57,
537
+ "candidate_color": -1,
538
+ "winner": 1,
539
+ "outcome": "loss",
540
+ "cutoff_reason": null,
541
+ "ply": 83
542
+ },
543
+ {
544
+ "game": 58,
545
+ "candidate_color": 1,
546
+ "winner": 1,
547
+ "outcome": "win",
548
+ "cutoff_reason": null,
549
+ "ply": 133
550
+ },
551
+ {
552
+ "game": 59,
553
+ "candidate_color": -1,
554
+ "winner": 1,
555
+ "outcome": "loss",
556
+ "cutoff_reason": null,
557
+ "ply": 49
558
+ },
559
+ {
560
+ "game": 60,
561
+ "candidate_color": 1,
562
+ "winner": 1,
563
+ "outcome": "win",
564
+ "cutoff_reason": null,
565
+ "ply": 103
566
+ },
567
+ {
568
+ "game": 61,
569
+ "candidate_color": -1,
570
+ "winner": 1,
571
+ "outcome": "loss",
572
+ "cutoff_reason": null,
573
+ "ply": 69
574
+ },
575
+ {
576
+ "game": 62,
577
+ "candidate_color": 1,
578
+ "winner": -1,
579
+ "outcome": "loss",
580
+ "cutoff_reason": null,
581
+ "ply": 54
582
+ },
583
+ {
584
+ "game": 63,
585
+ "candidate_color": -1,
586
+ "winner": 1,
587
+ "outcome": "loss",
588
+ "cutoff_reason": null,
589
+ "ply": 81
590
+ },
591
+ {
592
+ "game": 64,
593
+ "candidate_color": 1,
594
+ "winner": -1,
595
+ "outcome": "loss",
596
+ "cutoff_reason": null,
597
+ "ply": 38
598
+ },
599
+ {
600
+ "game": 65,
601
+ "candidate_color": -1,
602
+ "winner": 1,
603
+ "outcome": "loss",
604
+ "cutoff_reason": null,
605
+ "ply": 113
606
+ },
607
+ {
608
+ "game": 66,
609
+ "candidate_color": 1,
610
+ "winner": -1,
611
+ "outcome": "loss",
612
+ "cutoff_reason": null,
613
+ "ply": 54
614
+ },
615
+ {
616
+ "game": 67,
617
+ "candidate_color": -1,
618
+ "winner": 1,
619
+ "outcome": "loss",
620
+ "cutoff_reason": null,
621
+ "ply": 81
622
+ },
623
+ {
624
+ "game": 68,
625
+ "candidate_color": 1,
626
+ "winner": -1,
627
+ "outcome": "loss",
628
+ "cutoff_reason": null,
629
+ "ply": 72
630
+ },
631
+ {
632
+ "game": 69,
633
+ "candidate_color": -1,
634
+ "winner": 1,
635
+ "outcome": "loss",
636
+ "cutoff_reason": null,
637
+ "ply": 43
638
+ },
639
+ {
640
+ "game": 70,
641
+ "candidate_color": 1,
642
+ "winner": -1,
643
+ "outcome": "loss",
644
+ "cutoff_reason": null,
645
+ "ply": 28
646
+ },
647
+ {
648
+ "game": 71,
649
+ "candidate_color": -1,
650
+ "winner": 1,
651
+ "outcome": "loss",
652
+ "cutoff_reason": null,
653
+ "ply": 71
654
+ },
655
+ {
656
+ "game": 72,
657
+ "candidate_color": 1,
658
+ "winner": 1,
659
+ "outcome": "win",
660
+ "cutoff_reason": null,
661
+ "ply": 57
662
+ },
663
+ {
664
+ "game": 73,
665
+ "candidate_color": -1,
666
+ "winner": 1,
667
+ "outcome": "loss",
668
+ "cutoff_reason": null,
669
+ "ply": 33
670
+ },
671
+ {
672
+ "game": 74,
673
+ "candidate_color": 1,
674
+ "winner": -1,
675
+ "outcome": "loss",
676
+ "cutoff_reason": null,
677
+ "ply": 62
678
+ },
679
+ {
680
+ "game": 75,
681
+ "candidate_color": -1,
682
+ "winner": 1,
683
+ "outcome": "loss",
684
+ "cutoff_reason": null,
685
+ "ply": 49
686
+ },
687
+ {
688
+ "game": 76,
689
+ "candidate_color": 1,
690
+ "winner": -1,
691
+ "outcome": "loss",
692
+ "cutoff_reason": null,
693
+ "ply": 74
694
+ },
695
+ {
696
+ "game": 77,
697
+ "candidate_color": -1,
698
+ "winner": 1,
699
+ "outcome": "loss",
700
+ "cutoff_reason": null,
701
+ "ply": 86
702
+ },
703
+ {
704
+ "game": 78,
705
+ "candidate_color": 1,
706
+ "winner": -1,
707
+ "outcome": "loss",
708
+ "cutoff_reason": null,
709
+ "ply": 84
710
+ },
711
+ {
712
+ "game": 79,
713
+ "candidate_color": -1,
714
+ "winner": 1,
715
+ "outcome": "loss",
716
+ "cutoff_reason": null,
717
+ "ply": 35
718
+ }
719
+ ]
720
+ }
league-032337-duel.json ADDED
@@ -0,0 +1,720 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/checkpoints/league/candidate-032337.pt",
3
+ "champion": "/workspace/gipf/checkpoints/league/champion-032337.pt",
4
+ "hashes": [
5
+ "235d1e4d4ea756e4934f9e308a161c72698a05678457adb55efc2507d9f7f3e9",
6
+ "d5a7e3bda2966a79d22cf34f131976907260e679683d0e2680d935312f41c2f6"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 38076,
11
+ "iteration": 1165,
12
+ "config": {
13
+ "kind": "hexresnet",
14
+ "width": 32,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 43869,
21
+ "iteration": 727,
22
+ "config": {
23
+ "kind": "resnet",
24
+ "width": 32,
25
+ "blocks": 2,
26
+ "head": "flat"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 31,
31
+ "losses": 49,
32
+ "cutoffs": 0,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 0.3875,
35
+ "wilson_95_decisive": [
36
+ 0.2882456512530508,
37
+ 0.49706342789244295
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 3476,
41
+ "mean": 226.86104718066744,
42
+ "min": 41,
43
+ "max": 10000
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 3503,
47
+ "mean": 240.3930916357408,
48
+ "min": 51,
49
+ "max": 10000
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 3476,
53
+ "mean": 0.05028459619542089,
54
+ "min": 0.029697582125663757,
55
+ "max": 0.061029355973005295
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 3503,
59
+ "mean": 0.050180195522093955,
60
+ "min": 0.025200344622135162,
61
+ "max": 0.06006953492760658
62
+ },
63
+ "elapsed_seconds": 351.11589351668954,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/checkpoints/league/candidate-032337.pt",
66
+ "champion": "/workspace/gipf/checkpoints/league/champion-032337.pt",
67
+ "games": 80,
68
+ "batch": 1,
69
+ "simulations": 10000,
70
+ "seed": 332337,
71
+ "max_ply": 300,
72
+ "seconds": 1200,
73
+ "output": "/workspace/gipf/reports/league-032337-duel.json",
74
+ "device": "cpu",
75
+ "threads": 1,
76
+ "budget_ms": 50.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 0,
81
+ "candidate_color": 1,
82
+ "winner": 1,
83
+ "outcome": "win",
84
+ "cutoff_reason": null,
85
+ "ply": 73
86
+ },
87
+ {
88
+ "game": 1,
89
+ "candidate_color": -1,
90
+ "winner": -1,
91
+ "outcome": "win",
92
+ "cutoff_reason": null,
93
+ "ply": 140
94
+ },
95
+ {
96
+ "game": 2,
97
+ "candidate_color": 1,
98
+ "winner": -1,
99
+ "outcome": "loss",
100
+ "cutoff_reason": null,
101
+ "ply": 48
102
+ },
103
+ {
104
+ "game": 3,
105
+ "candidate_color": -1,
106
+ "winner": -1,
107
+ "outcome": "win",
108
+ "cutoff_reason": null,
109
+ "ply": 46
110
+ },
111
+ {
112
+ "game": 4,
113
+ "candidate_color": 1,
114
+ "winner": 1,
115
+ "outcome": "win",
116
+ "cutoff_reason": null,
117
+ "ply": 43
118
+ },
119
+ {
120
+ "game": 5,
121
+ "candidate_color": -1,
122
+ "winner": -1,
123
+ "outcome": "win",
124
+ "cutoff_reason": null,
125
+ "ply": 74
126
+ },
127
+ {
128
+ "game": 6,
129
+ "candidate_color": 1,
130
+ "winner": -1,
131
+ "outcome": "loss",
132
+ "cutoff_reason": null,
133
+ "ply": 62
134
+ },
135
+ {
136
+ "game": 7,
137
+ "candidate_color": -1,
138
+ "winner": -1,
139
+ "outcome": "win",
140
+ "cutoff_reason": null,
141
+ "ply": 36
142
+ },
143
+ {
144
+ "game": 8,
145
+ "candidate_color": 1,
146
+ "winner": 1,
147
+ "outcome": "win",
148
+ "cutoff_reason": null,
149
+ "ply": 53
150
+ },
151
+ {
152
+ "game": 9,
153
+ "candidate_color": -1,
154
+ "winner": 1,
155
+ "outcome": "loss",
156
+ "cutoff_reason": null,
157
+ "ply": 97
158
+ },
159
+ {
160
+ "game": 10,
161
+ "candidate_color": 1,
162
+ "winner": -1,
163
+ "outcome": "loss",
164
+ "cutoff_reason": null,
165
+ "ply": 50
166
+ },
167
+ {
168
+ "game": 11,
169
+ "candidate_color": -1,
170
+ "winner": -1,
171
+ "outcome": "win",
172
+ "cutoff_reason": null,
173
+ "ply": 86
174
+ },
175
+ {
176
+ "game": 12,
177
+ "candidate_color": 1,
178
+ "winner": 1,
179
+ "outcome": "win",
180
+ "cutoff_reason": null,
181
+ "ply": 77
182
+ },
183
+ {
184
+ "game": 13,
185
+ "candidate_color": -1,
186
+ "winner": 1,
187
+ "outcome": "loss",
188
+ "cutoff_reason": null,
189
+ "ply": 87
190
+ },
191
+ {
192
+ "game": 14,
193
+ "candidate_color": 1,
194
+ "winner": 1,
195
+ "outcome": "win",
196
+ "cutoff_reason": null,
197
+ "ply": 79
198
+ },
199
+ {
200
+ "game": 15,
201
+ "candidate_color": -1,
202
+ "winner": 1,
203
+ "outcome": "loss",
204
+ "cutoff_reason": null,
205
+ "ply": 79
206
+ },
207
+ {
208
+ "game": 16,
209
+ "candidate_color": 1,
210
+ "winner": -1,
211
+ "outcome": "loss",
212
+ "cutoff_reason": null,
213
+ "ply": 78
214
+ },
215
+ {
216
+ "game": 17,
217
+ "candidate_color": -1,
218
+ "winner": -1,
219
+ "outcome": "win",
220
+ "cutoff_reason": null,
221
+ "ply": 94
222
+ },
223
+ {
224
+ "game": 18,
225
+ "candidate_color": 1,
226
+ "winner": -1,
227
+ "outcome": "loss",
228
+ "cutoff_reason": null,
229
+ "ply": 82
230
+ },
231
+ {
232
+ "game": 19,
233
+ "candidate_color": -1,
234
+ "winner": 1,
235
+ "outcome": "loss",
236
+ "cutoff_reason": null,
237
+ "ply": 81
238
+ },
239
+ {
240
+ "game": 20,
241
+ "candidate_color": 1,
242
+ "winner": -1,
243
+ "outcome": "loss",
244
+ "cutoff_reason": null,
245
+ "ply": 40
246
+ },
247
+ {
248
+ "game": 21,
249
+ "candidate_color": -1,
250
+ "winner": -1,
251
+ "outcome": "win",
252
+ "cutoff_reason": null,
253
+ "ply": 30
254
+ },
255
+ {
256
+ "game": 22,
257
+ "candidate_color": 1,
258
+ "winner": -1,
259
+ "outcome": "loss",
260
+ "cutoff_reason": null,
261
+ "ply": 110
262
+ },
263
+ {
264
+ "game": 23,
265
+ "candidate_color": -1,
266
+ "winner": -1,
267
+ "outcome": "win",
268
+ "cutoff_reason": null,
269
+ "ply": 48
270
+ },
271
+ {
272
+ "game": 24,
273
+ "candidate_color": 1,
274
+ "winner": -1,
275
+ "outcome": "loss",
276
+ "cutoff_reason": null,
277
+ "ply": 78
278
+ },
279
+ {
280
+ "game": 25,
281
+ "candidate_color": -1,
282
+ "winner": 1,
283
+ "outcome": "loss",
284
+ "cutoff_reason": null,
285
+ "ply": 85
286
+ },
287
+ {
288
+ "game": 26,
289
+ "candidate_color": 1,
290
+ "winner": 1,
291
+ "outcome": "win",
292
+ "cutoff_reason": null,
293
+ "ply": 71
294
+ },
295
+ {
296
+ "game": 27,
297
+ "candidate_color": -1,
298
+ "winner": -1,
299
+ "outcome": "win",
300
+ "cutoff_reason": null,
301
+ "ply": 56
302
+ },
303
+ {
304
+ "game": 28,
305
+ "candidate_color": 1,
306
+ "winner": -1,
307
+ "outcome": "loss",
308
+ "cutoff_reason": null,
309
+ "ply": 118
310
+ },
311
+ {
312
+ "game": 29,
313
+ "candidate_color": -1,
314
+ "winner": -1,
315
+ "outcome": "win",
316
+ "cutoff_reason": null,
317
+ "ply": 80
318
+ },
319
+ {
320
+ "game": 30,
321
+ "candidate_color": 1,
322
+ "winner": 1,
323
+ "outcome": "win",
324
+ "cutoff_reason": null,
325
+ "ply": 59
326
+ },
327
+ {
328
+ "game": 31,
329
+ "candidate_color": -1,
330
+ "winner": 1,
331
+ "outcome": "loss",
332
+ "cutoff_reason": null,
333
+ "ply": 93
334
+ },
335
+ {
336
+ "game": 32,
337
+ "candidate_color": 1,
338
+ "winner": -1,
339
+ "outcome": "loss",
340
+ "cutoff_reason": null,
341
+ "ply": 90
342
+ },
343
+ {
344
+ "game": 33,
345
+ "candidate_color": -1,
346
+ "winner": 1,
347
+ "outcome": "loss",
348
+ "cutoff_reason": null,
349
