diff --git a/MODEL_CARD.md b/MODEL_CARD.md new file mode 100644 index 0000000000000000000000000000000000000000..8cafa2b856c7f28e5cb7c014ece1d1d75b0f3242 --- /dev/null +++ b/MODEL_CARD.md @@ -0,0 +1,7 @@ +# GIPF Zero checkpoint + +Custom PyTorch `state_dict` checkpoint for the [GIPF Zero source](https://github.com/wusche1/gipf-zero). Load with `training.model.load_model`. + +Source SHA-256: `aa533bb8922231a3e0dc53dd2ff947cab6899ef09a9245463c89ac99ea776ef0`. + +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. diff --git a/README.md b/README.md new file mode 100644 index 0000000000000000000000000000000000000000..f4b4fdb0cf467fedcbf9c9766fca1d8733c67759 --- /dev/null +++ b/README.md @@ -0,0 +1,18 @@ +--- +library_name: pytorch +tags: +- reinforcement-learning +- board-game +--- + +# GIPF Zero checkpoint + +Custom PyTorch `state_dict` checkpoint for the [GIPF Zero source](https://github.com/wusche1/gipf-zero). Load it with `training.model.load_model`. + +Source training checkpoint SHA-256: `aa533bb8922231a3e0dc53dd2ff947cab6899ef09a9245463c89ac99ea776ef0`. + +Served `champion.pt` artifact SHA-256: `d5a7e3bda2966a79d22cf34f131976907260e679683d0e2680d935312f41c2f6`. + +`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. + +[Full overnight comparison and results](RESULTS.md). diff --git a/RESULTS.md b/RESULTS.md new file mode 100644 index 0000000000000000000000000000000000000000..d4620b02def7d1900219243ee1a79d337cb7371c --- /dev/null +++ b/RESULTS.md @@ -0,0 +1,146 @@ +# Overnight evaluation — `20260909T235853Z` + +**Status:** `partial` + +> This experiment is incomplete or still running. Missing matches are shown as unplayed; no winner is inferred from partial coverage. + +## Protocol + +- Training: isolated **600 seconds per run**, 2 seed(s), native forest search: **True**. +- 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. +- Whole-match timeouts are 180 seconds (CPU) and 45 seconds (fixed simulations); unfinished games are reported separately. +- Training seconds are nominal allocations; the elapsed column includes checkpoint and shutdown overhead. +- Parameter counts are read from each run's initial `start` metric, before training. +- This is **not a pure architecture ablation**: the compared models use different output heads (flat, query). +- Results are automated game comparisons; this report makes no human-rating claim. + +## Training + +| Model | Seed | Games | Updates | Seconds | Parameters | State | +|---|---|---|---|---|---|---| +| mlp256 | 410101 | 8740 | 8640 | 609.3 | 979371 | complete | +| mlp256 | 410102 | 8122 | 8032 | 611.9 | 979371 | complete | +| squarecnn32 | 410101 | 6214 | 6144 | 603.0 | 1163523 | complete | +| squarecnn32 | 410102 | 8449 | 8320 | 602.7 | 1163523 | complete | +| hexcnn32 | 410101 | 7877 | 7744 | 602.7 | 1163523 | complete | +| hexcnn32 | 410102 | 6813 | 6688 | 602.6 | 1163523 | complete | +| querytransformer256x2 | 410101 | 524 | 480 | 601.2 | 1102124 | complete | +| querytransformer256x2 | 410102 | 717 | 704 | 601.0 | 1102124 | complete | +| querytransformer164x5 | 410101 | 582 | 550 | 600.7 | 1115664 | complete | +| querytransformer164x5 | 410102 | 476 | 448 | 601.7 | 1115664 | complete | + +## Pooled equal-CPU ranking + +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. + +| Model | W | L | C | Played | Unplayed | Win rate | Wilson 95% | +|---|---|---|---|---|---|---|---| +| squarecnn32 | 80 | 15 | 1 | 96 | 0 | 83.3% | 74.6%–89.5% | +| hexcnn32 | 74 | 15 | 7 | 96 | 0 | 77.1% | 67.7%–84.4% | +| mlp256 | 41 | 40 | 15 | 96 | 0 | 42.7% | 33.3%–52.7% | +| querytransformer256x2 | 9 | 74 | 13 | 96 | 0 | 9.4% | 5.0%–16.9% | +| querytransformer164x5 | 7 | 67 | 22 | 96 | 0 | 7.3% | 3.6%–14.3% | + +## Equal-CPU pairwise + +| Pair (left perspective) | Left W | Left L | C | Games | Wilson 95% (left) | +|---|---|---|---|---|---| +| hexcnn32 vs mlp256 | 20 | 2 | 2 | 24 | 64.1%–93.3% | +| hexcnn32 vs querytransformer164x5 | 19 | 0 | 5 | 24 | 59.5%–90.8% | +| hexcnn32 vs querytransformer256x2 | 24 | 0 | 0 | 24 | 86.2%–100.0% | +| hexcnn32 vs squarecnn32 | 11 | 13 | 0 | 24 | 27.9%–64.9% | +| mlp256 vs querytransformer164x5 | 16 | 0 | 8 | 24 | 46.7%–82.0% | +| mlp256 vs querytransformer256x2 | 19 | 0 | 5 | 24 | 59.5%–90.8% | +| mlp256 vs squarecnn32 | 4 | 20 | 0 | 24 | 6.7%–35.9% | +| querytransformer164x5 vs querytransformer256x2 | 7 | 9 | 8 | 24 | 14.9%–49.2% | +| querytransformer164x5 vs squarecnn32 | 0 | 23 | 1 | 24 | 0.0%–13.8% | +| querytransformer256x2 vs squarecnn32 | 0 | 24 | 0 | 24 | 0.0%–13.8% | + +## Fixed-simulation pairwise + +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. + +| Pair (left perspective) | Left W | Left L | C | Games | Wilson 95% (left) | +|---|---|---|---|---|---| +| hexcnn32 vs mlp256 | 18 | 1 | 5 | 24 | 55.1%–88.0% | +| hexcnn32 vs querytransformer164x5 | 17 | 0 | 7 | 24 | 50.8%–85.1% | +| hexcnn32 vs querytransformer256x2 | 20 | 2 | 2 | 24 | 64.1%–93.3% | +| hexcnn32 vs squarecnn32 | 7 | 14 | 3 | 24 | 14.9%–49.2% | +| mlp256 vs querytransformer164x5 | 14 | 2 | 8 | 24 | 38.8%–75.5% | +| mlp256 vs querytransformer256x2 | 13 | 2 | 9 | 24 | 35.1%–72.1% | +| mlp256 vs squarecnn32 | 4 | 15 | 5 | 24 | 6.7%–35.9% | +| querytransformer164x5 vs querytransformer256x2 | 2 | 14 | 8 | 24 | 2.3%–25.8% | +| querytransformer164x5 vs squarecnn32 | 1 | 21 | 2 | 24 | 0.7%–20.2% | +| querytransformer256x2 vs squarecnn32 | 0 | 21 | 3 | 24 | 0.0%–13.8% | + +## Extra tie-break attempts + +Attempt 3 / 80-game tie-breaks are excluded from the primary pools above. + +| Report | Candidate | Opponent | W | L | C | State | +|---|---|---|---|---|---|---| +| squarecnn32-seed410101-vs-hexcnn32-seed410102-equal_cpu_time-seed989001-attempt3.json | squarecnn32 | hexcnn32 | 42 | 38 | 0 | complete | + +## Continuation league evaluations + +These intermediate continuation snapshots are included for audit. They do not by themselves imply promotion; promotion requires its explicit gate. + +| Report | Evaluation | Candidate games | W | L | C | Unfinished | +|---|---|---|---|---|---|---| +| [league-024737-duel.json](league-024737-duel.json) | duel | 21515 | 13 | 65 | 2 | 0 | +| [league-032337-duel.json](league-032337-duel.json) | duel | 38076 | 31 | 49 | 0 | 0 | +| [league-035932-duel.json](league-035932-duel.json) | duel | 54044 | 34 | 46 | 0 | 0 | +| [league-043512-duel.json](league-043512-duel.json) | duel | 69700 | 39 | 41 | 0 | 0 | + +## Promotion and publication + +- Promotion gate: `False`. +- `duel_report`: [final-vs-served.json](final-vs-served.json) +- Hugging Face: [wuschelschulz/gipf-zero](https://huggingface.co/wuschelschulz/gipf-zero) + +## Raw duel reports + +- [hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json](hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json) +- [hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json](hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json) +- [hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json](hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json) +- [hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json](hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json) +- [hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json](hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json) +- [hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json](hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json) +- [hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json](hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json) +- [hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json](hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json) +- [hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json](hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json) +- [mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json](mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json) +- [mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json](mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json) +- [mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json](mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json) +- [mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json](mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json) +- [mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json](mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json) +- [mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json](mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json) +- [mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json](mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json) +- [mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.json](mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.json) +- [mlp256-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json](mlp256-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json) +- [mlp256-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json](mlp256-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json) +- [mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json](mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json) +- [mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json](mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json) +- [mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json](mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json) +- [mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json](mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json) +- [mlp256-seed410102-vs-squarecnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json](mlp256-seed410102-vs-squarecnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json) +- [mlp256-seed410102-vs-squarecnn32-seed410102-equal_simulations-seed1380102-attempt1.json](mlp256-seed410102-vs-squarecnn32-seed410102-equal_simulations-seed1380102-attempt1.json) +- [querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json](querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json) +- [querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json](querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json) +- [querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json](querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json) +- [querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json](querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json) +- [querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json](querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json) +- [querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-retry-complete.json](querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-retry-complete.json) +- [squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json](squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json) +- [squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json](squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json) +- [squarecnn32-seed410101-vs-hexcnn32-seed410102-equal_cpu_time-seed989001-attempt3.json](squarecnn32-seed410101-vs-hexcnn32-seed410102-equal_cpu_time-seed989001-attempt3.json) +- [squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json](squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json) +- [squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json](squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json) +- [squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json](squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json) +- [squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json](squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json) +- [squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json](squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json) +- [squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json](squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json) +- [squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json](squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json) +- [squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json](squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json) +- [squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json](squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json) +- [squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json](squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json) diff --git a/champion-metadata.json b/champion-metadata.json new file mode 100644 index 0000000000000000000000000000000000000000..382116593a8d6f31daac084984cd80e7117706f4 --- /dev/null +++ b/champion-metadata.json @@ -0,0 +1,34 @@ +{ + "name": "GIPF Zero \u2014 43,869 games", + "kind": "rl", + "games_trained": 43869, + "iteration": 727, + "architecture": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + }, + "source_checkpoint_sha256": "aa533bb8922231a3e0dc53dd2ff947cab6899ef09a9245463c89ac99ea776ef0", + "updated_utc": "2026-09-09T21:20:52.781670+00:00", + "evaluations": [ + { + "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 + } + ], + "evaluation_settings": "See linked reports for fixed simulation counts, baseline time budgets, paired opening seeds, and cutoffs.", + "served_artifact_sha256": "d5a7e3bda2966a79d22cf34f131976907260e679683d0e2680d935312f41c2f6" +} diff --git a/champion.pt b/champion.pt new file mode 100644 index 0000000000000000000000000000000000000000..901e505fee37dbe3c59d5e491ce2f0c625ecc6ad --- /dev/null +++ b/champion.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d5a7e3bda2966a79d22cf34f131976907260e679683d0e2680d935312f41c2f6 +size 4663985 diff --git a/comparison-checkpoints-manifest.json b/comparison-checkpoints-manifest.json new file mode 100644 index 0000000000000000000000000000000000000000..3f53ea9ca8fcef9f01d2e5618a2abc8a170839cf --- /dev/null +++ b/comparison-checkpoints-manifest.json @@ -0,0 +1,269 @@ +{ + "schema": "gipf-zero-comparison-checkpoints-v1", + "repository": "wuschelschulz/gipf-zero", + "tag": "20260909T235853Z", + "source_github_commit": "f4c83a5140625caba30aea6c045a11d92261f6be", + "source_github_url": "https://github.com/wusche1/gipf-zero/tree/f4c83a5140625caba30aea6c045a11d92261f6be", + "protocol": { + "training_seconds": 600, + "seeds": [ + 410101, + 410102 + ], + "architectures": [ + { + "name": "mlp256", + "kind": "mlp", + "head": "flat", + "width": 256, + "blocks": 2 + }, + { + "name": "squarecnn32", + "kind": "resnet", + "head": "flat", + "width": 32, + "blocks": 2 + }, + { + "name": "hexcnn32", + "kind": "hexresnet", + "head": "flat", + "width": 32, + "blocks": 2 + }, + { + "name": "querytransformer256x2", + "kind": "transformer", + "head": "query", + "width": 256, + "blocks": 2 + }, + { + "name": "querytransformer164x5", + "kind": "transformer", + "head": "query", + "width": 164, + "blocks": 5 + } + ], + "selfplay": { + "games": 64, + "games_per_iteration": 32, + "updates": 32, + "simulations": 32, + "batch_size": 256, + "lr": 0.001, + "max_ply": 300, + "replay_size": 100000, + "device": "cuda", + "native_forest": true + }, + "duels": { + "games": 12, + "seconds_cpu": 180, + "seconds_sim": 45, + "min_cpu_games_per_model": 96, + "equal_cpu_time": { + "device": "cpu", + "threads": 1, + "batch": 1, + "simulations": 10000, + "budget_ms": 50 + }, + "equal_simulations": { + "device": "cuda", + "threads": 2, + "batch": 64, + "simulations": 32, + "budget_ms": 0 + } + } + }, + "files": [ + { + "name": "mlp256-seed410101", + "seed": 410101, + "frozen_source": "checkpoints/overnight/20260909T235853Z/mlp256-seed410101-9acc8f0f15e5.pt", + "remote_path": "comparisons/20260909T235853Z/mlp256-seed410101-9acc8f0f15e5.pt", + "bytes": 11775385, + "summary_source_sha256": "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f", + "file_sha256": "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f", + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + }, + "games": 8740, + "iteration": 270 + }, + { + "name": "mlp256-seed410102", + "seed": 