+ "ply": 105
350
+ },
351
+ {
352
+ "game": 34,
353
+ "candidate_color": 1,
354
+ "winner": -1,
355
+ "outcome": "loss",
356
+ "cutoff_reason": null,
357
+ "ply": 82
358
+ },
359
+ {
360
+ "game": 35,
361
+ "candidate_color": -1,
362
+ "winner": -1,
363
+ "outcome": "win",
364
+ "cutoff_reason": null,
365
+ "ply": 56
366
+ },
367
+ {
368
+ "game": 36,
369
+ "candidate_color": 1,
370
+ "winner": 1,
371
+ "outcome": "win",
372
+ "cutoff_reason": null,
373
+ "ply": 73
374
+ },
375
+ {
376
+ "game": 37,
377
+ "candidate_color": -1,
378
+ "winner": 1,
379
+ "outcome": "loss",
380
+ "cutoff_reason": null,
381
+ "ply": 101
382
+ },
383
+ {
384
+ "game": 38,
385
+ "candidate_color": 1,
386
+ "winner": 1,
387
+ "outcome": "win",
388
+ "cutoff_reason": null,
389
+ "ply": 69
390
+ },
391
+ {
392
+ "game": 39,
393
+ "candidate_color": -1,
394
+ "winner": -1,
395
+ "outcome": "win",
396
+ "cutoff_reason": null,
397
+ "ply": 122
398
+ },
399
+ {
400
+ "game": 40,
401
+ "candidate_color": 1,
402
+ "winner": -1,
403
+ "outcome": "loss",
404
+ "cutoff_reason": null,
405
+ "ply": 60
406
+ },
407
+ {
408
+ "game": 41,
409
+ "candidate_color": -1,
410
+ "winner": 1,
411
+ "outcome": "loss",
412
+ "cutoff_reason": null,
413
+ "ply": 101
414
+ },
415
+ {
416
+ "game": 42,
417
+ "candidate_color": 1,
418
+ "winner": -1,
419
+ "outcome": "loss",
420
+ "cutoff_reason": null,
421
+ "ply": 68
422
+ },
423
+ {
424
+ "game": 43,
425
+ "candidate_color": -1,
426
+ "winner": 1,
427
+ "outcome": "loss",
428
+ "cutoff_reason": null,
429
+ "ply": 113
430
+ },
431
+ {
432
+ "game": 44,
433
+ "candidate_color": 1,
434
+ "winner": 1,
435
+ "outcome": "win",
436
+ "cutoff_reason": null,
437
+ "ply": 97
438
+ },
439
+ {
440
+ "game": 45,
441
+ "candidate_color": -1,
442
+ "winner": -1,
443
+ "outcome": "win",
444
+ "cutoff_reason": null,
445
+ "ply": 78
446
+ },
447
+ {
448
+ "game": 46,
449
+ "candidate_color": 1,
450
+ "winner": -1,
451
+ "outcome": "loss",
452
+ "cutoff_reason": null,
453
+ "ply": 88
454
+ },
455
+ {
456
+ "game": 47,
457
+ "candidate_color": -1,
458
+ "winner": 1,
459
+ "outcome": "loss",
460
+ "cutoff_reason": null,
461
+ "ply": 39
462
+ },
463
+ {
464
+ "game": 48,
465
+ "candidate_color": 1,
466
+ "winner": -1,
467
+ "outcome": "loss",
468
+ "cutoff_reason": null,
469
+ "ply": 82
470
+ },
471
+ {
472
+ "game": 49,
473
+ "candidate_color": -1,
474
+ "winner": 1,
475
+ "outcome": "loss",
476
+ "cutoff_reason": null,
477
+ "ply": 53
478
+ },
479
+ {
480
+ "game": 50,
481
+ "candidate_color": 1,
482
+ "winner": 1,
483
+ "outcome": "win",
484
+ "cutoff_reason": null,
485
+ "ply": 61
486
+ },
487
+ {
488
+ "game": 51,
489
+ "candidate_color": -1,
490
+ "winner": 1,
491
+ "outcome": "loss",
492
+ "cutoff_reason": null,
493
+ "ply": 55
494
+ },
495
+ {
496
+ "game": 52,
497
+ "candidate_color": 1,
498
+ "winner": 1,
499
+ "outcome": "win",
500
+ "cutoff_reason": null,
501
+ "ply": 55
502
+ },
503
+ {
504
+ "game": 53,
505
+ "candidate_color": -1,
506
+ "winner": 1,
507
+ "outcome": "loss",
508
+ "cutoff_reason": null,
509
+ "ply": 69
510
+ },
511
+ {
512
+ "game": 54,
513
+ "candidate_color": 1,
514
+ "winner": -1,
515
+ "outcome": "loss",
516
+ "cutoff_reason": null,
517
+ "ply": 46
518
+ },
519
+ {
520
+ "game": 55,
521
+ "candidate_color": -1,
522
+ "winner": 1,
523
+ "outcome": "loss",
524
+ "cutoff_reason": null,
525
+ "ply": 61
526
+ },
527
+ {
528
+ "game": 56,
529
+ "candidate_color": 1,
530
+ "winner": -1,
531
+ "outcome": "loss",
532
+ "cutoff_reason": null,
533
+ "ply": 64
534
+ },
535
+ {
536
+ "game": 57,
537
+ "candidate_color": -1,
538
+ "winner": -1,
539
+ "outcome": "win",
540
+ "cutoff_reason": null,
541
+ "ply": 76
542
+ },
543
+ {
544
+ "game": 58,
545
+ "candidate_color": 1,
546
+ "winner": 1,
547
+ "outcome": "win",
548
+ "cutoff_reason": null,
549
+ "ply": 41
550
+ },
551
+ {
552
+ "game": 59,
553
+ "candidate_color": -1,
554
+ "winner": 1,
555
+ "outcome": "loss",
556
+ "cutoff_reason": null,
557
+ "ply": 59
558
+ },
559
+ {
560
+ "game": 60,
561
+ "candidate_color": 1,
562
+ "winner": -1,
563
+ "outcome": "loss",
564
+ "cutoff_reason": null,
565
+ "ply": 34
566
+ },
567
+ {
568
+ "game": 61,
569
+ "candidate_color": -1,
570
+ "winner": -1,
571
+ "outcome": "win",
572
+ "cutoff_reason": null,
573
+ "ply": 80
574
+ },
575
+ {
576
+ "game": 62,
577
+ "candidate_color": 1,
578
+ "winner": 1,
579
+ "outcome": "win",
580
+ "cutoff_reason": null,
581
+ "ply": 71
582
+ },
583
+ {
584
+ "game": 63,
585
+ "candidate_color": -1,
586
+ "winner": 1,
587
+ "outcome": "loss",
588
+ "cutoff_reason": null,
589
+ "ply": 71
590
+ },
591
+ {
592
+ "game": 64,
593
+ "candidate_color": 1,
594
+ "winner": -1,
595
+ "outcome": "loss",
596
+ "cutoff_reason": null,
597
+ "ply": 72
598
+ },
599
+ {
600
+ "game": 65,
601
+ "candidate_color": -1,
602
+ "winner": 1,
603
+ "outcome": "loss",
604
+ "cutoff_reason": null,
605
+ "ply": 101
606
+ },
607
+ {
608
+ "game": 66,
609
+ "candidate_color": 1,
610
+ "winner": -1,
611
+ "outcome": "loss",
612
+ "cutoff_reason": null,
613
+ "ply": 90
614
+ },
615
+ {
616
+ "game": 67,
617
+ "candidate_color": -1,
618
+ "winner": -1,
619
+ "outcome": "win",
620
+ "cutoff_reason": null,
621
+ "ply": 80
622
+ },
623
+ {
624
+ "game": 68,
625
+ "candidate_color": 1,
626
+ "winner": -1,
627
+ "outcome": "loss",
628
+ "cutoff_reason": null,
629
+ "ply": 94
630
+ },
631
+ {
632
+ "game": 69,
633
+ "candidate_color": -1,
634
+ "winner": 1,
635
+ "outcome": "loss",
636
+ "cutoff_reason": null,
637
+ "ply": 59
638
+ },
639
+ {
640
+ "game": 70,
641
+ "candidate_color": 1,
642
+ "winner": -1,
643
+ "outcome": "loss",
644
+ "cutoff_reason": null,
645
+ "ply": 116
646
+ },
647
+ {
648
+ "game": 71,
649
+ "candidate_color": -1,
650
+ "winner": 1,
651
+ "outcome": "loss",
652
+ "cutoff_reason": null,
653
+ "ply": 93
654
+ },
655
+ {
656
+ "game": 72,
657
+ "candidate_color": 1,
658
+ "winner": -1,
659
+ "outcome": "loss",
660
+ "cutoff_reason": null,
661
+ "ply": 54
662
+ },
663
+ {
664
+ "game": 73,
665
+ "candidate_color": -1,
666
+ "winner": 1,
667
+ "outcome": "loss",
668
+ "cutoff_reason": null,
669
+ "ply": 91
670
+ },
671
+ {
672
+ "game": 74,
673
+ "candidate_color": 1,
674
+ "winner": -1,
675
+ "outcome": "loss",
676
+ "cutoff_reason": null,
677
+ "ply": 108
678
+ },
679
+ {
680
+ "game": 75,
681
+ "candidate_color": -1,
682
+ "winner": 1,
683
+ "outcome": "loss",
684
+ "cutoff_reason": null,
685
+ "ply": 83
686
+ },
687
+ {
688
+ "game": 76,
689
+ "candidate_color": 1,
690
+ "winner": -1,
691
+ "outcome": "loss",
692
+ "cutoff_reason": null,
693
+ "ply": 104
694
+ },
695
+ {
696
+ "game": 77,
697
+ "candidate_color": -1,
698
+ "winner": 1,
699
+ "outcome": "loss",
700
+ "cutoff_reason": null,
701
+ "ply": 95
702
+ },
703
+ {
704
+ "game": 78,
705
+ "candidate_color": 1,
706
+ "winner": 1,
707
+ "outcome": "win",
708
+ "cutoff_reason": null,
709
+ "ply": 85
710
+ },
711
+ {
712
+ "game": 79,
713
+ "candidate_color": -1,
714
+ "winner": 1,
715
+ "outcome": "loss",
716
+ "cutoff_reason": null,
717
+ "ply": 113
718
+ }
719
+ ]
720
+ }
league-035932-duel.json ADDED
@@ -0,0 +1,720 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/checkpoints/league/candidate-035932.pt",
3
+ "champion": "/workspace/gipf/checkpoints/league/champion-035932.pt",
4
+ "hashes": [
5
+ "319276139c115ed2302464201c50aad4a2244d84fc4671811d0bc1a1a6846553",
6
+ "d5a7e3bda2966a79d22cf34f131976907260e679683d0e2680d935312f41c2f6"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 54044,
11
+ "iteration": 1654,
12
+ "config": {
13
+ "kind": "hexresnet",
14
+ "width": 32,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 43869,
21
+ "iteration": 727,
22
+ "config": {
23
+ "kind": "resnet",
24
+ "width": 32,
25
+ "blocks": 2,
26
+ "head": "flat"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 34,
31
+ "losses": 46,
32
+ "cutoffs": 0,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 0.425,
35
+ "wilson_95_decisive": [
36
+ 0.3225656178960178,
37
+ 0.5343071015343114
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 3342,
41
+ "mean": 239.0718132854578,
42
+ "min": 12,
43
+ "max": 10000
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 3363,
47
+ "mean": 273.3193577163247,
48
+ "min": 42,
49
+ "max": 10000
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 3342,
53
+ "mean": 0.05027993674819731,
54
+ "min": 0.028711620718240738,
55
+ "max": 0.08356250822544098
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 3363,
59
+ "mean": 0.05021213266999637,
60
+ "min": 0.02769080176949501,
61
+ "max": 0.0755114033818245
62
+ },
63
+ "elapsed_seconds": 337.44456908106804,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/checkpoints/league/candidate-035932.pt",
66
+ "champion": "/workspace/gipf/checkpoints/league/champion-035932.pt",
67
+ "games": 80,
68
+ "batch": 1,
69
+ "simulations": 10000,
70
+ "seed": 335932,
71
+ "max_ply": 300,
72
+ "seconds": 1200,
73
+ "output": "/workspace/gipf/reports/league-035932-duel.json",
74
+ "device": "cpu",
75
+ "threads": 1,
76
+ "budget_ms": 50.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 0,
81
+ "candidate_color": 1,
82
+ "winner": 1,
83
+ "outcome": "win",
84
+ "cutoff_reason": null,
85
+ "ply": 47
86
+ },
87
+ {
88
+ "game": 1,
89
+ "candidate_color": -1,
90
+ "winner": 1,
91
+ "outcome": "loss",
92
+ "cutoff_reason": null,
93
+ "ply": 59
94
+ },
95
+ {
96
+ "game": 2,
97
+ "candidate_color": 1,
98
+ "winner": 1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 93
102
+ },
103
+ {
104
+ "game": 3,
105
+ "candidate_color": -1,
106
+ "winner": -1,
107
+ "outcome": "win",
108
+ "cutoff_reason": null,
109
+ "ply": 90
110
+ },
111
+ {
112
+ "game": 4,
113
+ "candidate_color": 1,
114
+ "winner": 1,
115
+ "outcome": "win",
116
+ "cutoff_reason": null,
117
+ "ply": 95
118
+ },
119
+ {
120
+ "game": 5,
121
+ "candidate_color": -1,
122
+ "winner": 1,
123
+ "outcome": "loss",
124
+ "cutoff_reason": null,
125
+ "ply": 41
126
+ },
127
+ {
128
+ "game": 6,
129
+ "candidate_color": 1,
130
+ "winner": 1,
131
+ "outcome": "win",
132
+ "cutoff_reason": null,
133
+ "ply": 47
134
+ },
135
+ {
136
+ "game": 7,
137
+ "candidate_color": -1,
138
+ "winner": 1,
139
+ "outcome": "loss",
140
+ "cutoff_reason": null,
141
+ "ply": 55
142
+ },
143
+ {
144
+ "game": 8,
145
+ "candidate_color": 1,
146
+ "winner": -1,
147
+ "outcome": "loss",
148
+ "cutoff_reason": null,
149
+ "ply": 82
150
+ },
151
+ {
152
+ "game": 9,
153
+ "candidate_color": -1,
154
+ "winner": -1,
155
+ "outcome": "win",
156
+ "cutoff_reason": null,
157
+ "ply": 62
158
+ },
159
+ {
160
+ "game": 10,
161
+ "candidate_color": 1,
162
+ "winner": -1,
163
+ "outcome": "loss",
164
+ "cutoff_reason": null,
165
+ "ply": 80
166
+ },
167
+ {
168
+ "game": 11,
169
+ "candidate_color": -1,
170
+ "winner": -1,
171
+ "outcome": "win",
172
+ "cutoff_reason": null,
173
+ "ply": 94
174
+ },
175
+ {
176
+ "game": 12,
177
+ "candidate_color": 1,
178
+ "winner": -1,
179
+ "outcome": "loss",
180
+ "cutoff_reason": null,
181
+ "ply": 58
182
+ },
183
+ {
184
+ "game": 13,
185
+ "candidate_color": -1,
186
+ "winner": 1,
187
+ "outcome": "loss",
188
+ "cutoff_reason": null,
189
+ "ply": 121
190
+ },
191
+ {
192
+ "game": 14,
193
+ "candidate_color": 1,
194
+ "winner": 1,
195
+ "outcome": "win",
196
+ "cutoff_reason": null,
197
+ "ply": 67
198
+ },
199
+ {
200
+ "game": 15,
201
+ "candidate_color": -1,
202
+ "winner": 1,
203
+ "outcome": "loss",
204
+ "cutoff_reason": null,
205
+ "ply": 113
206
+ },
207
+ {
208
+ "game": 16,
209
+ "candidate_color": 1,
210
+ "winner": 1,
211
+ "outcome": "win",
212
+ "cutoff_reason": null,
213
+ "ply": 117
214
+ },
215
+ {
216
+ "game": 17,
217
+ "candidate_color": -1,
218
+ "winner": -1,
219
+ "outcome": "win",
220
+ "cutoff_reason": null,
221
+ "ply": 50
222
+ },
223
+ {
224
+ "game": 18,
225
+ "candidate_color": 1,
226
+ "winner": -1,
227
+ "outcome": "loss",
228
+ "cutoff_reason": null,
229
+ "ply": 80
230
+ },
231
+ {
232
+ "game": 19,
233
+ "candidate_color": -1,
234
+ "winner": 1,
235
+ "outcome": "loss",
236
+ "cutoff_reason": null,
237
+ "ply": 37
238
+ },
239
+ {
240
+ "game": 20,
241
+ "candidate_color": 1,
242
+ "winner": -1,
243
+ "outcome": "loss",
244
+ "cutoff_reason": null,
245
+ "ply": 84
246
+ },
247
+ {
248
+ "game": 21,
249
+ "candidate_color": -1,
250
+ "winner": 1,
251
+ "outcome": "loss",
252
+ "cutoff_reason": null,
253
+ "ply": 63
254
+ },
255
+ {
256
+ "game": 22,
257
+ "candidate_color": 1,
258
+ "winner": 1,
259
+ "outcome": "win",
260
+ "cutoff_reason": null,
261
+ "ply": 125
262
+ },
263
+ {
264
+ "game": 23,
265
+ "candidate_color": -1,
266
+ "winner": -1,
267
+ "outcome": "win",
268
+ "cutoff_reason": null,
269
+ "ply": 68
270
+ },
271
+ {
272
+ "game": 24,
273
+ "candidate_color": 1,
274
+ "winner": 1,
275
+ "outcome": "win",
276
+ "cutoff_reason": null,
277
+ "ply": 59
278
+ },
279
+ {
280
+ "game": 25,
281
+ "candidate_color": -1,
282
+ "winner": 1,
283
+ "outcome": "loss",
284
+ "cutoff_reason": null,
285
+ "ply": 81
286
+ },
287
+ {
288
+ "game": 26,
289
+ "candidate_color": 1,
290
+ "winner": -1,
291
+ "outcome": "loss",
292
+ "cutoff_reason": null,
293
+ "ply": 68
294
+ },
295
+ {
296
+ "game": 27,
297
+ "candidate_color": -1,
298
+ "winner": 1,
299
+ "outcome": "loss",
300
+ "cutoff_reason": null,
301
+ "ply": 35
302
+ },
303
+ {
304
+ "game": 28,
305
+ "candidate_color": 1,
306
+ "winner": -1,
307
+ "outcome": "loss",
308
+ "cutoff_reason": null,
309
+ "ply": 64
310
+ },
311
+ {
312
+ "game": 29,
313
+ "candidate_color": -1,
314
+ "winner": -1,
315
+ "outcome": "win",
316
+ "cutoff_reason": null,
317
+ "ply": 80
318
+ },
319
+ {
320
+ "game": 30,
321
+ "candidate_color": 1,
322
+ "winner": -1,
323
+ "outcome": "loss",
324
+ "cutoff_reason": null,
325
+ "ply": 70
326
+ },
327
+ {
328
+ "game": 31,
329
+ "candidate_color": -1,
330
+ "winner": 1,
331
+ "outcome": "loss",
332
+ "cutoff_reason": null,
333
+ "ply": 95
334
+ },
335
+ {
336
+ "game": 32,
337
+ "candidate_color": 1,
338
+ "winner": 1,
339
+ "outcome": "win",
340
+ "cutoff_reason": null,
341
+ "ply": 55
342
+ },
343
+ {
344
+ "game": 33,
345
+ "candidate_color": -1,
346
+ "winner": -1,
347
+ "outcome": "win",
348
+ "cutoff_reason": null,
349
+ "ply": 38
350
+ },
351
+ {
352
+ "game": 34,
353
+ "candidate_color": 1,
354
+ "winner": -1,
355
+ "outcome": "loss",
356
+ "cutoff_reason": null,
357
+ "ply": 50
358
+ },
359
+ {
360
+ "game": 35,
361
+ "candidate_color": -1,
362
+ "winner": -1,
363
+ "outcome": "win",
364
+ "cutoff_reason": null,
365
+ "ply": 74
366
+ },
367
+ {
368
+ "game": 36,
369
+ "candidate_color": 1,
370
+ "winner": -1,
371
+ "outcome": "loss",
372
+ "cutoff_reason": null,
373
+ "ply": 44
374
+ },
375
+ {
376
+ "game": 37,
377
+ "candidate_color": -1,
378
+ "winner": 1,
379
+ "outcome": "loss",
380
+ "cutoff_reason": null,
381
+ "ply": 93
382
+ },
383
+ {
384
+ "game": 38,
385
+ "candidate_color": 1,
386
+ "winner": 1,
387
+ "outcome": "win",
388
+ "cutoff_reason": null,
389
+ "ply": 71
390
+ },
391
+ {
392
+ "game": 39,
393
+ "candidate_color": -1,
394
+ "winner": 1,
395
+ "outcome": "loss",
396
+ "cutoff_reason": null,
397
+ "ply": 67
398
+ },
399
+ {
400
+ "game": 40,
401
+ "candidate_color": 1,
402
+ "winner": -1,
403
+ "outcome": "loss",
404
+ "cutoff_reason": null,
405
+ "ply": 88
406
+ },
407
+ {
408
+ "game": 41,
409
+ "candidate_color": -1,
410
+ "winner": 1,
411
+ "outcome": "loss",
412
+ "cutoff_reason": null,
413
+ "ply": 79
414
+ },
415
+ {
416
+ "game": 42,
417
+ "candidate_color": 1,
418
+ "winner": 1,
419
+ "outcome": "win",
420
+ "cutoff_reason": null,
421
+ "ply": 105
422
+ },
423
+ {
424
+ "game": 43,
425
+ "candidate_color": -1,
426
+ "winner": 1,
427
+ "outcome": "loss",
428
+ "cutoff_reason": null,
429
+ "ply": 83
430
+ },
431
+ {
432
+ "game": 44,
433
+ "candidate_color": 1,
434
+ "winner": -1,
435
+ "outcome": "loss",
436
+ "cutoff_reason": null,
437
+ "ply": 84
438
+ },
439
+ {
440
+ "game": 45,
441
+ "candidate_color": -1,
442
+ "winner": 1,
443
+ "outcome": "loss",
444
+ "cutoff_reason": null,
445
+ "ply": 93
446
+ },
447
+ {
448
+ "game": 46,
449
+ "candidate_color": 1,
450
+ "winner": -1,
451
+ "outcome": "loss",
452
+ "cutoff_reason": null,
453
+ "ply": 60
454
+ },
455
+ {
456
+ "game": 47,
457
+ "candidate_color": -1,
458
+ "winner": 1,
459
+ "outcome": "loss",
460
+ "cutoff_reason": null,
461
+ "ply": 77
462
+ },
463
+ {
464
+ "game": 48,
465
+ "candidate_color": 1,
466
+ "winner": -1,
467
+ "outcome": "loss",
468
+ "cutoff_reason": null,
469
+ "ply": 58
470
+ },
471
+ {
472
+ "game": 49,
473
+ "candidate_color": -1,
474
+ "winner": 1,
475
+ "outcome": "loss",
476
+ "cutoff_reason": null,
477
+ "ply": 67
478
+ },
479
+ {
480
+ "game": 50,
481
+ "candidate_color": 1,
482
+ "winner": -1,
483
+ "outcome": "loss",
484
+ "cutoff_reason": null,
485
+ "ply": 78
486
+ },
487
+ {
488
+ "game": 51,
489
+ "candidate_color": -1,
490
+ "winner": -1,
491
+ "outcome": "win",
492
+ "cutoff_reason": null,
493
+ "ply": 36
494
+ },
495
+ {
496
+ "game": 52,
497
+ "candidate_color": 1,
498
+ "winner": -1,
499
+ "outcome": "loss",
500
+ "cutoff_reason": null,
501
+ "ply": 94
502
+ },
503
+ {
504
+ "game": 53,
505
+ "candidate_color": -1,
506
+ "winner": -1,
507
+ "outcome": "win",
508
+ "cutoff_reason": null,
509
+ "ply": 58
510
+ },
511
+ {
512
+ "game": 54,
513
+ "candidate_color": 1,
514
+ "winner": -1,
515
+ "outcome": "loss",
516
+ "cutoff_reason": null,
517
+ "ply": 66
518
+ },
519
+ {
520
+ "game": 55,
521
+ "candidate_color": -1,
522
+ "winner": 1,
523
+ "outcome": "loss",
524
+ "cutoff_reason": null,
525
+ "ply": 55
526
+ },
527
+ {
528
+ "game": 56,
529
+ "candidate_color": 1,
530
+ "winner": 1,
531
+ "outcome": "win",
532
+ "cutoff_reason": null,
533
+ "ply": 97
534
+ },
535
+ {
536
+ "game": 57,
537
+ "candidate_color": -1,
538
+ "winner": -1,
539
+ "outcome": "win",
540
+ "cutoff_reason": null,
541
+ "ply": 80
542
+ },
543
+ {
544
+ "game": 58,
545
+ "candidate_color": 1,
546
+ "winner": 1,
547
+ "outcome": "win",
548
+ "cutoff_reason": null,
549
+ "ply": 63
550
+ },
551
+ {
552
+ "game": 59,
553
+ "candidate_color": -1,
554
+ "winner": 1,
555
+ "outcome": "loss",
556
+ "cutoff_reason": null,
557
+ "ply": 73
558
+ },
559
+ {
560
+ "game": 60,
561
+ "candidate_color": 1,
562
+ "winner": 1,
563
+ "outcome": "win",
564
+ "cutoff_reason": null,
565
+ "ply": 81
566
+ },
567
+ {
568
+ "game": 61,
569
+ "candidate_color": -1,
570
+ "winner": 1,
571
+ "outcome": "loss",
572
+ "cutoff_reason": null,
573
+ "ply": 43
574
+ },
575
+ {
576
+ "game": 62,
577
+ "candidate_color": 1,
578
+ "winner": -1,
579
+ "outcome": "loss",
580
+ "cutoff_reason": null,
581
+ "ply": 44
582
+ },
583
+ {
584
+ "game": 63,
585
+ "candidate_color": -1,
586
+ "winner": 1,
587
+ "outcome": "loss",
588
+ "cutoff_reason": null,
589
+ "ply": 59
590
+ },
591
+ {
592
+ "game": 64,
593
+ "candidate_color": 1,
594
+ "winner": -1,
595
+ "outcome": "loss",
596
+ "cutoff_reason": null,
597
+ "ply": 96
598
+ },
599
+ {
600
+ "game": 65,
601
+ "candidate_color": -1,
602
+ "winner": 1,
603
+ "outcome": "loss",
604
+ "cutoff_reason": null,
605
+ "ply": 93
606
+ },
607
+ {
608
+ "game": 66,
609
+ "candidate_color": 1,
610
+ "winner": 1,
611
+ "outcome": "win",
612
+ "cutoff_reason": null,
613
+ "ply": 69
614
+ },
615
+ {
616
+ "game": 67,
617
+ "candidate_color": -1,
618
+ "winner": 1,
619
+ "outcome": "loss",
620
+ "cutoff_reason": null,
621
+ "ply": 75
622
+ },
623
+ {
624
+ "game": 68,
625
+ "candidate_color": 1,
626
+ "winner": 1,
627
+ "outcome": "win",
628
+ "cutoff_reason": null,
629
+ "ply": 64
630
+ },
631
+ {
632
+ "game": 69,
633
+ "candidate_color": -1,
634
+ "winner": -1,
635
+ "outcome": "win",
636
+ "cutoff_reason": null,
637
+ "ply": 90
638
+ },
639
+ {
640
+ "game": 70,
641
+ "candidate_color": 1,
642
+ "winner": -1,
643
+ "outcome": "loss",
644
+ "cutoff_reason": null,
645
+ "ply": 118
646
+ },
647
+ {
648
+ "game": 71,
649
+ "candidate_color": -1,
650
+ "winner": 1,
651
+ "outcome": "loss",
652
+ "cutoff_reason": null,
653
+ "ply": 67
654
+ },
655
+ {
656
+ "game": 72,
657
+ "candidate_color": 1,
658
+ "winner": 1,
659
+ "outcome": "win",
660
+ "cutoff_reason": null,
661
+ "ply": 81
662
+ },
663
+ {
664
+ "game": 73,
665
+ "candidate_color": -1,
666
+ "winner": -1,
667
+ "outcome": "win",
668
+ "cutoff_reason": null,
669
+ "ply": 56
670
+ },
671
+ {
672
+ "game": 74,
673
+ "candidate_color": 1,
674
+ "winner": 1,
675
+ "outcome": "win",
676
+ "cutoff_reason": null,
677
+ "ply": 25
678
+ },
679
+ {
680
+ "game": 75,
681
+ "candidate_color": -1,
682
+ "winner": -1,
683
+ "outcome": "win",
684
+ "cutoff_reason": null,
685
+ "ply": 76
686
+ },
687
+ {
688
+ "game": 76,
689
+ "candidate_color": 1,
690
+ "winner": 1,
691
+ "outcome": "win",
692
+ "cutoff_reason": null,
693
+ "ply": 115
694
+ },
695
+ {
696
+ "game": 77,
697
+ "candidate_color": -1,
698
+ "winner": 1,
699
+ "outcome": "loss",
700
+ "cutoff_reason": null,
701
+ "ply": 65
702
+ },
703
+ {
704
+ "game": 78,
705
+ "candidate_color": 1,
706
+ "winner": -1,
707
+ "outcome": "loss",
708
+ "cutoff_reason": null,
709
+ "ply": 112
710
+ },
711
+ {
712
+ "game": 79,
713
+ "candidate_color": -1,
714
+ "winner": -1,
715
+ "outcome": "win",
716
+ "cutoff_reason": null,
717
+ "ply": 76
718
+ }
719
+ ]
720
+ }
league-043512-duel.json ADDED