410102, + "frozen_source": "checkpoints/overnight/20260909T235853Z/mlp256-seed410102-9ca408753dfb.pt", + "remote_path": "comparisons/20260909T235853Z/mlp256-seed410102-9ca408753dfb.pt", + "bytes": 11775385, + "summary_source_sha256": "9ca408753dfbe4c2103016e9694be0e5033ed517459bf8a7cef2cb56164fa9a7", + "file_sha256": "9ca408753dfbe4c2103016e9694be0e5033ed517459bf8a7cef2cb56164fa9a7", + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + }, + "games": 8122, + "iteration": 251 + }, + { + "name": "squarecnn32-seed410101", + "seed": 410101, + "frozen_source": "checkpoints/overnight/20260909T235853Z/squarecnn32-seed410101-e2a8946b1405.pt", + "remote_path": "comparisons/20260909T235853Z/squarecnn32-seed410101-e2a8946b1405.pt", + "bytes": 14001446, + "summary_source_sha256": "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1", + "file_sha256": "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1", + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + }, + "games": 6214, + "iteration": 192 + }, + { + "name": "squarecnn32-seed410102", + "seed": 410102, + "frozen_source": "checkpoints/overnight/20260909T235853Z/squarecnn32-seed410102-d3e8eb4f1e75.pt", + "remote_path": "comparisons/20260909T235853Z/squarecnn32-seed410102-d3e8eb4f1e75.pt", + "bytes": 14001446, + "summary_source_sha256": "d3e8eb4f1e75d18bbeb28bef1b46c6c3f84b1ab6f009e63cc950a88f647c8025", + "file_sha256": "d3e8eb4f1e75d18bbeb28bef1b46c6c3f84b1ab6f009e63cc950a88f647c8025", + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + }, + "games": 8449, + "iteration": 260 + }, + { + "name": "hexcnn32-seed410101", + "seed": 410101, + "frozen_source": "checkpoints/overnight/20260909T235853Z/hexcnn32-seed410101-d6f1c7081cdb.pt", + "remote_path": "comparisons/20260909T235853Z/hexcnn32-seed410101-d6f1c7081cdb.pt", + "bytes": 14003159, + "summary_source_sha256": "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5", + "file_sha256": "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5", + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + }, + "games": 7877, + "iteration": 242 + }, + { + "name": "hexcnn32-seed410102", + "seed": 410102, + "frozen_source": "checkpoints/overnight/20260909T235853Z/hexcnn32-seed410102-c7d5bd9e9ee0.pt", + "remote_path": "comparisons/20260909T235853Z/hexcnn32-seed410102-c7d5bd9e9ee0.pt", + "bytes": 14003159, + "summary_source_sha256": "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d", + "file_sha256": "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d", + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + }, + "games": 6813, + "iteration": 209 + }, + { + "name": "querytransformer256x2-seed410101", + "seed": 410101, + "frozen_source": "checkpoints/overnight/20260909T235853Z/querytransformer256x2-seed410101-454d54089b4a.pt", + "remote_path": "comparisons/20260909T235853Z/querytransformer256x2-seed410101-454d54089b4a.pt", + "bytes": 13274255, + "summary_source_sha256": "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4", + "file_sha256": "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4", + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + }, + "games": 524, + "iteration": 15 + }, + { + "name": "querytransformer256x2-seed410102", + "seed": 410102, + "frozen_source": "checkpoints/overnight/20260909T235853Z/querytransformer256x2-seed410102-a0a1d4c39231.pt", + "remote_path": "comparisons/20260909T235853Z/querytransformer256x2-seed410102-a0a1d4c39231.pt", + "bytes": 13274191, + "summary_source_sha256": "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486", + "file_sha256": "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486", + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + }, + "games": 717, + "iteration": 22 + }, + { + "name": "querytransformer164x5-seed410101", + "seed": 410101, + "frozen_source": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt", + "remote_path": "comparisons/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt", + "bytes": 13479583, + "summary_source_sha256": "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99", + "file_sha256": "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99", + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + }, + "games": 582, + "iteration": 18 + }, + { + "name": "querytransformer164x5-seed410102", + "seed": 410102, + "frozen_source": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410102-02b9f1909248.pt", + "remote_path": "comparisons/20260909T235853Z/querytransformer164x5-seed410102-02b9f1909248.pt", + "bytes": 13479583, + "summary_source_sha256": "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c", + "file_sha256": "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c", + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + }, + "games": 476, + "iteration": 14 + } + ], + "total_files": 10, + "total_bytes": 133067592, + "source_github_commit_provenance": { + "kind": "inferred_repository_tree_at_backend_freeze", + "checkpoint_embedded_git_commit": null, + "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.", + "verified_backend_tree_sha256": { + "training/model.py": "1130a6f7e23524c04790adc702b8ecc7a9d92008bb354a398f377735b379dd71", + "training/native_search.py": "88beacc193f1fcfae082e46d93b3e1c8ca5edf76bf99c2c6298bd3d6cbe359f0", + "training/query_transformer.py": "9ff230547d8a0f9fd393fe05414a496724b3c514f76839a8f4c956e941f70385", + "training/search.py": "54587d2fde6bc62a671d9ff22ca2681f6fff521bbc5ed439c9cdd2b9c1786664", + "training/train.py": "8cbb2ed0b29d35c5101b6d9882cc2e83dc919277a32816b322a059abf97bd648" + } + } +} diff --git a/comparison-checkpoints-publication.json b/comparison-checkpoints-publication.json new file mode 100644 index 0000000000000000000000000000000000000000..576fc416c9c425afdf655efbdf2fd755efaf7242 --- /dev/null +++ b/comparison-checkpoints-publication.json @@ -0,0 +1,133 @@ +{ + "schema": "gipf-zero-comparison-publication-receipt-v1", + "repository": "wuschelschulz/gipf-zero", + "tag": "20260909T235853Z", + "repository_commit": "3996ec0178ef203edefeaa052cffa78c11eb2093", + "source_github_commit": "f4c83a5140625caba30aea6c045a11d92261f6be", + "source_github_url": "https://github.com/wusche1/gipf-zero/tree/f4c83a5140625caba30aea6c045a11d92261f6be", + "manifest": "comparison-checkpoints-manifest.json", + "remote_manifest_path": "comparisons/20260909T235853Z/manifest.json", + "remote_manifest_sha256": "19504f4cc194e7b527bcfe470a1c38e3103bb4d6decdc9af232e018a8a019a25", + "files_uploaded": 10, + "total_bytes": 133067592, + "network_protocol": { + "per_process_timeout_seconds": 180, + "maximum_retries": 2, + "attempts_per_file": 1 + }, + "remote_lfs_sha256_verification": true, + "files": [ + { + "name": "mlp256-seed410101", + "remote_path": "comparisons/20260909T235853Z/mlp256-seed410101-9acc8f0f15e5.pt", + "bytes": 11775385, + "local_sha256": "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f", + "remote_lfs_sha256": "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f", + "remote_lfs_size": 11775385, + "verified": true, + "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/mlp256-seed410101-9acc8f0f15e5.pt" + }, + { + "name": "mlp256-seed410102", + "remote_path": "comparisons/20260909T235853Z/mlp256-seed410102-9ca408753dfb.pt", + "bytes": 11775385, + "local_sha256": "9ca408753dfbe4c2103016e9694be0e5033ed517459bf8a7cef2cb56164fa9a7", + "remote_lfs_sha256": "9ca408753dfbe4c2103016e9694be0e5033ed517459bf8a7cef2cb56164fa9a7", + "remote_lfs_size": 11775385, + "verified": true, + "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/mlp256-seed410102-9ca408753dfb.pt" + }, + { + "name": "squarecnn32-seed410101", + "remote_path": "comparisons/20260909T235853Z/squarecnn32-seed410101-e2a8946b1405.pt", + "bytes": 14001446, + "local_sha256": "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1", + "remote_lfs_sha256": "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1", + "remote_lfs_size": 14001446, + "verified": true, + "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/squarecnn32-seed410101-e2a8946b1405.pt" + }, + { + "name": "squarecnn32-seed410102", + "remote_path": "comparisons/20260909T235853Z/squarecnn32-seed410102-d3e8eb4f1e75.pt", + "bytes": 14001446, + "local_sha256": "d3e8eb4f1e75d18bbeb28bef1b46c6c3f84b1ab6f009e63cc950a88f647c8025", + "remote_lfs_sha256": "d3e8eb4f1e75d18bbeb28bef1b46c6c3f84b1ab6f009e63cc950a88f647c8025", + "remote_lfs_size": 14001446, + "verified": true, + "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/squarecnn32-seed410102-d3e8eb4f1e75.pt" + }, + { + "name": "hexcnn32-seed410101", + "remote_path": "comparisons/20260909T235853Z/hexcnn32-seed410101-d6f1c7081cdb.pt", + "bytes": 14003159, + "local_sha256": "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5", + "remote_lfs_sha256": "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5", + "remote_lfs_size": 14003159, + "verified": true, + "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/hexcnn32-seed410101-d6f1c7081cdb.pt" + }, + { + "name": "hexcnn32-seed410102", + "remote_path": "comparisons/20260909T235853Z/hexcnn32-seed410102-c7d5bd9e9ee0.pt", + "bytes": 14003159, + "local_sha256": "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d", + "remote_lfs_sha256": "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d", + "remote_lfs_size": 14003159, + "verified": true, + "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/hexcnn32-seed410102-c7d5bd9e9ee0.pt" + }, + { + "name": "querytransformer256x2-seed410101", + "remote_path": "comparisons/20260909T235853Z/querytransformer256x2-seed410101-454d54089b4a.pt", + "bytes": 13274255, + "local_sha256": "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4", + "remote_lfs_sha256": "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4", + "remote_lfs_size": 13274255, + "verified": true, + "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/querytransformer256x2-seed410101-454d54089b4a.pt" + }, + { + "name": "querytransformer256x2-seed410102", + "remote_path": "comparisons/20260909T235853Z/querytransformer256x2-seed410102-a0a1d4c39231.pt", + "bytes": 13274191, + "local_sha256": "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486", + "remote_lfs_sha256": "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486", + "remote_lfs_size": 13274191, + "verified": true, + "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/querytransformer256x2-seed410102-a0a1d4c39231.pt" + }, + { + "name": "querytransformer164x5-seed410101", + "remote_path": "comparisons/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt", + "bytes": 13479583, + "local_sha256": "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99", + "remote_lfs_sha256": "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99", + "remote_lfs_size": 13479583, + "verified": true, + "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt" + }, + { + "name": "querytransformer164x5-seed410102", + "remote_path": "comparisons/20260909T235853Z/querytransformer164x5-seed410102-02b9f1909248.pt", + "bytes": 13479583, + "local_sha256": "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c", + "remote_lfs_sha256": "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c", + "remote_lfs_size": 13479583, + "verified": true, + "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/comparisons/20260909T235853Z/querytransformer164x5-seed410102-02b9f1909248.pt" + } + ], + "source_github_commit_provenance": { + "kind": "inferred_repository_tree_at_backend_freeze", + "checkpoint_embedded_git_commit": null, + "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.", + "verified_backend_tree_sha256": { + "training/model.py": "1130a6f7e23524c04790adc702b8ecc7a9d92008bb354a398f377735b379dd71", + "training/native_search.py": "88beacc193f1fcfae082e46d93b3e1c8ca5edf76bf99c2c6298bd3d6cbe359f0", + "training/query_transformer.py": "9ff230547d8a0f9fd393fe05414a496724b3c514f76839a8f4c956e941f70385", + "training/search.py": "54587d2fde6bc62a671d9ff22ca2681f6fff521bbc5ed439c9cdd2b9c1786664", + "training/train.py": "8cbb2ed0b29d35c5101b6d9882cc2e83dc919277a32816b322a059abf97bd648" + } + } +} diff --git a/config.json b/config.json new file mode 100644 index 0000000000000000000000000000000000000000..67bf05b9956120bf3767edf29f31507de3b162b0 --- /dev/null +++ b/config.json @@ -0,0 +1,6 @@ +{ + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" +} diff --git a/final-checkpoint-publication.json b/final-checkpoint-publication.json new file mode 100644 index 0000000000000000000000000000000000000000..f83909a331a7b511035a735148b34af77a8e85b9 --- /dev/null +++ b/final-checkpoint-publication.json @@ -0,0 +1,30 @@ +{ + "schema": "gipf-zero-final-checkpoint-publication-v1", + "repository": "wuschelschulz/gipf-zero", + "tag": "20260909T235853Z", + "checkpoint": "checkpoints/overnight/20260909T235853Z/selected-final-cb19bc298969.pt", + "remote_path": "finalists/20260909T235853Z/selected-final-cb19bc298969.pt", + "url": "https://huggingface.co/wuschelschulz/gipf-zero/resolve/main/finalists/20260909T235853Z/selected-final-cb19bc298969.pt", + "bytes": 14003543, + "file_sha256": "cb19bc298969043987463b267b909ca38c7ead84bab3c0b46571b6ef30e6d00f", + "remote_lfs_sha256": "cb19bc298969043987463b267b909ca38c7ead84bab3c0b46571b6ef30e6d00f", + "remote_lfs_size": 14003543, + "remote_lfs_sha256_verified": true, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + }, + "games": 93916, + "iteration": 2873, + "evaluation_pending": true, + "evaluation_report": "reports/overnight/20260909T235853Z/final-vs-served.json", + "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.", + "upload_protocol": { + "per_process_timeout_seconds": 180, + "maximum_retries": 1, + "network_upload_attempts": 1 + }, + "repository_commit": "8682c0c683117b697b3f1c36ff63008acf3295b8" +} diff --git a/final-vs-served.json b/final-vs-served.json new file mode 100644 index 0000000000000000000000000000000000000000..a507c3f256b2bb231be461a0b7328151092ebaac --- /dev/null +++ b/final-vs-served.json @@ -0,0 +1,720 @@ +{ + "candidate": "/workspace/gipf/checkpoints/overnight/20260909T235853Z/selected-final-cb19bc298969.pt", + "champion": "/workspace/gipf/checkpoints/overnight/20260909T235853Z/served-champion-d5a7e3bda296.pt", + "hashes": [ + "cb19bc298969043987463b267b909ca38c7ead84bab3c0b46571b6ef30e6d00f", + "d5a7e3bda2966a79d22cf34f131976907260e679683d0e2680d935312f41c2f6" + ], + "models": [ + { + "games": 93916, + "iteration": 2873, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 43869, + "iteration": 727, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 38, + "losses": 42, + "cutoffs": 0, + "unfinished": 0, + "win_rate_decisive": 0.475, + "wilson_95_decisive": [ + 0.36924733297407575, + 0.5830435735027006 + ], + "candidate_root_simulations": { + "count": 3526, + "mean": 251.41775382870108, + "min": 44, + "max": 10000 + }, + "champion_root_simulations": { + "count": 3523, + "mean": 243.38575078058471, + "min": 41, + "max": 10000 + }, + "candidate_search_seconds": { + "count": 3526, + "mean": 0.05023197256244182, + "min": 0.02666553482413292, + "max": 0.06883102655410767 + }, + "champion_search_seconds": { + "count": 3523, + "mean": 0.05029253078465464, + "min": 0.02774936333298683, + "max": 0.10136080533266068 + }, + "elapsed_seconds": 354.87011195346713, + "config": { + "candidate": "/workspace/gipf/checkpoints/overnight/20260909T235853Z/selected-final-cb19bc298969.pt", + "champion": "/workspace/gipf/checkpoints/overnight/20260909T235853Z/served-champion-d5a7e3bda296.pt", + "games": 80, + "batch": 1, + "simulations": 10000, + "seed": 950001, + "max_ply": 300, + "seconds": 600, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/final-vs-served.