@@ -0,0 +1,720 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/checkpoints/league/candidate-043512.pt",
3
+ "champion": "/workspace/gipf/checkpoints/league/champion-043512.pt",
4
+ "hashes": [
5
+ "2cd04db0930f19b5dc231738b97607bc1dd88cbf3b65a00d9dd49e7279aeb788",
6
+ "d5a7e3bda2966a79d22cf34f131976907260e679683d0e2680d935312f41c2f6"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 69700,
11
+ "iteration": 2133,
12
+ "config": {
13
+ "kind": "hexresnet",
14
+ "width": 32,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 43869,
21
+ "iteration": 727,
22
+ "config": {
23
+ "kind": "resnet",
24
+ "width": 32,
25
+ "blocks": 2,
26
+ "head": "flat"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 39,
31
+ "losses": 41,
32
+ "cutoffs": 0,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 0.4875,
35
+ "wilson_95_decisive": [
36
+ 0.38107879540381084,
37
+ 0.5950666578345775
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 3276,
41
+ "mean": 253.34615384615384,
42
+ "min": 44,
43
+ "max": 10000
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 3293,
47
+ "mean": 232.6450045551169,
48
+ "min": 46,
49
+ "max": 10000
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 3276,
53
+ "mean": 0.050195788872520346,
54
+ "min": 0.028529252856969833,
55
+ "max": 0.06653248891234398
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 3293,
59
+ "mean": 0.050200892989618454,
60
+ "min": 0.029827728867530823,
61
+ "max": 0.06366511434316635
62
+ },
63
+ "elapsed_seconds": 330.2759933806956,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/checkpoints/league/candidate-043512.pt",
66
+ "champion": "/workspace/gipf/checkpoints/league/champion-043512.pt",
67
+ "games": 80,
68
+ "batch": 1,
69
+ "simulations": 10000,
70
+ "seed": 343512,
71
+ "max_ply": 300,
72
+ "seconds": 1200,
73
+ "output": "/workspace/gipf/reports/league-043512-duel.json",
74
+ "device": "cpu",
75
+ "threads": 1,
76
+ "budget_ms": 50.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 0,
81
+ "candidate_color": 1,
82
+ "winner": 1,
83
+ "outcome": "win",
84
+ "cutoff_reason": null,
85
+ "ply": 55
86
+ },
87
+ {
88
+ "game": 1,
89
+ "candidate_color": -1,
90
+ "winner": -1,
91
+ "outcome": "win",
92
+ "cutoff_reason": null,
93
+ "ply": 56
94
+ },
95
+ {
96
+ "game": 2,
97
+ "candidate_color": 1,
98
+ "winner": 1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 67
102
+ },
103
+ {
104
+ "game": 3,
105
+ "candidate_color": -1,
106
+ "winner": -1,
107
+ "outcome": "win",
108
+ "cutoff_reason": null,
109
+ "ply": 88
110
+ },
111
+ {
112
+ "game": 4,
113
+ "candidate_color": 1,
114
+ "winner": -1,
115
+ "outcome": "loss",
116
+ "cutoff_reason": null,
117
+ "ply": 90
118
+ },
119
+ {
120
+ "game": 5,
121
+ "candidate_color": -1,
122
+ "winner": -1,
123
+ "outcome": "win",
124
+ "cutoff_reason": null,
125
+ "ply": 102
126
+ },
127
+ {
128
+ "game": 6,
129
+ "candidate_color": 1,
130
+ "winner": -1,
131
+ "outcome": "loss",
132
+ "cutoff_reason": null,
133
+ "ply": 86
134
+ },
135
+ {
136
+ "game": 7,
137
+ "candidate_color": -1,
138
+ "winner": 1,
139
+ "outcome": "loss",
140
+ "cutoff_reason": null,
141
+ "ply": 57
142
+ },
143
+ {
144
+ "game": 8,
145
+ "candidate_color": 1,
146
+ "winner": -1,
147
+ "outcome": "loss",
148
+ "cutoff_reason": null,
149
+ "ply": 76
150
+ },
151
+ {
152
+ "game": 9,
153
+ "candidate_color": -1,
154
+ "winner": 1,
155
+ "outcome": "loss",
156
+ "cutoff_reason": null,
157
+ "ply": 73
158
+ },
159
+ {
160
+ "game": 10,
161
+ "candidate_color": 1,
162
+ "winner": 1,
163
+ "outcome": "win",
164
+ "cutoff_reason": null,
165
+ "ply": 47
166
+ },
167
+ {
168
+ "game": 11,
169
+ "candidate_color": -1,
170
+ "winner": -1,
171
+ "outcome": "win",
172
+ "cutoff_reason": null,
173
+ "ply": 64
174
+ },
175
+ {
176
+ "game": 12,
177
+ "candidate_color": 1,
178
+ "winner": 1,
179
+ "outcome": "win",
180
+ "cutoff_reason": null,
181
+ "ply": 63
182
+ },
183
+ {
184
+ "game": 13,
185
+ "candidate_color": -1,
186
+ "winner": -1,
187
+ "outcome": "win",
188
+ "cutoff_reason": null,
189
+ "ply": 132
190
+ },
191
+ {
192
+ "game": 14,
193
+ "candidate_color": 1,
194
+ "winner": -1,
195
+ "outcome": "loss",
196
+ "cutoff_reason": null,
197
+ "ply": 56
198
+ },
199
+ {
200
+ "game": 15,
201
+ "candidate_color": -1,
202
+ "winner": -1,
203
+ "outcome": "win",
204
+ "cutoff_reason": null,
205
+ "ply": 56
206
+ },
207
+ {
208
+ "game": 16,
209
+ "candidate_color": 1,
210
+ "winner": -1,
211
+ "outcome": "loss",
212
+ "cutoff_reason": null,
213
+ "ply": 46
214
+ },
215
+ {
216
+ "game": 17,
217
+ "candidate_color": -1,
218
+ "winner": 1,
219
+ "outcome": "loss",
220
+ "cutoff_reason": null,
221
+ "ply": 69
222
+ },
223
+ {
224
+ "game": 18,
225
+ "candidate_color": 1,
226
+ "winner": -1,
227
+ "outcome": "loss",
228
+ "cutoff_reason": null,
229
+ "ply": 94
230
+ },
231
+ {
232
+ "game": 19,
233
+ "candidate_color": -1,
234
+ "winner": -1,
235
+ "outcome": "win",
236
+ "cutoff_reason": null,
237
+ "ply": 112
238
+ },
239
+ {
240
+ "game": 20,
241
+ "candidate_color": 1,
242
+ "winner": -1,
243
+ "outcome": "loss",
244
+ "cutoff_reason": null,
245
+ "ply": 76
246
+ },
247
+ {
248
+ "game": 21,
249
+ "candidate_color": -1,
250
+ "winner": -1,
251
+ "outcome": "win",
252
+ "cutoff_reason": null,
253
+ "ply": 78
254
+ },
255
+ {
256
+ "game": 22,
257
+ "candidate_color": 1,
258
+ "winner": -1,
259
+ "outcome": "loss",
260
+ "cutoff_reason": null,
261
+ "ply": 80
262
+ },
263
+ {
264
+ "game": 23,
265
+ "candidate_color": -1,
266
+ "winner": -1,
267
+ "outcome": "win",
268
+ "cutoff_reason": null,
269
+ "ply": 64
270
+ },
271
+ {
272
+ "game": 24,
273
+ "candidate_color": 1,
274
+ "winner": 1,
275
+ "outcome": "win",
276
+ "cutoff_reason": null,
277
+ "ply": 61
278
+ },
279
+ {
280
+ "game": 25,
281
+ "candidate_color": -1,
282
+ "winner": 1,
283
+ "outcome": "loss",
284
+ "cutoff_reason": null,
285
+ "ply": 63
286
+ },
287
+ {
288
+ "game": 26,
289
+ "candidate_color": 1,
290
+ "winner": 1,
291
+ "outcome": "win",
292
+ "cutoff_reason": null,
293
+ "ply": 65
294
+ },
295
+ {
296
+ "game": 27,
297
+ "candidate_color": -1,
298
+ "winner": 1,
299
+ "outcome": "loss",
300
+ "cutoff_reason": null,
301
+ "ply": 87
302
+ },
303
+ {
304
+ "game": 28,
305
+ "candidate_color": 1,
306
+ "winner": -1,
307
+ "outcome": "loss",
308
+ "cutoff_reason": null,
309
+ "ply": 106
310
+ },
311
+ {
312
+ "game": 29,
313
+ "candidate_color": -1,
314
+ "winner": 1,
315
+ "outcome": "loss",
316
+ "cutoff_reason": null,
317
+ "ply": 101
318
+ },
319
+ {
320
+ "game": 30,
321
+ "candidate_color": 1,
322
+ "winner": -1,
323
+ "outcome": "loss",
324
+ "cutoff_reason": null,
325
+ "ply": 30
326
+ },
327
+ {
328
+ "game": 31,
329
+ "candidate_color": -1,
330
+ "winner": 1,
331
+ "outcome": "loss",
332
+ "cutoff_reason": null,
333
+ "ply": 71
334
+ },
335
+ {
336
+ "game": 32,
337
+ "candidate_color": 1,
338
+ "winner": 1,
339
+ "outcome": "win",
340
+ "cutoff_reason": null,
341
+ "ply": 71
342
+ },
343
+ {
344
+ "game": 33,
345
+ "candidate_color": -1,
346
+ "winner": 1,
347
+ "outcome": "loss",
348
+ "cutoff_reason": null,
349
+ "ply": 67
350
+ },
351
+ {
352
+ "game": 34,
353
+ "candidate_color": 1,
354
+ "winner": 1,
355
+ "outcome": "win",
356
+ "cutoff_reason": null,
357
+ "ply": 47
358
+ },
359
+ {
360
+ "game": 35,
361
+ "candidate_color": -1,
362
+ "winner": -1,
363
+ "outcome": "win",
364
+ "cutoff_reason": null,
365
+ "ply": 32
366
+ },
367
+ {
368
+ "game": 36,
369
+ "candidate_color": 1,
370
+ "winner": 1,
371
+ "outcome": "win",
372
+ "cutoff_reason": null,
373
+ "ply": 73
374
+ },
375
+ {
376
+ "game": 37,
377
+ "candidate_color": -1,
378
+ "winner": 1,
379
+ "outcome": "loss",
380
+ "cutoff_reason": null,
381
+ "ply": 105
382
+ },
383
+ {
384
+ "game": 38,
385
+ "candidate_color": 1,
386
+ "winner": 1,
387
+ "outcome": "win",
388
+ "cutoff_reason": null,
389
+ "ply": 77
390
+ },
391
+ {
392
+ "game": 39,
393
+ "candidate_color": -1,
394
+ "winner": 1,
395
+ "outcome": "loss",
396
+ "cutoff_reason": null,
397
+ "ply": 81
398
+ },
399
+ {
400
+ "game": 40,
401
+ "candidate_color": 1,
402
+ "winner": 1,
403
+ "outcome": "win",
404
+ "cutoff_reason": null,
405
+ "ply": 81
406
+ },
407
+ {
408
+ "game": 41,
409
+ "candidate_color": -1,
410
+ "winner": 1,
411
+ "outcome": "loss",
412
+ "cutoff_reason": null,
413
+ "ply": 111
414
+ },
415
+ {
416
+ "game": 42,
417
+ "candidate_color": 1,
418
+ "winner": -1,
419
+ "outcome": "loss",
420
+ "cutoff_reason": null,
421
+ "ply": 72
422
+ },
423
+ {
424
+ "game": 43,
425
+ "candidate_color": -1,
426
+ "winner": -1,
427
+ "outcome": "win",
428
+ "cutoff_reason": null,
429
+ "ply": 42
430
+ },
431
+ {
432
+ "game": 44,
433
+ "candidate_color": 1,
434
+ "winner": 1,
435
+ "outcome": "win",
436
+ "cutoff_reason": null,
437
+ "ply": 73
438
+ },
439
+ {
440
+ "game": 45,
441
+ "candidate_color": -1,
442
+ "winner": -1,
443
+ "outcome": "win",
444
+ "cutoff_reason": null,
445
+ "ply": 66
446
+ },
447
+ {
448
+ "game": 46,
449
+ "candidate_color": 1,
450
+ "winner": 1,
451
+ "outcome": "win",
452
+ "cutoff_reason": null,
453
+ "ply": 77
454
+ },
455
+ {
456
+ "game": 47,
457
+ "candidate_color": -1,
458
+ "winner": 1,
459
+ "outcome": "loss",
460
+ "cutoff_reason": null,
461
+ "ply": 57
462
+ },
463
+ {
464
+ "game": 48,
465
+ "candidate_color": 1,
466
+ "winner": -1,
467
+ "outcome": "loss",
468
+ "cutoff_reason": null,
469
+ "ply": 76
470
+ },
471
+ {
472
+ "game": 49,
473
+ "candidate_color": -1,
474
+ "winner": -1,
475
+ "outcome": "win",
476
+ "cutoff_reason": null,
477
+ "ply": 70
478
+ },
479
+ {
480
+ "game": 50,
481
+ "candidate_color": 1,
482
+ "winner": 1,
483
+ "outcome": "win",
484
+ "cutoff_reason": null,
485
+ "ply": 41
486
+ },
487
+ {
488
+ "game": 51,
489
+ "candidate_color": -1,
490
+ "winner": 1,
491
+ "outcome": "loss",
492
+ "cutoff_reason": null,
493
+ "ply": 85
494
+ },
495
+ {
496
+ "game": 52,
497
+ "candidate_color": 1,
498
+ "winner": -1,
499
+ "outcome": "loss",
500
+ "cutoff_reason": null,
501
+ "ply": 104
502
+ },
503
+ {
504
+ "game": 53,
505
+ "candidate_color": -1,
506
+ "winner": 1,
507
+ "outcome": "loss",
508
+ "cutoff_reason": null,
509
+ "ply": 59
510
+ },
511
+ {
512
+ "game": 54,
513
+ "candidate_color": 1,
514
+ "winner": -1,
515
+ "outcome": "loss",
516
+ "cutoff_reason": null,
517
+ "ply": 46
518
+ },
519
+ {
520
+ "game": 55,
521
+ "candidate_color": -1,
522
+ "winner": -1,
523
+ "outcome": "win",
524
+ "cutoff_reason": null,
525
+ "ply": 40
526
+ },
527
+ {
528
+ "game": 56,
529
+ "candidate_color": 1,
530
+ "winner": 1,
531
+ "outcome": "win",
532
+ "cutoff_reason": null,
533
+ "ply": 49
534
+ },
535
+ {
536
+ "game": 57,
537
+ "candidate_color": -1,