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 44 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 66 + }, + { + "game": 2, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 98 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 40 + }, + { + "game": 4, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 122 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 37 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 39 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 48 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 87 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 59 + }, + { + "game": 10, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 108 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 67 + }, + { + "game": 12, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 75 + }, + { + "game": 13, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 79 + }, + { + "game": 14, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 66 + }, + { + "game": 15, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 111 + }, + { + "game": 16, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 102 + }, + { + "game": 17, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 18, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 92 + }, + { + "game": 19, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 86 + }, + { + "game": 20, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 96 + }, + { + "game": 21, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 84 + }, + { + "game": 22, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 118 + }, + { + "game": 23, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 283 + }, + { + "game": 24, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 46 + }, + { + "game": 25, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 79 + }, + { + "game": 26, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 66 + }, + { + "game": 27, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 63 + }, + { + "game": 28, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 117 + }, + { + "game": 29, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 81 + }, + { + "game": 30, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 87 + }, + { + "game": 31, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 97 + }, + { + "game": 32, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 48 + }, + { + "game": 33, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 64 + }, + { + "game": 34, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 67 + }, + { + "game": 35, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 98 + }, + { + "game": 36, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 67 + }, + { + "game": 37, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 66 + }, + { + "game": 38, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 93 + }, + { + "game": 39, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 73 + }, + { + "game": 40, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 77 + }, + { + "game": 41, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 59 + }, + { + "game": 42, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 66 + }, + { + "game": 43, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 68 + }, + { + "game": 44, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 42 + }, + { + "game": 45, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 80 + }, + { + "game": 46, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 65 + }, + { + "game": 47, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 73 + }, + { + "game": 48, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 37 + }, + { + "game": 49, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 159 + }, + { + "game": 50, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 76 + }, + { + "game": 51, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 62 + }, + { + "game": 52, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 66 + }, + { + "game": 53, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 80 + }, + { + "game": 54, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 70 + }, + { + "game": 55, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 66 + }, + { + "game": 56, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 57, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 62 + }, + { + "game": 58, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 96 + }, + { + "game": 59, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 72 + }, + { + "game": 60, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 79 + }, + { + "game": 61, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 71 + }, + { + "game": 62, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 79 + }, + { + "game": 63, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 73 + }, + { + "game": 64, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 89 + }, + { + "game": 65, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 35 + }, + { + "game": 66, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 59 + }, + { + "game": 67, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 57 + }, + { + "game": 68, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 56 + }, + { + "game": 69, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 84 + }, + { + "game": 70, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 84 + }, + { + "game": 71, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 88 + }, + { + "game": 72, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 73, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 82 + }, + { + "game": 74, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 89 + }, + { + "game": 75, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 80 + }, + { + "game": 76, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 38 + }, + { + "game": 77, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 115 + }, + { + "game": 78, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 79 + }, + { + "game": 79, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 73 + } + ] +} diff --git a/final-vs-served.jsonl b/final-vs-served.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..3f23ee3f95dbbfa376df9c7e35dfd9ac071fef86 --- /dev/null +++ b/final-vs-served.jsonl @@ -0,0 +1,80 @@ +{"game": 0, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 44} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 66} +{"game": 2, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 98} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 40} +{"game": 4, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 122} +{"game": 5, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 37} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 39} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 48} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 87} +{"game": 9, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 59} +{"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 108} +{"game": 11, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 67} +{"game": 12, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 75} +{"game": 13, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 79} +{"game": 14, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 66} +{"game": 15, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 111} +{"game": 16, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 102} +{"game": 17, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 49} +{"game": 18, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 92} +{"game": 19, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 86} +{"game": 20, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 96} +{"game": 21, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 84} +{"game": 22, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 118} +{"game": 23, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 283} +{"game": 24, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 46} +{"game": 25, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 79} +{"game": 26, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 66} +{"game": 27, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 63} +{"game": 28, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 117} +{"game": 29, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 81} +{"game": 30, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 87} +{"game": 31, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 97} +{"game": 32, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 48} +{"game": 33, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 64} +{"game": 34, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 67} +{"game": 35, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 98} +{"game": 36, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 67} +{"game": 37, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 66} +{"game": 38, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 93} +{"game": 39, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 73} +{"game": 40, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 77} +{"game": 41, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 59} +{"game": 42, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 66} +{"game": 43, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 68} +{"game": 44, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 42} +{"game": 45, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 80} +{"game": 46, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 65} +{"game": 47, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 73} +{"game": 48, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 37} +{"game": 49, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 159} +{"game": 50, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 76} +{"game": 51, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 62} +{"game": 52, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 66} +{"game": 53, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 80} +{"game": 54, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 70} +{"game": 55, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 66} +{"game": 56, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41} +{"game": 57, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 62} +{"game": 58, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 96} +{"game": 59, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 72} +{"game": 60, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 79} +{"game": 61, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 71} +{"game": 62, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 79} +{"game": 63, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 73} +{"game": 64, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 89} +{"game": 65, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 35} +{"game": 66, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 59} +{"game": 67, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 57} +{"game": 68, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 56} +{"game": 69, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 84} +{"game": 70, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 84} +{"game": 71, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 88} +{"game": 72, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 49} +{"game": 73, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 82} +{"game": 74, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 89} +{"game": 75, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 80} +{"game": 76, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 38} +{"game": 77, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 115} +{"game": 78, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 79} +{"game": 79, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 73} diff --git a/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json b/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..617b8b92298014f91c200ce2b71e850d6e10dc8f --- /dev/null +++ b/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt", + "hashes": [ + "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5", + "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99" + ], + "models": [ + { + "games": 7877, + "iteration": 242, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 582, + "iteration": 18, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 8, + "losses": 0, + "cutoffs": 4, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.6755924351161198, + 1.0 + ], + "candidate_root_simulations": { + "count": 925, + "mean": 197.71243243243242, + "min": 49, + "max": 10000 + }, + "champion_root_simulations": { + "count": 911, + "mean": 16.644346871569702, + "min": 6, + "max": 21 + }, + "candidate_search_seconds": { + "count": 925, + "mean": 0.05026920535274454, + "min": 0.030910689383745193, + "max": 0.059041790664196014 + }, + "champion_search_seconds": { + "count": 911, + "mean": 0.0512864647921825, + "min": 0.05002184584736824, + "max": 0.06940874457359314 + }, + "elapsed_seconds": 93.33740724623203, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380101, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 71 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 50 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 19 + }, + { + "game": 3, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 20 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 37 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 54 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 45 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 34 + } + ] +} diff --git a/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl b/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..e22207c5bc59768971cf7a0c0462ac39992c6ef8 --- /dev/null +++ b/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 71} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 50} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 19} +{"game": 3, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 4, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 20} +{"game": 6, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 7, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 37} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 54} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 45} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 34} diff --git a/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json b/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..bb33baa410c7141eeca665344336d31e1dd26f63 --- /dev/null +++ b/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json @@ -0,0 +1,128 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt", + "hashes": [ + "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5", + "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99" + ], + "models": [ + { + "games": 7877, + "iteration": 242, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 582, + "iteration": 18, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 6, + "losses": 0, + "cutoffs": 0, + "unfinished": 6, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.6096657120978348, + 1.0 + ], + "candidate_root_simulations": { + "count": 1184, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 1172, + "mean": 31.907849829351537, + "min": 5, + "max": 32 + }, + "candidate_search_seconds": { + "count": 344, + "mean": 0.03558920219887134, + "min": 0.026320748031139374, + "max": 0.3319869302213192 + }, + "champion_search_seconds": { + "count": 344, + "mean": 0.09499198118181423, + "min": 0.020720209926366806, + "max": 0.18930847942829132 + }, + "elapsed_seconds": 45.010730154812336, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380101, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 28 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 32 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 47 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 50 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 71 + } + ] +} diff --git a/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.jsonl b/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..2c5ce2168b6e19078b618a27597a2a659b8ff251 --- /dev/null +++ b/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.jsonl @@ -0,0 +1,6 @@ +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 28} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 32} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41} +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 47} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 50} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 71} diff --git a/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json b/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json new file mode 100644 index 0000000000000000000000000000000000000000..50d44392a02a8e66ce4a995b98cf4e9338bbaa90 --- /dev/null +++ b/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json @@ -0,0 +1,176 @@ +{ + "candidate": "checkpoints/overnight/20260909T235853Z/hexcnn32-seed410101-d6f1c7081cdb.pt", + "champion": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt", + "hashes": [ + "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5", + "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99" + ], + "models": [ + { + "games": 7877, + "iteration": 242, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 582, + "iteration": 18, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 6, + "losses": 0, + "cutoffs": 6, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.6096657120978348, + 1.0 + ], + "candidate_root_simulations": { + "count": 1259, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 1247, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 369, + "mean": 0.033492098312552385, + "min": 0.0261453315615654, + "max": 0.28923046216368675 + }, + "champion_search_seconds": { + "count": 370, + "mean": 0.08941787915253961, + "min": 0.05584017187356949, + "max": 0.17711329832673073 + }, + "elapsed_seconds": 45.535510551184416, + "config": { + "candidate": "checkpoints/overnight/20260909T235853Z/hexcnn32-seed410101-d6f1c7081cdb.pt", + "champion": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380101, + "max_ply": 300, + "seconds": 180, + "output": "reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json", + "device": "cuda", + "threads": 1, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 28 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 32 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 47 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 50 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 71 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.jsonl b/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..e5b0bfea659f1572d254bafc4b344b6a58ea5202 --- /dev/null +++ b/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.jsonl @@ -0,0 +1,12 @@ +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 28} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 32} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41} +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 47} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 50} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 71} +{"game": 4, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 1, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 2, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 5, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 7, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 6, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json b/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..2ee3a6e0c9102bca4a475aa252c715c5bfca13f3 --- /dev/null +++ b/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt", + "hashes": [ + "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5", + "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4" + ], + "models": [ + { + "games": 7877, + "iteration": 242, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 524, + "iteration": 15, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + } + ], + "wins": 12, + "losses": 0, + "cutoffs": 0, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.7575059933447593, + 1.0 + ], + "candidate_root_simulations": { + "count": 213, + "mean": 1017.1032863849765, + "min": 61, + "max": 10000 + }, + "champion_root_simulations": { + "count": 196, + "mean": 22.714285714285715, + "min": 14, + "max": 27 + }, + "candidate_search_seconds": { + "count": 213, + "mean": 0.049378569102343256, + "min": 0.029349185526371002, + "max": 0.05088867247104645 + }, + "champion_search_seconds": { + "count": 196, + "mean": 0.051197532659434546, + "min": 0.05004933848977089, + "max": 0.055476345121860504 + }, + "elapsed_seconds": 20.587458580732346, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380101, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 45 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 20 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 33 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 22 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 55 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 20 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 23 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 40 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 28 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 23 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 28 + } + ] +} diff --git a/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl b/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..f31b2cbb8250241bfb99577e7214fb9c2355a450 --- /dev/null +++ b/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 45} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 20} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 33} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 22} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 55} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 20} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 23} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 40} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 49} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 28} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 23} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 28} diff --git a/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json b/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..48565522659bbc287f20de1634c56479b7fd8a85 --- /dev/null +++ b/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt", + "hashes": [ + "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5", + "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4" + ], + "models": [ + { + "games": 7877, + "iteration": 242, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 524, + "iteration": 15, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + } + ], + "wins": 9, + "losses": 2, + "cutoffs": 1, + "unfinished": 0, + "win_rate_decisive": 0.8181818181818182, + "wilson_95_decisive": [ + 0.523019438039156, + 0.9486323102539151 + ], + "candidate_root_simulations": { + "count": 359, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 352, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 210, + "mean": 0.035037198857892124, + "min": 0.002302609384059906, + "max": 0.31664491817355156 + }, + "champion_search_seconds": { + "count": 210, + "mean": 0.051772340440324374, + "min": 0.034844595938920975, + "max": 0.17626779153943062 + }, + "elapsed_seconds": 18.266604859381914, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380101, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 18 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 20 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 20 + }, + { + "game": 0, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 24 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 22 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 28 + }, + { + "game": 10, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 32 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 33 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 36 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 47 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 51 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.jsonl b/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..639c6683bb5aa9ca614e6a0ca8bb35bca5dc9632 --- /dev/null +++ b/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 18} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 20} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 20} +{"game": 0, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 24} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 22} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 28} +{"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 32} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 33} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 36} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 47} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 51} +{"game": 8, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json b/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..0ff0c28af3e9974b64dbbf167fc81508636afb28 --- /dev/null +++ b/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt", + "hashes": [ + "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d", + "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c" + ], + "models": [ + { + "games": 6813, + "iteration": 209, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 476, + "iteration": 14, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 11, + "losses": 0, + "cutoffs": 1, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.7411670330319684, + 1.0 + ], + "candidate_root_simulations": { + "count": 368, + "mean": 533.054347826087, + "min": 58, + "max": 10000 + }, + "champion_root_simulations": { + "count": 354, + "mean": 16.86723163841808, + "min": 8, + "max": 21 + }, + "candidate_search_seconds": { + "count": 368, + "mean": 0.04981631146865371, + "min": 0.029736820608377457, + "max": 0.05099663510918617 + }, + "champion_search_seconds": { + "count": 354, + "mean": 0.05114389503490453, + "min": 0.050041306763887405, + "max": 0.06353601440787315 + }, + "elapsed_seconds": 36.48766961693764, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380102, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 39 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 30 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 59 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 16 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 29 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 34 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 29 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 24 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 34 + } + ] +} diff --git a/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl b/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..e89352e9317ae6029a4a935248e007a98e705b1e --- /dev/null +++ b/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 39} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 30} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 59} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 16} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 29} +{"game": 5, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 34} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 29} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 24} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 34} diff --git a/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json b/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..ab5d83fe83d6e441365a5316bd3126063db29fed --- /dev/null +++ b/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt", + "hashes": [ + "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d", + "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c" + ], + "models": [ + { + "games": 6813, + "iteration": 209, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 476, + "iteration": 14, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 11, + "losses": 0, + "cutoffs": 1, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.7411670330319684, + 1.0 + ], + "candidate_root_simulations": { + "count": 367, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 353, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 208, + "mean": 0.03744002062684068, + "min": 0.0012018196284770966, + "max": 0.3364960700273514 + }, + "champion_search_seconds": { + "count": 207, + "mean": 0.08625425635904506, + "min": 0.06370929256081581, + "max": 0.20670939981937408 + }, + "elapsed_seconds": 25.690135687589645, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380102, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 19 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 20 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 22 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 27 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 29 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 30 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 34 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 34 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 35 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 37 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.jsonl b/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..5a655c4d0995acc8e5c4561cf7c5c351fa004cff --- /dev/null +++ b/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 19} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 20} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 22} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 27} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 29} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 30} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 34} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 34} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 35} +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 37} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 49} +{"game": 5, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json b/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..5711cf49115b8d1a63b0a99c8b6318cec408ed84 --- /dev/null +++ b/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt", + "hashes": [ + "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d", + "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486" + ], + "models": [ + { + "games": 6813, + "iteration": 209, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 717, + "iteration": 22, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + } + ], + "wins": 12, + "losses": 0, + "cutoffs": 0, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.7575059933447593, + 1.0 + ], + "candidate_root_simulations": { + "count": 117, + "mean": 1548.6581196581196, + "min": 64, + "max": 10000 + }, + "champion_root_simulations": { + "count": 110, + "mean": 21.654545454545456, + "min": 15, + "max": 25 + }, + "candidate_search_seconds": { + "count": 117, + "mean": 0.049162707172143154, + "min": 0.031186256557703018, + "max": 0.05291455611586571 + }, + "champion_search_seconds": { + "count": 110, + "mean": 0.05114460059187629, + "min": 0.05002905800938606, + "max": 0.05573977157473564 + }, + "elapsed_seconds": 11.402687206864357, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380102, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 21 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 16 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 29 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 30 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 21 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 16 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 13 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 18 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 17 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 24 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 17 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 16 + } + ] +} diff --git a/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl b/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..2957fe7dadcdb3313917245962672b35644e307c --- /dev/null +++ b/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 21} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 16} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 29} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 30} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 21} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 16} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 13} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 18} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 17} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 24} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 17} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 16} diff --git a/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json b/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..c069c6b289fc2e6f228a5ebe1e898c655c983fb7 --- /dev/null +++ b/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt", + "hashes": [ + "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d", + "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486" + ], + "models": [ + { + "games": 6813, + "iteration": 209, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 717, + "iteration": 22, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + } + ], + "wins": 11, + "losses": 0, + "cutoffs": 1, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.7411670330319684, + 1.0 + ], + "candidate_root_simulations": { + "count": 339, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 324, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 205, + "mean": 0.03397062422662246, + "min": 0.001332070678472519, + "max": 0.30108968541026115 + }, + "champion_search_seconds": { + "count": 203, + "mean": 0.05458392078066107, + "min": 0.03652486577630043, + "max": 0.17201025411486626 + }, + "elapsed_seconds": 18.07961431518197, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380102, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 15 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 17 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 17 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 22 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 22 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 24 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 31 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 33 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 34 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 50 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.jsonl b/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..1d66e3923424b86817934f3f385fce3fe1b15508 --- /dev/null +++ b/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 15} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 17} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 17} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 22} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 22} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 24} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 31} +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 33} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 34} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 50} +{"game": 11, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/league-024737-duel.json b/league-024737-duel.json new file mode 100644 index 0000000000000000000000000000000000000000..0f11c8dfde1fa761bbc640bbd0751df388abb73b --- /dev/null +++ b/league-024737-duel.json @@ -0,0 +1,720 @@ +{ + "candidate": "/workspace/gipf/checkpoints/league/candidate-024737.pt", + "champion": "/workspace/gipf/checkpoints/league/champion-024737.pt", + "hashes": [ + "2afe8b00ed3ef5b5052f28e2ede2853ed79ff76e1b843492e4d1e2d83059adbb", + "d5a7e3bda2966a79d22cf34f131976907260e679683d0e2680d935312f41c2f6" + ], + "models": [ + { + "games": 21515, + "iteration": 658, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 43869, + "iteration": 727, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 13, + "losses": 65, + "cutoffs": 2, + "unfinished": 0, + "win_rate_decisive": 0.16666666666666666, + "wilson_95_decisive": [ + 0.1000694541672165, + 0.26455575230757966 + ], + "candidate_root_simulations": { + "count": 3471, + "mean": 168.01267646211465, + "min": 14, + "max": 10000 + }, + "champion_root_simulations": { + "count": 3545, + "mean": 289.8758815232722, + "min": 42, + "max": 10000 + }, + "candidate_search_seconds": { + "count": 3471, + "mean": 0.050376033062156635, + "min": 0.025996044278144836, + "max": 0.07281234115362167 + }, + "champion_search_seconds": { + "count": 3545, + "mean": 0.05013121535163504, + "min": 0.02710983157157898, + "max": 0.06725717708468437 + }, + "elapsed_seconds": 353.1475488767028, + "config": { + "candidate": "/workspace/gipf/checkpoints/league/candidate-024737.pt", + "champion": "/workspace/gipf/checkpoints/league/champion-024737.pt", + "games": 80, + "batch": 1, + "simulations": 10000, + "seed": 324737, + "max_ply": 300, + "seconds": 1200, + "output": "/workspace/gipf/reports/league-024737-duel.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 78 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 75 + }, + { + "game": 2, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 50 + }, + { + "game": 3, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 79 + }, + { + "game": 4, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 74 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 91 + }, + { + "game": 6, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 48 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 77 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 39 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 63 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 89 + }, + { + "game": 12, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 82 + }, + { + "game": 13, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 63 + }, + { + "game": 14, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 85 + }, + { + "game": 15, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 69 + }, + { + "game": 16, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 64 + }, + { + "game": 17, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 85 + }, + { + "game": 18, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 74 + }, + { + "game": 19, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 51 + }, + { + "game": 20, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 94 + }, + { + "game": 21, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 55 + }, + { + "game": 22, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 97 + }, + { + "game": 23, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 75 + }, + { + "game": 24, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 84 + }, + { + "game": 25, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 59 + }, + { + "game": 26, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 77 + }, + { + "game": 27, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 47 + }, + { + "game": 28, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 106 + }, + { + "game": 29, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 61 + }, + { + "game": 30, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 42 + }, + { + "game": 31, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 65 + }, + { + "game": 32, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 38 + }, + { + "game": 33, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 73 + }, + { + "game": 34, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 21 + }, + { + "game": 35, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 63 + }, + { + "game": 36, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 164 + }, + { + "game": 37, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 89 + }, + { + "game": 38, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 70 + }, + { + "game": 39, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 113 + }, + { + "game": 40, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 95 + }, + { + "game": 41, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 78 + }, + { + "game": 42, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 73 + }, + { + "game": 43, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 59 + }, + { + "game": 44, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 66 + }, + { + "game": 45, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 93 + }, + { + "game": 46, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 74 + }, + { + "game": 47, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 63 + }, + { + "game": 48, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 53 + }, + { + "game": 49, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 105 + }, + { + "game": 50, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 86 + }, + { + "game": 51, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 52, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 42 + }, + { + "game": 53, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 100 + }, + { + "game": 54, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 58 + }, + { + "game": 55, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 43 + }, + { + "game": 56, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 57, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 83 + }, + { + "game": 58, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 133 + }, + { + "game": 59, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 60, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 103 + }, + { + "game": 61, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 69 + }, + { + "game": 62, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 54 + }, + { + "game": 63, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 81 + }, + { + "game": 64, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 38 + }, + { + "game": 65, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 113 + }, + { + "game": 66, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 54 + }, + { + "game": 67, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 81 + }, + { + "game": 68, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 72 + }, + { + "game": 69, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 43 + }, + { + "game": 70, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 28 + }, + { + "game": 71, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 71 + }, + { + "game": 72, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 57 + }, + { + "game": 73, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 33 + }, + { + "game": 74, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 62 + }, + { + "game": 75, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 76, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 74 + }, + { + "game": 77, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 86 + }, + { + "game": 78, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 84 + }, + { + "game": 79, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 35 + } + ] +} diff --git a/league-032337-duel.json b/league-032337-duel.json new file mode 100644 index 0000000000000000000000000000000000000000..51ebd09f88365587773014ab8ebe7fd6fc493cb9 --- /dev/null +++ b/league-032337-duel.json @@ -0,0 +1,720 @@ +{ + "candidate": "/workspace/gipf/checkpoints/league/candidate-032337.pt", + "champion": "/workspace/gipf/checkpoints/league/champion-032337.pt", + "hashes": [ + "235d1e4d4ea756e4934f9e308a161c72698a05678457adb55efc2507d9f7f3e9", + "d5a7e3bda2966a79d22cf34f131976907260e679683d0e2680d935312f41c2f6" + ], + "models": [ + { + "games": 38076, + "iteration": 1165, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 43869, + "iteration": 727, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 31, + "losses": 49, + "cutoffs": 0, + "unfinished": 0, + "win_rate_decisive": 0.3875, + "wilson_95_decisive": [ + 0.2882456512530508, + 0.49706342789244295 + ], + "candidate_root_simulations": { + "count": 3476, + "mean": 226.86104718066744, + "min": 41, + "max": 10000 + }, + "champion_root_simulations": { + "count": 3503, + "mean": 240.3930916357408, + "min": 51, + "max": 10000 + }, + "candidate_search_seconds": { + "count": 3476, + "mean": 0.05028459619542089, + "min": 0.029697582125663757, + "max": 0.061029355973005295 + }, + "champion_search_seconds": { + "count": 3503, + "mean": 0.050180195522093955, + "min": 0.025200344622135162, + "max": 0.06006953492760658 + }, + "elapsed_seconds": 351.11589351668954, + "config": { + "candidate": "/workspace/gipf/checkpoints/league/candidate-032337.pt", + "champion": "/workspace/gipf/checkpoints/league/champion-032337.pt", + "games": 80, + "batch": 1, + "simulations": 10000, + "seed": 332337, + "max_ply": 300, + "seconds": 1200, + "output": "/workspace/gipf/reports/league-032337-duel.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 73 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 140 + }, + { + "game": 2, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 48 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 46 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 43 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 74 + }, + { + "game": 6, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 62 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 36 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 53 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 97 + }, + { + "game": 10, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 50 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 86 + }, + { + "game": 12, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 77 + }, + { + "game": 13, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 87 + }, + { + "game": 14, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 79 + }, + { + "game": 15, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 79 + }, + { + "game": 16, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 78 + }, + { + "game": 17, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 94 + }, + { + "game": 18, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 82 + }, + { + "game": 19, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 81 + }, + { + "game": 20, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 40 + }, + { + "game": 21, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 30 + }, + { + "game": 22, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 110 + }, + { + "game": 23, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 48 + }, + { + "game": 24, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 78 + }, + { + "game": 25, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 85 + }, + { + "game": 26, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 71 + }, + { + "game": 27, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 56 + }, + { + "game": 28, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 118 + }, + { + "game": 29, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 80 + }, + { + "game": 30, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 59 + }, + { + "game": 31, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 93 + }, + { + "game": 32, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 90 + }, + { + "game": 33, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 105 + }, + { + "game": 34, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 82 + }, + { + "game": 35, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 56 + }, + { + "game": 36, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 73 + }, + { + "game": 37, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 101 + }, + { + "game": 38, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 69 + }, + { + "game": 39, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 122 + }, + { + "game": 40, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 60 + }, + { + "game": 41, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 101 + }, + { + "game": 42, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 