538
+ "winner": 1,
539
+ "outcome": "loss",
540
+ "cutoff_reason": null,
541
+ "ply": 65
542
+ },
543
+ {
544
+ "game": 58,
545
+ "candidate_color": 1,
546
+ "winner": -1,
547
+ "outcome": "loss",
548
+ "cutoff_reason": null,
549
+ "ply": 80
550
+ },
551
+ {
552
+ "game": 59,
553
+ "candidate_color": -1,
554
+ "winner": 1,
555
+ "outcome": "loss",
556
+ "cutoff_reason": null,
557
+ "ply": 123
558
+ },
559
+ {
560
+ "game": 60,
561
+ "candidate_color": 1,
562
+ "winner": 1,
563
+ "outcome": "win",
564
+ "cutoff_reason": null,
565
+ "ply": 51
566
+ },
567
+ {
568
+ "game": 61,
569
+ "candidate_color": -1,
570
+ "winner": 1,
571
+ "outcome": "loss",
572
+ "cutoff_reason": null,
573
+ "ply": 89
574
+ },
575
+ {
576
+ "game": 62,
577
+ "candidate_color": 1,
578
+ "winner": -1,
579
+ "outcome": "loss",
580
+ "cutoff_reason": null,
581
+ "ply": 58
582
+ },
583
+ {
584
+ "game": 63,
585
+ "candidate_color": -1,
586
+ "winner": -1,
587
+ "outcome": "win",
588
+ "cutoff_reason": null,
589
+ "ply": 84
590
+ },
591
+ {
592
+ "game": 64,
593
+ "candidate_color": 1,
594
+ "winner": -1,
595
+ "outcome": "loss",
596
+ "cutoff_reason": null,
597
+ "ply": 92
598
+ },
599
+ {
600
+ "game": 65,
601
+ "candidate_color": -1,
602
+ "winner": 1,
603
+ "outcome": "loss",
604
+ "cutoff_reason": null,
605
+ "ply": 81
606
+ },
607
+ {
608
+ "game": 66,
609
+ "candidate_color": 1,
610
+ "winner": 1,
611
+ "outcome": "win",
612
+ "cutoff_reason": null,
613
+ "ply": 49
614
+ },
615
+ {
616
+ "game": 67,
617
+ "candidate_color": -1,
618
+ "winner": -1,
619
+ "outcome": "win",
620
+ "cutoff_reason": null,
621
+ "ply": 100
622
+ },
623
+ {
624
+ "game": 68,
625
+ "candidate_color": 1,
626
+ "winner": -1,
627
+ "outcome": "loss",
628
+ "cutoff_reason": null,
629
+ "ply": 40
630
+ },
631
+ {
632
+ "game": 69,
633
+ "candidate_color": -1,
634
+ "winner": 1,
635
+ "outcome": "loss",
636
+ "cutoff_reason": null,
637
+ "ply": 49
638
+ },
639
+ {
640
+ "game": 70,
641
+ "candidate_color": 1,
642
+ "winner": 1,
643
+ "outcome": "win",
644
+ "cutoff_reason": null,
645
+ "ply": 75
646
+ },
647
+ {
648
+ "game": 71,
649
+ "candidate_color": -1,
650
+ "winner": -1,
651
+ "outcome": "win",
652
+ "cutoff_reason": null,
653
+ "ply": 104
654
+ },
655
+ {
656
+ "game": 72,
657
+ "candidate_color": 1,
658
+ "winner": -1,
659
+ "outcome": "loss",
660
+ "cutoff_reason": null,
661
+ "ply": 70
662
+ },
663
+ {
664
+ "game": 73,
665
+ "candidate_color": -1,
666
+ "winner": -1,
667
+ "outcome": "win",
668
+ "cutoff_reason": null,
669
+ "ply": 86
670
+ },
671
+ {
672
+ "game": 74,
673
+ "candidate_color": 1,
674
+ "winner": -1,
675
+ "outcome": "loss",
676
+ "cutoff_reason": null,
677
+ "ply": 62
678
+ },
679
+ {
680
+ "game": 75,
681
+ "candidate_color": -1,
682
+ "winner": 1,
683
+ "outcome": "loss",
684
+ "cutoff_reason": null,
685
+ "ply": 33
686
+ },
687
+ {
688
+ "game": 76,
689
+ "candidate_color": 1,
690
+ "winner": 1,
691
+ "outcome": "win",
692
+ "cutoff_reason": null,
693
+ "ply": 95
694
+ },
695
+ {
696
+ "game": 77,
697
+ "candidate_color": -1,
698
+ "winner": 1,
699
+ "outcome": "loss",
700
+ "cutoff_reason": null,
701
+ "ply": 71
702
+ },
703
+ {
704
+ "game": 78,
705
+ "candidate_color": 1,
706
+ "winner": 1,
707
+ "outcome": "win",
708
+ "cutoff_reason": null,
709
+ "ply": 53
710
+ },
711
+ {
712
+ "game": 79,
713
+ "candidate_color": -1,
714
+ "winner": -1,
715
+ "outcome": "win",
716
+ "cutoff_reason": null,
717
+ "ply": 84
718
+ }
719
+ ]
720
+ }
mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt",
3
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt",
4
+ "hashes": [
5
+ "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f",
6
+ "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 8740,
11
+ "iteration": 270,
12
+ "config": {
13
+ "kind": "mlp",
14
+ "width": 256,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 7877,
21
+ "iteration": 242,
22
+ "config": {
23
+ "kind": "hexresnet",
24
+ "width": 32,
25
+ "blocks": 2,
26
+ "head": "flat"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 2,
31
+ "losses": 8,
32
+ "cutoffs": 2,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 0.2,
35
+ "wilson_95_decisive": [
36
+ 0.056682151454375246,
37
+ 0.5098375284633582
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 801,
41
+ "mean": 228.27715355805245,
42
+ "min": 100,
43
+ "max": 10000
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 801,
47
+ "mean": 201.11610486891385,
48
+ "min": 41,
49
+ "max": 10000
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 801,
53
+ "mean": 0.05022746998440014,
54
+ "min": 0.030646514147520065,
55
+ "max": 0.06282975524663925
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 801,
59
+ "mean": 0.05027599407176251,
60
+ "min": 0.02731853723526001,
61
+ "max": 0.05738721042871475
62
+ },
63
+ "elapsed_seconds": 80.68824608251452,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt",
66
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt",
67
+ "games": 12,
68
+ "batch": 1,
69
+ "simulations": 10000,
70
+ "seed": 1380101,
71
+ "max_ply": 300,
72
+ "seconds": 180,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json",
74
+ "device": "cpu",
75
+ "threads": 1,
76
+ "budget_ms": 50.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 0,
81
+ "candidate_color": 1,
82
+ "winner": -1,
83
+ "outcome": "loss",
84
+ "cutoff_reason": null,
85
+ "ply": 68
86
+ },
87
+ {
88
+ "game": 1,
89
+ "candidate_color": -1,
90
+ "winner": 1,
91
+ "outcome": "loss",
92
+ "cutoff_reason": null,
93
+ "ply": 57
94
+ },
95
+ {
96
+ "game": 2,
97
+ "candidate_color": 1,
98
+ "winner": 1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 79
102
+ },
103
+ {
104
+ "game": 3,
105
+ "candidate_color": -1,
106
+ "winner": 1,
107
+ "outcome": "loss",
108
+ "cutoff_reason": null,
109
+ "ply": 73
110
+ },
111
+ {
112
+ "game": 4,
113
+ "candidate_color": 1,
114
+ "winner": 0,
115
+ "outcome": "cutoff",
116
+ "cutoff_reason": "insertion",
117
+ "ply": 300
118
+ },
119
+ {
120
+ "game": 5,
121
+ "candidate_color": -1,
122
+ "winner": 1,
123
+ "outcome": "loss",
124
+ "cutoff_reason": null,
125
+ "ply": 67
126
+ },
127
+ {
128
+ "game": 6,
129
+ "candidate_color": 1,
130
+ "winner": -1,
131
+ "outcome": "loss",
132
+ "cutoff_reason": null,
133
+ "ply": 118
134
+ },
135
+ {
136
+ "game": 7,
137
+ "candidate_color": -1,
138
+ "winner": -1,
139
+ "outcome": "win",
140
+ "cutoff_reason": null,
141
+ "ply": 72
142
+ },
143
+ {
144
+ "game": 8,
145
+ "candidate_color": 1,
146
+ "winner": -1,
147
+ "outcome": "loss",
148
+ "cutoff_reason": null,
149
+ "ply": 60
150
+ },
151
+ {
152
+ "game": 9,
153
+ "candidate_color": -1,
154
+ "winner": 0,
155
+ "outcome": "cutoff",
156
+ "cutoff_reason": "insertion",
157
+ "ply": 300
158
+ },
159
+ {
160
+ "game": 10,
161
+ "candidate_color": 1,
162
+ "winner": -1,
163
+ "outcome": "loss",
164
+ "cutoff_reason": null,
165
+ "ply": 66
166
+ },
167
+ {
168
+ "game": 11,
169
+ "candidate_color": -1,
170
+ "winner": 1,
171
+ "outcome": "loss",
172
+ "cutoff_reason": null,
173
+ "ply": 91
174
+ }
175
+ ]
176
+ }
mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"game": 0, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 68}
2
+ {"game": 1, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 57}
3
+ {"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 79}
4
+ {"game": 3, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 73}
5
+ {"game": 4, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
6
+ {"game": 5, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 67}
7
+ {"game": 6, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 118}
8
+ {"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 72}
9
+ {"game": 8, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 60}
10
+ {"game": 9, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
11
+ {"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 66}
12
+ {"game": 11, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 91}
mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt",
3
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt",
4
+ "hashes": [
5
+ "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f",
6
+ "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 8740,
11
+ "iteration": 270,
12
+ "config": {
13
+ "kind": "mlp",
14
+ "width": 256,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 7877,
21
+ "iteration": 242,
22
+ "config": {
23
+ "kind": "hexresnet",
24
+ "width": 32,
25
+ "blocks": 2,
26
+ "head": "flat"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 1,
31
+ "losses": 9,
32
+ "cutoffs": 2,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 0.1,
35
+ "wilson_95_decisive": [
36
+ 0.017876213095072896,
37
+ 0.40415002679523837
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 731,
41
+ "mean": 32.0,
42
+ "min": 32,
43
+ "max": 32
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 739,
47
+ "mean": 32.0,
48
+ "min": 32,
49
+ "max": 32
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 328,
53
+ "mean": 0.014640243871637234,
54
+ "min": 0.010727941989898682,
55
+ "max": 0.12138830497860909
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 328,
59
+ "mean": 0.03528110070790096,
60
+ "min": 0.02560272067785263,
61
+ "max": 0.18772567063570023
62
+ },
63
+ "elapsed_seconds": 16.43799015507102,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt",
66
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt",
67
+ "games": 12,
68
+ "batch": 64,
69
+ "simulations": 32,
70
+ "seed": 1380101,
71
+ "max_ply": 300,
72
+ "seconds": 45,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json",
74
+ "device": "cuda",
75
+ "threads": 2,
76
+ "budget_ms": 0.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 8,
81
+ "candidate_color": 1,
82
+ "winner": -1,
83
+ "outcome": "loss",
84
+ "cutoff_reason": null,
85
+ "ply": 24
86
+ },
87
+ {
88
+ "game": 3,
89
+ "candidate_color": -1,
90
+ "winner": 1,
91
+ "outcome": "loss",
92
+ "cutoff_reason": null,
93
+ "ply": 29
94
+ },
95
+ {
96
+ "game": 11,
97
+ "candidate_color": -1,
98
+ "winner": 1,
99
+ "outcome": "loss",
100
+ "cutoff_reason": null,
101
+ "ply": 37
102
+ },
103
+ {
104
+ "game": 10,
105
+ "candidate_color": 1,
106
+ "winner": 1,
107
+ "outcome": "win",
108
+ "cutoff_reason": null,
109
+ "ply": 39
110
+ },
111
+ {
112
+ "game": 0,
113
+ "candidate_color": 1,
114
+ "winner": -1,
115
+ "outcome": "loss",
116
+ "cutoff_reason": null,
117
+ "ply": 44
118
+ },
119
+ {
120
+ "game": 1,
121
+ "candidate_color": -1,
122
+ "winner": 1,
123
+ "outcome": "loss",
124
+ "cutoff_reason": null,
125
+ "ply": 49
126
+ },
127
+ {
128
+ "game": 6,
129
+ "candidate_color": 1,
130
+ "winner": -1,
131