68 + }, + { + "game": 43, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 113 + }, + { + "game": 44, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 97 + }, + { + "game": 45, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 78 + }, + { + "game": 46, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 88 + }, + { + "game": 47, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 39 + }, + { + "game": 48, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 82 + }, + { + "game": 49, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 53 + }, + { + "game": 50, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 61 + }, + { + "game": 51, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 55 + }, + { + "game": 52, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 55 + }, + { + "game": 53, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 69 + }, + { + "game": 54, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 46 + }, + { + "game": 55, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 61 + }, + { + "game": 56, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 64 + }, + { + "game": 57, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 76 + }, + { + "game": 58, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 59, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 59 + }, + { + "game": 60, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 34 + }, + { + "game": 61, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 80 + }, + { + "game": 62, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 71 + }, + { + "game": 63, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 71 + }, + { + "game": 64, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 72 + }, + { + "game": 65, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 101 + }, + { + "game": 66, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 90 + }, + { + "game": 67, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 80 + }, + { + "game": 68, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 94 + }, + { + "game": 69, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 59 + }, + { + "game": 70, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 116 + }, + { + "game": 71, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 93 + }, + { + "game": 72, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 54 + }, + { + "game": 73, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 91 + }, + { + "game": 74, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 108 + }, + { + "game": 75, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 83 + }, + { + "game": 76, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 104 + }, + { + "game": 77, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 95 + }, + { + "game": 78, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 85 + }, + { + "game": 79, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 113 + } + ] +} diff --git a/league-035932-duel.json b/league-035932-duel.json new file mode 100644 index 0000000000000000000000000000000000000000..e10bef6d9e0ef6719deea0568b00daf28464908e --- /dev/null +++ b/league-035932-duel.json @@ -0,0 +1,720 @@ +{ + "candidate": "/workspace/gipf/checkpoints/league/candidate-035932.pt", + "champion": "/workspace/gipf/checkpoints/league/champion-035932.pt", + "hashes": [ + "319276139c115ed2302464201c50aad4a2244d84fc4671811d0bc1a1a6846553", + "d5a7e3bda2966a79d22cf34f131976907260e679683d0e2680d935312f41c2f6" + ], + "models": [ + { + "games": 54044, + "iteration": 1654, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 43869, + "iteration": 727, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 34, + "losses": 46, + "cutoffs": 0, + "unfinished": 0, + "win_rate_decisive": 0.425, + "wilson_95_decisive": [ + 0.3225656178960178, + 0.5343071015343114 + ], + "candidate_root_simulations": { + "count": 3342, + "mean": 239.0718132854578, + "min": 12, + "max": 10000 + }, + "champion_root_simulations": { + "count": 3363, + "mean": 273.3193577163247, + "min": 42, + "max": 10000 + }, + "candidate_search_seconds": { + "count": 3342, + "mean": 0.05027993674819731, + "min": 0.028711620718240738, + "max": 0.08356250822544098 + }, + "champion_search_seconds": { + "count": 3363, + "mean": 0.05021213266999637, + "min": 0.02769080176949501, + "max": 0.0755114033818245 + }, + "elapsed_seconds": 337.44456908106804, + "config": { + "candidate": "/workspace/gipf/checkpoints/league/candidate-035932.pt", + "champion": "/workspace/gipf/checkpoints/league/champion-035932.pt", + "games": 80, + "batch": 1, + "simulations": 10000, + "seed": 335932, + "max_ply": 300, + "seconds": 1200, + "output": "/workspace/gipf/reports/league-035932-duel.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 47 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 59 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 93 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 90 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 95 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 47 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 55 + }, + { + "game": 8, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 82 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 62 + }, + { + "game": 10, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 80 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 94 + }, + { + "game": 12, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 58 + }, + { + "game": 13, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 121 + }, + { + "game": 14, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 67 + }, + { + "game": 15, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 113 + }, + { + "game": 16, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 117 + }, + { + "game": 17, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 50 + }, + { + "game": 18, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 80 + }, + { + "game": 19, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 37 + }, + { + "game": 20, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 84 + }, + { + "game": 21, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 63 + }, + { + "game": 22, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 125 + }, + { + "game": 23, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 68 + }, + { + "game": 24, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 59 + }, + { + "game": 25, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 81 + }, + { + "game": 26, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 68 + }, + { + "game": 27, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 35 + }, + { + "game": 28, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 64 + }, + { + "game": 29, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 80 + }, + { + "game": 30, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 70 + }, + { + "game": 31, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 95 + }, + { + "game": 32, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 55 + }, + { + "game": 33, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 38 + }, + { + "game": 34, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 50 + }, + { + "game": 35, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 74 + }, + { + "game": 36, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 44 + }, + { + "game": 37, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 93 + }, + { + "game": 38, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 71 + }, + { + "game": 39, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 67 + }, + { + "game": 40, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 88 + }, + { + "game": 41, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 79 + }, + { + "game": 42, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 105 + }, + { + "game": 43, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 83 + }, + { + "game": 44, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 84 + }, + { + "game": 45, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 93 + }, + { + "game": 46, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 60 + }, + { + "game": 47, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 77 + }, + { + "game": 48, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 58 + }, + { + "game": 49, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 67 + }, + { + "game": 50, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 78 + }, + { + "game": 51, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 36 + }, + { + "game": 52, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 94 + }, + { + "game": 53, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 58 + }, + { + "game": 54, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 66 + }, + { + "game": 55, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 55 + }, + { + "game": 56, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 97 + }, + { + "game": 57, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 80 + }, + { + "game": 58, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 63 + }, + { + "game": 59, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 73 + }, + { + "game": 60, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 81 + }, + { + "game": 61, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 43 + }, + { + "game": 62, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 44 + }, + { + "game": 63, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 59 + }, + { + "game": 64, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 96 + }, + { + "game": 65, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 93 + }, + { + "game": 66, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 69 + }, + { + "game": 67, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 75 + }, + { + "game": 68, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 64 + }, + { + "game": 69, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 90 + }, + { + "game": 70, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 118 + }, + { + "game": 71, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 67 + }, + { + "game": 72, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 81 + }, + { + "game": 73, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 56 + }, + { + "game": 74, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 75, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 76 + }, + { + "game": 76, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 115 + }, + { + "game": 77, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 65 + }, + { + "game": 78, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 112 + }, + { + "game": 79, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 76 + } + ] +} diff --git a/league-043512-duel.json b/league-043512-duel.json new file mode 100644 index 0000000000000000000000000000000000000000..a090b7d4b459e882fdecc8ee8f949a89da79be18 --- /dev/null +++ b/league-043512-duel.json @@ -0,0 +1,720 @@ +{ + "candidate": "/workspace/gipf/checkpoints/league/candidate-043512.pt", + "champion": "/workspace/gipf/checkpoints/league/champion-043512.pt", + "hashes": [ + "2cd04db0930f19b5dc231738b97607bc1dd88cbf3b65a00d9dd49e7279aeb788", + "d5a7e3bda2966a79d22cf34f131976907260e679683d0e2680d935312f41c2f6" + ], + "models": [ + { + "games": 69700, + "iteration": 2133, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 43869, + "iteration": 727, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 39, + "losses": 41, + "cutoffs": 0, + "unfinished": 0, + "win_rate_decisive": 0.4875, + "wilson_95_decisive": [ + 0.38107879540381084, + 0.5950666578345775 + ], + "candidate_root_simulations": { + "count": 3276, + "mean": 253.34615384615384, + "min": 44, + "max": 10000 + }, + "champion_root_simulations": { + "count": 3293, + "mean": 232.6450045551169, + "min": 46, + "max": 10000 + }, + "candidate_search_seconds": { + "count": 3276, + "mean": 0.050195788872520346, + "min": 0.028529252856969833, + "max": 0.06653248891234398 + }, + "champion_search_seconds": { + "count": 3293, + "mean": 0.050200892989618454, + "min": 0.029827728867530823, + "max": 0.06366511434316635 + }, + "elapsed_seconds": 330.2759933806956, + "config": { + "candidate": "/workspace/gipf/checkpoints/league/candidate-043512.pt", + "champion": "/workspace/gipf/checkpoints/league/champion-043512.pt", + "games": 80, + "batch": 1, + "simulations": 10000, + "seed": 343512, + "max_ply": 300, + "seconds": 1200, + "output": "/workspace/gipf/reports/league-043512-duel.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 55 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 56 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 67 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 88 + }, + { + "game": 4, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 90 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 102 + }, + { + "game": 6, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 86 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 57 + }, + { + "game": 8, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 76 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 73 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 47 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 64 + }, + { + "game": 12, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 63 + }, + { + "game": 13, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 132 + }, + { + "game": 14, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 56 + }, + { + "game": 15, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 56 + }, + { + "game": 16, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 46 + }, + { + "game": 17, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 69 + }, + { + "game": 18, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 94 + }, + { + "game": 19, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 112 + }, + { + "game": 20, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 76 + }, + { + "game": 21, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 78 + }, + { + "game": 22, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 80 + }, + { + "game": 23, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 64 + }, + { + "game": 24, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 61 + }, + { + "game": 25, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 63 + }, + { + "game": 26, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 65 + }, + { + "game": 27, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 87 + }, + { + "game": 28, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 106 + }, + { + "game": 29, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 101 + }, + { + "game": 30, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 30 + }, + { + "game": 31, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 71 + }, + { + "game": 32, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 71 + }, + { + "game": 33, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 67 + }, + { + "game": 34, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 47 + }, + { + "game": 35, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 32 + }, + { + "game": 36, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 73 + }, + { + "game": 37, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 105 + }, + { + "game": 38, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 77 + }, + { + "game": 39, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 81 + }, + { + "game": 40, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 81 + }, + { + "game": 41, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 111 + }, + { + "game": 42, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 72 + }, + { + "game": 43, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 42 + }, + { + "game": 44, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 73 + }, + { + "game": 45, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 66 + }, + { + "game": 46, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 