+ "outcome": "loss",
132
+ "cutoff_reason": null,
133
+ "ply": 68
134
+ },
135
+ {
136
+ "game": 2,
137
+ "candidate_color": 1,
138
+ "winner": -1,
139
+ "outcome": "loss",
140
+ "cutoff_reason": null,
141
+ "ply": 70
142
+ },
143
+ {
144
+ "game": 9,
145
+ "candidate_color": -1,
146
+ "winner": 1,
147
+ "outcome": "loss",
148
+ "cutoff_reason": null,
149
+ "ply": 99
150
+ },
151
+ {
152
+ "game": 5,
153
+ "candidate_color": -1,
154
+ "winner": 1,
155
+ "outcome": "loss",
156
+ "cutoff_reason": null,
157
+ "ply": 191
158
+ },
159
+ {
160
+ "game": 4,
161
+ "candidate_color": 1,
162
+ "winner": 0,
163
+ "outcome": "cutoff",
164
+ "cutoff_reason": "insertion",
165
+ "ply": 300
166
+ },
167
+ {
168
+ "game": 7,
169
+ "candidate_color": -1,
170
+ "winner": 0,
171
+ "outcome": "cutoff",
172
+ "cutoff_reason": "insertion",
173
+ "ply": 300
174
+ }
175
+ ]
176
+ }
mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.jsonl ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"game": 8, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 24}
2
+ {"game": 3, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 29}
3
+ {"game": 11, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 37}
4
+ {"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 39}
5
+ {"game": 0, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 44}
6
+ {"game": 1, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 49}
7
+ {"game": 6, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 68}
8
+ {"game": 2, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 70}
9
+ {"game": 9, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 99}
10
+ {"game": 5, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 191}
11
+ {"game": 4, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
12
+ {"game": 7, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt",
3
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt",
4
+ "hashes": [
5
+ "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f",
6
+ "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 8740,
11
+ "iteration": 270,
12
+ "config": {
13
+ "kind": "mlp",
14
+ "width": 256,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 582,
21
+ "iteration": 18,
22
+ "config": {
23
+ "kind": "transformer",
24
+ "width": 164,
25
+ "blocks": 5,
26
+ "head": "query"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 11,
31
+ "losses": 0,
32
+ "cutoffs": 1,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 1.0,
35
+ "wilson_95_decisive": [
36
+ 0.7411670330319684,
37
+ 1.0
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 449,
41
+ "mean": 538.0890868596882,
42
+ "min": 89,
43
+ "max": 10000
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 426,
47
+ "mean": 15.556338028169014,
48
+ "min": 7,
49
+ "max": 21
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 449,
53
+ "mean": 0.04968965270936091,
54
+ "min": 0.027370132505893707,
55
+ "max": 0.056996483355760574
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 426,
59
+ "mean": 0.05177791051986352,
60
+ "min": 0.05002857372164726,
61
+ "max": 0.07907987758517265
62
+ },
63
+ "elapsed_seconds": 44.45715307071805,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt",
66
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt",
67
+ "games": 12,
68
+ "batch": 1,
69
+ "simulations": 10000,
70
+ "seed": 1380101,
71
+ "max_ply": 300,
72
+ "seconds": 180,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json",
74
+ "device": "cpu",
75
+ "threads": 1,
76
+ "budget_ms": 50.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 0,
81
+ "candidate_color": 1,
82
+ "winner": 1,
83
+ "outcome": "win",
84
+ "cutoff_reason": null,
85
+ "ply": 33
86
+ },
87
+ {
88
+ "game": 1,
89
+ "candidate_color": -1,
90
+ "winner": -1,
91
+ "outcome": "win",
92
+ "cutoff_reason": null,
93
+ "ply": 32
94
+ },
95
+ {
96
+ "game": 2,
97
+ "candidate_color": 1,
98
+ "winner": 1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 25
102
+ },
103
+ {
104
+ "game": 3,
105
+ "candidate_color": -1,
106
+ "winner": -1,
107
+ "outcome": "win",
108
+ "cutoff_reason": null,
109
+ "ply": 42
110
+ },
111
+ {
112
+ "game": 4,
113
+ "candidate_color": 1,
114
+ "winner": 1,
115
+ "outcome": "win",
116
+ "cutoff_reason": null,
117
+ "ply": 49
118
+ },
119
+ {
120
+ "game": 5,
121
+ "candidate_color": -1,
122
+ "winner": -1,
123
+ "outcome": "win",
124
+ "cutoff_reason": null,
125
+ "ply": 40
126
+ },
127
+ {
128
+ "game": 6,
129
+ "candidate_color": 1,
130
+ "winner": 0,
131
+ "outcome": "cutoff",
132
+ "cutoff_reason": "insertion",
133
+ "ply": 300
134
+ },
135
+ {
136
+ "game": 7,
137
+ "candidate_color": -1,
138
+ "winner": -1,
139
+ "outcome": "win",
140
+ "cutoff_reason": null,
141
+ "ply": 32
142
+ },
143
+ {
144
+ "game": 8,
145
+ "candidate_color": 1,
146
+ "winner": 1,
147
+ "outcome": "win",
148
+ "cutoff_reason": null,
149
+ "ply": 39
150
+ },
151
+ {
152
+ "game": 9,
153
+ "candidate_color": -1,
154
+ "winner": -1,
155
+ "outcome": "win",
156
+ "cutoff_reason": null,
157
+ "ply": 108
158
+ },
159
+ {
160
+ "game": 10,
161
+ "candidate_color": 1,
162
+ "winner": 1,
163
+ "outcome": "win",
164
+ "cutoff_reason": null,
165
+ "ply": 53
166
+ },
167
+ {
168
+ "game": 11,
169
+ "candidate_color": -1,
170
+ "winner": -1,
171
+ "outcome": "win",
172
+ "cutoff_reason": null,
173
+ "ply": 24
174
+ }
175
+ ]
176
+ }
mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 33}
2
+ {"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 32}
3
+ {"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25}
4
+ {"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 42}
5
+ {"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 49}
6
+ {"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 40}
7
+ {"game": 6, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
8
+ {"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 32}
9
+ {"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 39}
10
+ {"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 108}
11
+ {"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 53}
12
+ {"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 24}
mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt",
3
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt",
4
+ "hashes": [
5
+ "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f",
6
+ "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 8740,
11
+ "iteration": 270,
12
+ "config": {
13
+ "kind": "mlp",
14
+ "width": 256,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 582,
21
+ "iteration": 18,
22
+ "config": {
23
+ "kind": "transformer",
24
+ "width": 164,
25
+ "blocks": 5,
26
+ "head": "query"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 11,
31
+ "losses": 1,
32
+ "cutoffs": 0,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 0.9166666666666666,
35
+ "wilson_95_decisive": [
36
+ 0.6461200888588832,
37
+ 0.9851349055950829
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 473,
41
+ "mean": 32.0,
42
+ "min": 32,
43
+ "max": 32
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 454,
47
+ "mean": 32.0,
48
+ "min": 32,
49
+ "max": 32
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 171,
53
+ "mean": 0.013640849104924508,
54
+ "min": 0.0005120299756526947,
55
+ "max": 0.12017641961574554
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 163,
59
+ "mean": 0.10111915811530651,
60
+ "min": 0.04979807510972023,
61
+ "max": 0.2054123431444168
62
+ },
63
+ "elapsed_seconds": 18.855112049728632,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt",
66
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt",
67
+ "games": 12,
68
+ "batch": 64,
69
+ "simulations": 32,
70
+ "seed": 1380101,
71
+ "max_ply": 300,
72
+ "seconds": 45,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json",
74
+ "device": "cuda",
75
+ "threads": 2,
76
+ "budget_ms": 0.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 5,
81
+ "candidate_color": -1,
82
+ "winner": 1,
83
+ "outcome": "loss",
84
+ "cutoff_reason": null,
85
+ "ply": 31
86
+ },
87
+ {
88
+ "game": 7,
89
+ "candidate_color": -1,
90
+ "winner": -1,
91
+ "outcome": "win",
92
+ "cutoff_reason": null,
93
+ "ply": 32
94
+ },
95
+ {
96
+ "game": 8,
97
+ "candidate_color": 1,
98
+ "winner": 1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 41
102
+ },
103
+ {
104
+ "game": 4,
105
+ "candidate_color": 1,
106
+ "winner": 1,
107
+ "outcome": "win",
108
+ "cutoff_reason": null,
109
+ "ply": 41
110
+ },
111
+ {
112
+ "game": 10,
113
+ "candidate_color": 1,
114
+ "winner": 1,
115
+ "outcome": "win",
116
+ "cutoff_reason": null,
117
+ "ply": 41
118
+ },
119
+ {
120
+ "game": 3,
121
+ "candidate_color": -1,
122
+ "winner": -1,
123
+ "outcome": "win",
124
+ "cutoff_reason": null,
125
+ "ply": 46
126
+ },
127
+ {
128
+ "game": 9,
129
+ "candidate_color": -1,
130
+ "winner": -1,
131
+ "outcome": "win",
132
+ "cutoff_reason": null,
133
+ "ply": 64
134
+ },
135
+ {
136
+ "game": 11,
137
+ "candidate_color": -1,
138
+ "winner": -1,
139
+ "outcome": "win",
140
+ "cutoff_reason": null,
141
+ "ply": 66
142
+ },
143
+ {
144
+ "game": 6,
145
+ "candidate_color": 1,
146
+ "winner": 1,
147
+ "outcome": "win",
148
+ "cutoff_reason": null,
149
+ "ply": 71
150
+ },
151
+ {
152
+ "game": 0,
153
+ "candidate_color": 1,
154
+ "winner": 1,
155
+ "outcome": "win",
156
+ "cutoff_reason": null,
157
+ "ply": 91
158
+ },
159
+ {
160
+ "game": 2,
161
+ "candidate_color": 1,
162
+ "winner": 1,
163
+ "outcome": "win",
164
+ "cutoff_reason": null,
165
+ "ply": 115
166
+ },
167
+ {
168
+ "game": 1,
169
+ "candidate_color": -1,
170
+ "winner": -1,
171
+ "outcome": "win",
172
+ "cutoff_reason": null,
173
+ "ply": 184
174
+ }
175
+ ]
176
+ }
mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.jsonl ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"game": 5, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 31}
2
+ {"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 32}
3
+ {"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41}
4
+ {"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41}
5
+ {"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41}
6
+ {"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 46}
7
+ {"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 64}
8
+ {"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 66}
9
+ {"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 71}
10
+ {"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 91}
11
+ {"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 115}
12
+ {"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 184}
mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt",
3
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt",
4
+ "hashes": [
5
+ "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f",
6
+ "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 8740,
11
+ "iteration": 270,
12
+ "config": {
13
+ "kind": "mlp",
14
+ "width": 256,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 524,
21
+ "iteration": 15,
22
+ "config": {
23
+ "kind": "transformer",
24
+ "width": 256,
25
+ "blocks": 2,
26
+ "head": "query"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 10,
31
+ "losses": 0,
32
+ "cutoffs": 2,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 1.0,
35
+ "wilson_95_decisive": [
36
+ 0.7224672001371109,
37
+ 1.0
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 812,
41
+ "mean": 358.23522167487687,
42
+ "min": 123,
43
+ "max": 10000
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 790,
47
+ "mean": 21.61139240506329,
48
+ "min": 10,
49
+ "max": 28
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 812,
53
+ "mean": 0.050130335912291935,
54
+ "min": 0.03028266131877899,
55
+ "max": 0.10844805836677551
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 790,
59
+ "mean": 0.05129407604661169,
60
+ "min": 0.05002252012491226,
61
+ "max": 0.06667163968086243
62
+ },
63
+ "elapsed_seconds": 81.39602221176028,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt",
66
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt",
67
+ "games": 12,
68
+ "batch": 1,
69
+ "simulations": 10000,
70
+ "seed": 1380101,
71
+ "max_ply": 300,
72
+ "seconds": 180,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json",
74
+ "device": "cpu",
75
+ "threads": 1,
76
+ "budget_ms": 50.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 0,
81
+ "candidate_color": 1,
82
+ "winner": 1,
83
+ "outcome": "win",
84
+ "cutoff_reason": null,
85
+ "ply": 41
86
+ },
87
+ {
88
+ "game": 1,
89
+ "candidate_color": -1,
90
+ "winner": 0,
91
+ "outcome": "cutoff",
92
+ "cutoff_reason": "insertion",
93
+ "ply": 300
94
+ },
95
+ {
96
+ "game": 2,
97
+ "candidate_color": 1,
98
+ "winner": 1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 189
102
+ },
103
+ {
104
+ "game": 3,
105
+ "candidate_color": -1,
106
+ "winner": -1,
107
+ "outcome": "win",
108
+ "cutoff_reason": null,
109
+ "ply": 76
110
+ },
111
+ {
112
+ "game": 4,
113
+ "candidate_color": 1,
114
+ "winner": 1,
115
+ "outcome": "win",
116
+ "cutoff_reason": null,
117
+ "ply": 57
118
+ },
119
+ {
120
+ "game": 5,
121
+ "candidate_color": -1,
122
+ "winner": -1,
123
+ "outcome": "win",
124
+ "cutoff_reason": null,
125
+ "ply": 104
126
+ },
127
+ {
128
+ "game": 6,
129
+ "candidate_color": 1,
130
+ "winner": 1,
131
+ "outcome": "win",
132
+ "cutoff_reason": null,
133
+ "ply": 69
134
+ },
135
+ {
136
+ "game": 7,
137
+ "candidate_color": -1,
138
+ "winner": -1,
139
+ "outcome": "win",
140
+ "cutoff_reason": null,
141
+ "ply": 36
142
+ },
143
+ {
144
+ "game": 8,
145
+ "candidate_color": 1,
146
+ "winner": 1,
147
+ "outcome": "win",
148
+ "cutoff_reason": null,
149
+ "ply": 61
150
+ },
151
+ {
152
+ "game": 9,
153
+ "candidate_color": -1,
154
+ "winner": -1,
155
+ "outcome": "win",
156
+ "cutoff_reason": null,
157
+ "ply": 66
158
+ },
159
+ {
160
+ "game": 10,
161
+ "candidate_color": 1,
162
+ "winner": 1,
163
+ "outcome": "win",
164
+ "cutoff_reason": null,
165
+ "ply": 63
166
+ },
167
+ {
168
+ "game": 11,
169
+ "candidate_color": -1,
170
+ "winner": 0,
171
+ "outcome": "cutoff",
172
+ "cutoff_reason": "insertion",
173
+ "ply": 300
174
+ }
175
+ ]
176
+ }
mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41}
2
+ {"game": 1, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
3
+ {"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 189}
4
+ {"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 76}
5
+ {"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 57}
6
+ {"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 104}
7
+ {"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 69}
8
+ {"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 36}
9
+ {"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 61}
10
+ {"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 66}
11
+ {"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 63}
12
+ {"game": 11, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt",
3
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt",
4
+ "hashes": [
5
+ "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f",
6
+ "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 8740,
11
+ "iteration": 270,
12
+ "config": {
13
+ "kind": "mlp",
14
+ "width": 256,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 524,
21
+ "iteration": 15,
22
+ "config": {
23
+ "kind": "transformer",
24
+ "width": 256,
25
+ "blocks": 2,
26
+ "head": "query"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 5,
31
+ "losses": 2,
32
+ "cutoffs": 5,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 0.7142857142857143,
35
+ "wilson_95_decisive": [
36
+ 0.3589344518326194,
37
+ 0.9177810759959433
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 1245,
41
+ "mean": 32.0,
42
+ "min": 32,
43
+ "max": 32
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 1217,
47
+ "mean": 32.0,
48
+ "min": 32,
49
+ "max": 32
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 372,
53
+ "mean": 0.014497957342574674,
54
+ "min": 0.010189041495323181,
55
+ "max": 0.10919997468590736
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 373,
59
+ "mean": 0.07265937728832299,
60
+ "min": 0.037330176681280136,
61
+ "max": 0.17097918316721916
62
+ },
63
+ "elapsed_seconds": 32.5906348451972,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt",
66
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt",
67
+ "games": 12,
68
+ "batch": 64,
69
+ "simulations": 32,
70
+ "seed": 1380101,
71
+ "max_ply": 300,
72
+ "seconds": 45,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json",
74
+ "device": "cuda",
75
+ "threads": 2,
76
+ "budget_ms": 0.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 4,
81
+ "candidate_color": 1,
82
+ "winner": -1,
83
+ "outcome": "loss",
84
+ "cutoff_reason": null,
85
+ "ply": 38
86
+ },
87
+ {
88
+ "game": 11,
89
+ "candidate_color": -1,
90
+ "winner": -1,
91
+ "outcome": "win",
92
+ "cutoff_reason": null,
93
+ "ply": 46
94
+ },
95
+ {
96
+ "game": 5,
97
+ "candidate_color": -1,
98
+ "winner": -1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 50
102
+ },
103
+ {
104
+ "game": 3,
105
+ "candidate_color": -1,
106
+ "winner": -1,
107
+ "outcome": "win",
108
+ "cutoff_reason": null,
109
+ "ply": 60
110
+ },
111
+ {
112
+ "game": 9,
113
+ "candidate_color": -1,
114
+ "winner": -1,
115
+ "outcome": "win",
116
+ "cutoff_reason": null,
117
+ "ply": 72
118
+ },
119
+ {
120
+ "game": 6,
121
+ "candidate_color": 1,
122
+ "winner": -1,
123
+ "outcome": "loss",
124
+ "cutoff_reason": null,
125
+ "ply": 90
126
+ },
127
+ {
128
+ "game": 1,
129
+ "candidate_color": -1,
130
+ "winner": -1,
131
+ "outcome": "win",
132
+ "cutoff_reason": null,
133
+ "ply": 176
134
+ },
135
+ {
136
+ "game": 2,
137
+ "candidate_color": 1,
138
+ "winner": 0,
139
+ "outcome": "cutoff",
140
+ "cutoff_reason": "insertion",
141
+ "ply": 300
142
+ },
143
+ {
144
+ "game": 7,
145
+ "candidate_color": -1,
146
+ "winner": 0,
147
+ "outcome": "cutoff",
148
+ "cutoff_reason": "insertion",
149
+ "ply": 300
150
+ },
151
+ {
152
+ "game": 10,
153
+ "candidate_color": 1,
154
+ "winner": 0,
155
+ "outcome": "cutoff",
156
+ "cutoff_reason": "insertion",
157
+ "ply": 300
158
+ },
159
+ {
160
+ "game": 0,
161
+ "candidate_color": 1,
162
+ "winner": 0,
163
+ "outcome": "cutoff",
164
+ "cutoff_reason": "insertion",
165
+ "ply": 300
166
+ },
167
+ {
168
+ "game": 8,
169
+ "candidate_color": 1,
170
+ "winner": 0,
171
+ "outcome": "cutoff",
172
+ "cutoff_reason": "insertion",
173
+ "ply": 300
174
+ }
175
+ ]
176
+ }
mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.jsonl ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"game": 4, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 38}
2
+ {"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 46}
3
+ {"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 50}
4
+ {"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 60}
5
+ {"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 72}
6
+ {"game": 6, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 90}
7
+ {"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 176}
8
+ {"game": 2, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
9
+ {"game": 7, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
10
+ {"game": 10, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
11
+ {"game": 0, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
12
+ {"game": 8, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt",
3
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt",
4
+ "hashes": [
5
+ "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f",
6
+ "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 8740,
11
+ "iteration": 270,
12
+ "config": {
13
+ "kind": "mlp",
14
+ "width": 256,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 6214,
21
+ "iteration": 192,
22
+ "config": {
23
+ "kind": "resnet",
24
+ "width": 32,
25
+ "blocks": 2,
26
+ "head": "flat"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 3,
31
+ "losses": 9,
32
+ "cutoffs": 0,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 0.25,
35
+ "wilson_95_decisive": [
36
+ 0.08894166839405476,
37
+ 0.5323053349335656
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 975,
41
+ "mean": 268.34871794871793,
42
+ "min": 119,
43
+ "max": 10000
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 981,
47
+ "mean": 199.56574923547402,
48
+ "min": 42,
49
+ "max": 10000
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 975,
53
+ "mean": 0.05015508570732215,
54
+ "min": 0.02668626606464386,
55
+ "max": 0.05868809297680855
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 981,
59
+ "mean": 0.050266843832171046,
60
+ "min": 0.03076682612299919,
61
+ "max": 0.06585546210408211
62
+ },
63
+ "elapsed_seconds": 98.44315692782402,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt",
66
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt",
67
+ "games": 12,
68
+ "batch": 1,
69
+ "simulations": 10000,
70
+ "seed": 1380101,
71
+ "max_ply": 300,
72
+ "seconds": 180,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json",
74
+ "device": "cpu",
75
+ "threads": 1,
76
+ "budget_ms": 50.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 0,
81
+ "candidate_color": 1,
82
+ "winner": -1,
83
+ "outcome": "loss",
84
+ "cutoff_reason": null,
85
+ "ply": 126
86
+ },
87
+ {
88
+ "game": 1,
89
+ "candidate_color": -1,
90
+ "winner": 1,
91
+ "outcome": "loss",
92
+ "cutoff_reason": null,
93
+ "ply": 51
94
+ },
95
+ {
96
+ "game": 2,
97
+ "candidate_color": 1,
98
+ "winner": 1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 199
102
+ },
103
+ {
104
+ "game": 3,
105
+ "candidate_color": -1,
106
+ "winner": 1,
107
+ "outcome": "loss",
108
+ "cutoff_reason": null,
109
+ "ply": 125
110
+ },
111
+ {
112
+ "game": 4,
113
+ "candidate_color": 1,
114
+ "winner": 1,
115
+ "outcome": "win",
116
+ "cutoff_reason": null,
117
+ "ply": 87
118
+ },
119
+ {
120
+ "game": 5,
121
+ "candidate_color": -1,
122
+ "winner": -1,
123
+ "outcome": "win",
124
+ "cutoff_reason": null,
125
+ "ply": 128
126
+ },
127
+ {
128
+ "game": 6,
129
+ "candidate_color": 1,
130
+ "winner": -1,
131
+ "outcome": "loss",
132
+ "cutoff_reason": null,
133
+ "ply": 212
134
+ },
135
+ {
136
+ "game": 7,
137
+ "candidate_color": -1,
138
+ "winner": 1,
139
+ "outcome": "loss",
140
+ "cutoff_reason": null,
141
+ "ply": 245
142
+ },
143
+ {
144
+ "game": 8,
145
+ "candidate_color": 1,
146
+ "winner": -1,
147
+ "outcome": "loss",
148
+ "cutoff_reason": null,
149
+ "ply": 128
150
+ },
151
+ {
152
+ "game": 9,
153
+ "candidate_color": -1,
154
+ "winner": 1,
155
+ "outcome": "loss",
156
+ "cutoff_reason": null,
157
+ "ply": 57
158
+ },
159
+ {
160
+ "game": 10,
161
+ "candidate_color": 1,
162
+ "winner": -1,
163
+ "outcome": "loss",
164
+ "cutoff_reason": null,
165
+ "ply": 154
166
+ },
167
+ {
168
+ "game": 11,
169
+ "candidate_color": -1,
170
+ "winner": 1,
171
+ "outcome": "loss",
172
+ "cutoff_reason": null,
173
+ "ply": 131
174
+ }
175
+ ]
176
+ }
mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"game": 0, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 126}
2
+ {"game": 1, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 51}
3
+ {"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 199}
4
+ {"game": 3, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 125}
5
+ {"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 87}
6
+ {"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 128}
7
+ {"game": 6, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 212}
8
+ {"game": 7, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 245}
9
+ {"game": 8, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 128}
10
+ {"game": 9, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 57}
11
+ {"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 154}
12
+ {"game": 11, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 131}
mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt",
3
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt",
4
+ "hashes": [
5
+ "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f",
6
+ "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 8740,
11
+ "iteration": 270,
12
+ "config": {
13
+ "kind": "mlp",
14
+ "width": 256,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 6214,
21
+ "iteration": 192,
22
+ "config": {
23
+ "kind": "resnet",
24
+ "width": 32,
25
+ "blocks": 2,
26
+ "head": "flat"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 3,
31
+ "losses": 7,
32
+ "cutoffs": 2,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 0.3,
35
+ "wilson_95_decisive": [
36
+ 0.10779126740630102,
37
+ 0.6032218525388546
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 808,
41
+ "mean": 32.0,
42
+ "min": 32,
43
+ "max": 32
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 819,
47
+ "mean": 32.0,
48
+ "min": 32,
49
+ "max": 32
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 344,
53
+ "mean": 0.013873586084607035,
54
+ "min": 0.010192811489105225,
55
+ "max": 0.1148918941617012
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 348,
59
+ "mean": 0.031710495883277776,
60
+ "min": 0.023946691304445267,
61
+ "max": 0.1708715409040451
62
+ },
63
+ "elapsed_seconds": 15.873456325381994,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt",
66
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt",
67
+ "games": 12,
68
+ "batch": 64,
69
+ "simulations": 32,
70
+ "seed": 1380101,
71
+ "max_ply": 300,
72
+ "seconds": 45,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.json",
74
+ "device": "cuda",
75
+ "threads": 2,
76
+ "budget_ms": 0.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 1,
81
+ "candidate_color": -1,
82
+ "winner": 1,
83
+ "outcome": "loss",
84
+ "cutoff_reason": null,
85
+ "ply": 31
86
+ },
87
+ {
88
+ "game": 2,
89
+ "candidate_color": 1,
90
+ "winner": -1,
91
+ "outcome": "loss",
92
+ "cutoff_reason": null,
93
+ "ply": 40
94
+ },
95
+ {
96
+ "game": 11,
97
+ "candidate_color": -1,
98
+ "winner": -1,
99
+ "outcome": "win",
100
+ "cutoff_reason": null,
101
+ "ply": 44
102
+ },
103
+ {
104
+ "game": 9,
105
+ "candidate_color": -1,
106
+ "winner": 1,
107
+ "outcome": "loss",
108
+ "cutoff_reason": null,
109
+ "ply": 47
110
+ },
111
+ {
112
+ "game": 5,
113
+ "candidate_color": -1,
114
+ "winner": 1,
115
+ "outcome": "loss",
116
+ "cutoff_reason": null,
117
+ "ply": 57
118
+ },
119
+ {
120
+ "game": 4,
121
+ "candidate_color": 1,
122
+ "winner": -1,
123
+ "outcome": "loss",
124
+ "cutoff_reason": null,
125
+ "ply": 60
126
+ },
127
+ {
128
+ "game": 0,
129
+ "candidate_color": 1,
130
+ "winner": 1,
131
+ "outcome": "win",
132
+ "cutoff_reason": null,
133
+ "ply": 61
134
+ },
135
+ {
136
+ "game": 7,
137
+ "candidate_color": -1,
138
+ "winner": 1,
139
+ "outcome": "loss",
140
+ "cutoff_reason": null,
141
+ "ply": 67
142
+ },
143
+ {
144
+ "game": 10,
145
+ "candidate_color": 1,
146
+ "winner": -1,
147
+ "outcome": "loss",
148
+ "cutoff_reason": null,
149
+ "ply": 96
150
+ },
151
+ {
152
+ "game": 3,
153
+ "candidate_color": -1,
154
+ "winner": -1,
155
+ "outcome": "win",
156
+ "cutoff_reason": null,
157
+ "ply": 278
158
+ },
159
+ {
160
+ "game": 6,
161
+ "candidate_color": 1,
162
+ "winner": 0,
163
+ "outcome": "cutoff",
164
+ "cutoff_reason": "insertion",
165
+ "ply": 300
166
+ },
167
+ {
168
+ "game": 8,
169
+ "candidate_color": 1,
170
+ "winner": 0,
171
+ "outcome": "cutoff",
172
+ "cutoff_reason": "insertion",
173
+ "ply": 300
174
+ }
175
+ ]
176
+ }
mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.jsonl ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"game": 1, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 31}
2
+ {"game": 2, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 40}
3
+ {"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 44}
4
+ {"game": 9, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 47}
5
+ {"game": 5, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 57}
6
+ {"game": 4, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 60}
7
+ {"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 61}
8
+ {"game": 7, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 67}
9
+ {"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 96}
10
+ {"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 278}
11
+ {"game": 6, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
12
+ {"game": 8, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300}
mlp256-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt",
3
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt",
4
+ "hashes": [
5
+ "9ca408753dfbe4c2103016e9694be0e5033ed517459bf8a7cef2cb56164fa9a7",
6
+ "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d"
7
+ ],
8
+ "models": [
9
+ {
10
+ "games": 8122,
11
+ "iteration": 251,
12
+ "config": {
13
+ "kind": "mlp",
14
+ "width": 256,
15
+ "blocks": 2,
16
+ "head": "flat"
17
+ }
18
+ },
19
+ {
20
+ "games": 6813,
21
+ "iteration": 209,
22
+ "config": {
23
+ "kind": "hexresnet",
24
+ "width": 32,
25
+ "blocks": 2,
26
+ "head": "flat"
27
+ }
28
+ }
29
+ ],
30
+ "wins": 0,
31
+ "losses": 12,
32
+ "cutoffs": 0,
33
+ "unfinished": 0,
34
+ "win_rate_decisive": 0.0,
35
+ "wilson_95_decisive": [
36
+ 0.0,
37
+ 0.2424940066552408
38
+ ],
39
+ "candidate_root_simulations": {
40
+ "count": 413,
41
+ "mean": 201.4721549636804,
42
+ "min": 139,
43
+ "max": 5467
44
+ },
45
+ "champion_root_simulations": {
46
+ "count": 414,
47
+ "mean": 524.7777777777778,
48
+ "min": 44,
49
+ "max": 10000
50
+ },
51
+ "candidate_search_seconds": {
52
+ "count": 413,
53
+ "mean": 0.05022500817084428,
54
+ "min": 0.05002162232995033,
55
+ "max": 0.05760979652404785
56
+ },
57
+ "champion_search_seconds": {
58
+ "count": 414,
59
+ "mean": 0.04998849508699012,
60
+ "min": 0.028176337480545044,
61
+ "max": 0.08139600604772568
62
+ },
63
+ "elapsed_seconds": 41.53895569965243,
64
+ "config": {
65
+ "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt",
66
+ "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt",
67
+ "games": 12,
68
+ "batch": 1,
69
+ "simulations": 10000,
70
+ "seed": 1380102,
71
+ "max_ply": 300,
72
+ "seconds": 180,
73
+ "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json",
74
+ "device": "cpu",
75
+ "threads": 1,
76
+ "budget_ms": 50.0
77
+ },
78
+ "records": [
79
+ {
80
+ "game": 0,
81
+ "candidate_color": 1,
82
+ "winner": -1,
83
+ "outcome": "loss",
84
+ "cutoff_reason": null,
85
+ "ply": 28
86
+ },
87
+ {
88
+ "game": 1,
89
+ "candidate_color": -1,
90
+ "winner": 1,
91
+ "outcome": "loss",
92
+ "cutoff_reason": null,
93
+ "ply": 41
94
+ },
95
+ {
96
+ "game": 2,
97
+ "candidate_color": 1,
98
+ "winner": -1,
99
+ "outcome": "loss",
100
+ "cutoff_reason": null,
101
+ "ply": 44
102
+ },
103
+ {
104
+ "game": 3,
105
+ "candidate_color": -1,
106
+ "winner": 1,
107
+ "outcome": "loss",
108
+ "cutoff_reason": null,
109
+ "ply": 171
110
+ },
111
+ {
112
+ "game": 4,
113
+ "candidate_color": 1,
114
+ "winner": -1,
115
+ "outcome": "loss",
116
+ "cutoff_reason": null,
117
+ "ply": 20
118
+ },
119
+ {
120
+ "game": 5,
121
+ "candidate_color": -1,
122
+ "winner": 1,
123
+ "outcome": "loss",
124
+ "cutoff_reason": null,
125
+ "ply": 41
126
+ },
127
+ {
128
+ "game": 6,
129
+ "candidate_color": 1,
130
+ "winner": -1,
131
+ "outcome": "loss",
132
+ "cutoff_reason": null,
133
+ "ply": 70
134
+ },
135
+ {
136
+ "game": 7,
137
+ "candidate_color": -1,
138
+ "winner": 1,
139
+ "outcome": "loss",
140
+ "cutoff_reason": null,
141
+ "ply": 57
142
+ },
143
+ {
144
+ "game": 8,
145
+ "candidate_color": 1,
146
+ "winner": -1,
147
+ "outcome": "loss",
148
+ "cutoff_reason": null,
149
+ "ply": 46
150
+ },
151
+ {
152
+ "game": 9,
153
+ "candidate_color": -1,
154
+ "winner": 1,
155
+ "outcome": "loss",
156
+ "cutoff_reason": null,
157
+ "ply": 63
158
+ },
159
+ {
160
+ "game": 10,
161
+ "candidate_color": 1,
162
+ "winner": -1,
163
+ "outcome": "loss",
164
+ "cutoff_reason": null,
165
+ "ply": 88
166
+ },
167
+ {
168
+ "game": 11,
169
+ "candidate_color": -1,
170
+ "winner": 1,
171
+ "outcome": "loss",
172
+ "cutoff_reason": null,
173
+ "ply": 43
174
+ }
175
+ ]
176
+ }