77 + }, + { + "game": 47, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 57 + }, + { + "game": 48, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 76 + }, + { + "game": 49, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 70 + }, + { + "game": 50, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 51, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 85 + }, + { + "game": 52, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 104 + }, + { + "game": 53, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 59 + }, + { + "game": 54, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 46 + }, + { + "game": 55, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 40 + }, + { + "game": 56, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 57, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 65 + }, + { + "game": 58, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 80 + }, + { + "game": 59, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 123 + }, + { + "game": 60, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 51 + }, + { + "game": 61, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 89 + }, + { + "game": 62, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 58 + }, + { + "game": 63, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 84 + }, + { + "game": 64, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 92 + }, + { + "game": 65, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 81 + }, + { + "game": 66, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 67, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 100 + }, + { + "game": 68, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 40 + }, + { + "game": 69, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 70, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 75 + }, + { + "game": 71, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 104 + }, + { + "game": 72, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 70 + }, + { + "game": 73, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 86 + }, + { + "game": 74, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 62 + }, + { + "game": 75, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 33 + }, + { + "game": 76, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 95 + }, + { + "game": 77, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 71 + }, + { + "game": 78, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 53 + }, + { + "game": 79, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 84 + } + ] +} diff --git a/mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json b/mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..d5cb32e3fa7f0dd47e534e1a3b08fec50f7409ef --- /dev/null +++ b/mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt", + "hashes": [ + "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f", + "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5" + ], + "models": [ + { + "games": 8740, + "iteration": 270, + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 7877, + "iteration": 242, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 2, + "losses": 8, + "cutoffs": 2, + "unfinished": 0, + "win_rate_decisive": 0.2, + "wilson_95_decisive": [ + 0.056682151454375246, + 0.5098375284633582 + ], + "candidate_root_simulations": { + "count": 801, + "mean": 228.27715355805245, + "min": 100, + "max": 10000 + }, + "champion_root_simulations": { + "count": 801, + "mean": 201.11610486891385, + "min": 41, + "max": 10000 + }, + "candidate_search_seconds": { + "count": 801, + "mean": 0.05022746998440014, + "min": 0.030646514147520065, + "max": 0.06282975524663925 + }, + "champion_search_seconds": { + "count": 801, + "mean": 0.05027599407176251, + "min": 0.02731853723526001, + "max": 0.05738721042871475 + }, + "elapsed_seconds": 80.68824608251452, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380101, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 68 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 57 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 79 + }, + { + "game": 3, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 73 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 67 + }, + { + "game": 6, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 118 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 72 + }, + { + "game": 8, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 60 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 10, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 66 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 91 + } + ] +} diff --git a/mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl b/mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..e1fa81627ca08d671fef9b32e9ac6abe29d1ab74 --- /dev/null +++ b/mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 68} +{"game": 1, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 57} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 79} +{"game": 3, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 73} +{"game": 4, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 5, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 67} +{"game": 6, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 118} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 72} +{"game": 8, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 60} +{"game": 9, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 66} +{"game": 11, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 91} diff --git a/mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json b/mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..c2df08b4e610ec45f883cb0aa0c640ba8a98990c --- /dev/null +++ b/mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt", + "hashes": [ + "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f", + "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5" + ], + "models": [ + { + "games": 8740, + "iteration": 270, + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 7877, + "iteration": 242, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 1, + "losses": 9, + "cutoffs": 2, + "unfinished": 0, + "win_rate_decisive": 0.1, + "wilson_95_decisive": [ + 0.017876213095072896, + 0.40415002679523837 + ], + "candidate_root_simulations": { + "count": 731, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 739, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 328, + "mean": 0.014640243871637234, + "min": 0.010727941989898682, + "max": 0.12138830497860909 + }, + "champion_search_seconds": { + "count": 328, + "mean": 0.03528110070790096, + "min": 0.02560272067785263, + "max": 0.18772567063570023 + }, + "elapsed_seconds": 16.43799015507102, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380101, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 8, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 24 + }, + { + "game": 3, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 29 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 37 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 39 + }, + { + "game": 0, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 44 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 6, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 68 + }, + { + "game": 2, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 70 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 99 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 191 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.jsonl b/mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..b15796aacb1626d58385e6f9e58ab1440c31b644 --- /dev/null +++ b/mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 8, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 24} +{"game": 3, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 29} +{"game": 11, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 37} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 39} +{"game": 0, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 44} +{"game": 1, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 49} +{"game": 6, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 68} +{"game": 2, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 70} +{"game": 9, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 99} +{"game": 5, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 191} +{"game": 4, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 7, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json b/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..8bec4d2cb6755cd79b4909de60bd660e08ee10a4 --- /dev/null +++ b/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt", + "hashes": [ + "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f", + "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99" + ], + "models": [ + { + "games": 8740, + "iteration": 270, + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 582, + "iteration": 18, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 11, + "losses": 0, + "cutoffs": 1, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.7411670330319684, + 1.0 + ], + "candidate_root_simulations": { + "count": 449, + "mean": 538.0890868596882, + "min": 89, + "max": 10000 + }, + "champion_root_simulations": { + "count": 426, + "mean": 15.556338028169014, + "min": 7, + "max": 21 + }, + "candidate_search_seconds": { + "count": 449, + "mean": 0.04968965270936091, + "min": 0.027370132505893707, + "max": 0.056996483355760574 + }, + "champion_search_seconds": { + "count": 426, + "mean": 0.05177791051986352, + "min": 0.05002857372164726, + "max": 0.07907987758517265 + }, + "elapsed_seconds": 44.45715307071805, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380101, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 33 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 32 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 42 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 40 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 32 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 39 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 108 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 53 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 24 + } + ] +} diff --git a/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl b/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..f79a88fd4cc7c9d1ca4b7956ed461770c54f7d45 --- /dev/null +++ b/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 33} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 32} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 42} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 49} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 40} +{"game": 6, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 32} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 39} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 108} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 53} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 24} diff --git a/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json b/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..e1503bdc3975c393a6ff7be0a2138c0bd8001af8 --- /dev/null +++ b/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt", + "hashes": [ + "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f", + "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99" + ], + "models": [ + { + "games": 8740, + "iteration": 270, + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 582, + "iteration": 18, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 11, + "losses": 1, + "cutoffs": 0, + "unfinished": 0, + "win_rate_decisive": 0.9166666666666666, + "wilson_95_decisive": [ + 0.6461200888588832, + 0.9851349055950829 + ], + "candidate_root_simulations": { + "count": 473, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 454, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 171, + "mean": 0.013640849104924508, + "min": 0.0005120299756526947, + "max": 0.12017641961574554 + }, + "champion_search_seconds": { + "count": 163, + "mean": 0.10111915811530651, + "min": 0.04979807510972023, + "max": 0.2054123431444168 + }, + "elapsed_seconds": 18.855112049728632, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380101, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 5, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 31 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 32 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 46 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 64 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 66 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 71 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 91 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 115 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 184 + } + ] +} diff --git a/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.jsonl b/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..83303a161035623d3b75483605addf903a7f1e57 --- /dev/null +++ b/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 5, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 31} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 32} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 46} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 64} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 66} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 71} +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 91} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 115} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 184} diff --git a/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json b/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..69d72b6d51fcd5b3115e20cc261a3559dcf9f1c6 --- /dev/null +++ b/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt", + "hashes": [ + "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f", + "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4" + ], + "models": [ + { + "games": 8740, + "iteration": 270, + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 524, + "iteration": 15, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + } + ], + "wins": 10, + "losses": 0, + "cutoffs": 2, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.7224672001371109, + 1.0 + ], + "candidate_root_simulations": { + "count": 812, + "mean": 358.23522167487687, + "min": 123, + "max": 10000 + }, + "champion_root_simulations": { + "count": 790, + "mean": 21.61139240506329, + "min": 10, + "max": 28 + }, + "candidate_search_seconds": { + "count": 812, + "mean": 0.050130335912291935, + "min": 0.03028266131877899, + "max": 0.10844805836677551 + }, + "champion_search_seconds": { + "count": 790, + "mean": 0.05129407604661169, + "min": 0.05002252012491226, + "max": 0.06667163968086243 + }, + "elapsed_seconds": 81.39602221176028, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380101, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 189 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 76 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 57 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 104 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 69 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 36 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 61 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 66 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 63 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl b/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..f92c97a5ad97a26b4633635a84fb0efbc06d5164 --- /dev/null +++ b/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41} +{"game": 1, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 189} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 76} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 57} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 104} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 69} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 36} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 61} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 66} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 63} +{"game": 11, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json b/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..706bfa5aca063c8d8e36a4bb9221d4dda875e9db --- /dev/null +++ b/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt", + "hashes": [ + "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f", + "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4" + ], + "models": [ + { + "games": 8740, + "iteration": 270, + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 524, + "iteration": 15, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + } + ], + "wins": 5, + "losses": 2, + "cutoffs": 5, + "unfinished": 0, + "win_rate_decisive": 0.7142857142857143, + "wilson_95_decisive": [ + 0.3589344518326194, + 0.9177810759959433 + ], + "candidate_root_simulations": { + "count": 1245, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 1217, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 372, + "mean": 0.014497957342574674, + "min": 0.010189041495323181, + "max": 0.10919997468590736 + }, + "champion_search_seconds": { + "count": 373, + "mean": 0.07265937728832299, + "min": 0.037330176681280136, + "max": 0.17097918316721916 + }, + "elapsed_seconds": 32.5906348451972, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380101, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 4, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 38 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 46 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 50 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 60 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 72 + }, + { + "game": 6, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 90 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 176 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.jsonl b/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..3004c5d3d6267ba62990308248916d2a21f7638a --- /dev/null +++ b/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 4, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 38} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 46} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 50} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 60} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 72} +{"game": 6, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 90} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 176} +{"game": 2, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 7, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 10, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 0, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 8, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json b/mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..702269d72b03f965bc95de4940c78112d7ead586 --- /dev/null +++ b/mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "hashes": [ + "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f", + "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1" + ], + "models": [ + { + "games": 8740, + "iteration": 270, + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 6214, + "iteration": 192, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 3, + "losses": 9, + "cutoffs": 0, + "unfinished": 0, + "win_rate_decisive": 0.25, + "wilson_95_decisive": [ + 0.08894166839405476, + 0.5323053349335656 + ], + "candidate_root_simulations": { + "count": 975, + "mean": 268.34871794871793, + "min": 119, + "max": 10000 + }, + "champion_root_simulations": { + "count": 981, + "mean": 199.56574923547402, + "min": 42, + "max": 10000 + }, + "candidate_search_seconds": { + "count": 975, + "mean": 0.05015508570732215, + "min": 0.02668626606464386, + "max": 0.05868809297680855 + }, + "champion_search_seconds": { + "count": 981, + "mean": 0.050266843832171046, + "min": 0.03076682612299919, + "max": 0.06585546210408211 + }, + "elapsed_seconds": 98.44315692782402, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380101, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 126 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 51 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 199 + }, + { + "game": 3, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 125 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 87 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 128 + }, + { + "game": 6, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 212 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 245 + }, + { + "game": 8, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 128 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 57 + }, + { + "game": 10, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 154 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 131 + } + ] +} diff --git a/mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl b/mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..7013e0528a4a269a70eed0bffb040c4ab4209fc4 --- /dev/null +++ b/mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 126} +{"game": 1, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 51} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 199} +{"game": 3, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 125} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 87} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 128} +{"game": 6, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 212} +{"game": 7, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 245} +{"game": 8, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 128} +{"game": 9, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 57} +{"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 154} +{"game": 11, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 131} diff --git a/mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.json b/mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..99d50c335037aa83de637f40c9c160d6a18bd54e --- /dev/null +++ b/mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "hashes": [ + "9acc8f0f15e5831249949239bac566467a951a3559d4953066f18631e7aef64f", + "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1" + ], + "models": [ + { + "games": 8740, + "iteration": 270, + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 6214, + "iteration": 192, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 3, + "losses": 7, + "cutoffs": 2, + "unfinished": 0, + "win_rate_decisive": 0.3, + "wilson_95_decisive": [ + 0.10779126740630102, + 0.6032218525388546 + ], + "candidate_root_simulations": { + "count": 808, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 819, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 344, + "mean": 0.013873586084607035, + "min": 0.010192811489105225, + "max": 0.1148918941617012 + }, + "champion_search_seconds": { + "count": 348, + "mean": 0.031710495883277776, + "min": 0.023946691304445267, + "max": 0.1708715409040451 + }, + "elapsed_seconds": 15.873456325381994, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380101, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 1, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 31 + }, + { + "game": 2, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 40 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 44 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 47 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 57 + }, + { + "game": 4, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 60 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 61 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 67 + }, + { + "game": 10, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 96 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 278 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.jsonl b/mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..7493e86a84f6f34d1c14d584b6350ce2a13fea24 --- /dev/null +++ b/mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 1, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 31} +{"game": 2, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 40} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 44} +{"game": 9, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 47} +{"game": 5, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 57} +{"game": 4, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 60} +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 61} +{"game": 7, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 67} +{"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 96} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 278} +{"game": 6, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 8, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/mlp256-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json b/mlp256-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..2ff6e5870406eaf4d3c88a439036d4f13bfb1e15 --- /dev/null +++ b/mlp256-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "hashes": [ + "9ca408753dfbe4c2103016e9694be0e5033ed517459bf8a7cef2cb56164fa9a7", + "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d" + ], + "models": [ + { + "games": 8122, + "iteration": 251, + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 6813, + "iteration": 209, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 0, + "losses": 12, + "cutoffs": 0, + "unfinished": 0, + "win_rate_decisive": 0.0, + "wilson_95_decisive": [ + 0.0, + 0.2424940066552408 + ], + "candidate_root_simulations": { + "count": 413, + "mean": 201.4721549636804, + "min": 139, + "max": 5467 + }, + "champion_root_simulations": { + "count": 414, + "mean": 524.7777777777778, + "min": 44, + "max": 10000 + }, + "candidate_search_seconds": { + "count": 413, + "mean": 0.05022500817084428, + "min": 0.05002162232995033, + "max": 0.05760979652404785 + }, + "champion_search_seconds": { + "count": 414, + "mean": 0.04998849508699012, + "min": 0.028176337480545044, + "max": 0.08139600604772568 + }, + "elapsed_seconds": 41.53895569965243, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380102, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 28 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 2, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 44 + }, + { + "game": 3, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 171 + }, + { + "game": 4, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 20 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 6, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 70 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 57 + }, + { + "game": 8, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 46 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 63 + }, + { + "game": 10, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 88 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 43 + } + ] +} diff --git a/mlp256-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl b/mlp256-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..56c8b51196e6518663806c5c29c9d4255c8634c6 --- /dev/null +++ b/mlp256-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 28} +{"game": 1, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 41} +{"game": 2, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 44} +{"game": 3, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 171} +{"game": 4, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 20} +{"game": 5, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 41} +{"game": 6, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 70} +{"game": 7, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 57} +{"game": 8, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 46} +{"game": 9, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 63} +{"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 88} +{"game": 11, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 43} diff --git a/mlp256-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json b/mlp256-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..489d7076307d09555540e1bec54ba53b01b378ab --- /dev/null +++ b/mlp256-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "hashes": [ + "9ca408753dfbe4c2103016e9694be0e5033ed517459bf8a7cef2cb56164fa9a7", + "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d" + ], + "models": [ + { + "games": 8122, + "iteration": 251, + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 6813, + "iteration": 209, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 0, + "losses": 9, + "cutoffs": 3, + "unfinished": 0, + "win_rate_decisive": 0.0, + "wilson_95_decisive": [ + 0.0, + 0.299145048419544 + ], + "candidate_root_simulations": { + "count": 726, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 722, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 339, + "mean": 0.014525834921538654, + "min": 0.00997798889875412, + "max": 0.11943966895341873 + }, + "champion_search_seconds": { + "count": 337, + "mean": 0.035479011782375926, + "min": 0.0011329799890518188, + "max": 0.2066914066672325 + }, + "elapsed_seconds": 16.942541033029556, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380102, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 4, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 20 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 21 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 8, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 28 + }, + { + "game": 0, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 28 + }, + { + "game": 2, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 30 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 6, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 44 + }, + { + "game": 3, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 71 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/mlp256-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.jsonl b/mlp256-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..f96cd660dc143335bf7f5b0401873dede0934f85 --- /dev/null +++ b/mlp256-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 4, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 20} +{"game": 7, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 21} +{"game": 1, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 25} +{"game": 8, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 28} +{"game": 0, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 28} +{"game": 2, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 30} +{"game": 11, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 41} +{"game": 6, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 44} +{"game": 3, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 71} +{"game": 5, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 10, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 9, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json b/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..2ac8e98fc3903bca9db8f3370932d8a883933306 --- /dev/null +++ b/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt", + "hashes": [ + "9ca408753dfbe4c2103016e9694be0e5033ed517459bf8a7cef2cb56164fa9a7", + "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c" + ], + "models": [ + { + "games": 8122, + "iteration": 251, + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 476, + "iteration": 14, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 5, + "losses": 0, + "cutoffs": 7, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.5655175352168254, + 1.0 + ], + "candidate_root_simulations": { + "count": 1456, + "mean": 245.4114010989011, + "min": 32, + "max": 10000 + }, + "champion_root_simulations": { + "count": 1439, + "mean": 16.82209867963864, + "min": 6, + "max": 22 + }, + "candidate_search_seconds": { + "count": 1456, + "mean": 0.05023818979408223, + "min": 0.029932472854852676, + "max": 0.09951138123869896 + }, + "champion_search_seconds": { + "count": 1439, + "mean": 0.051358162349735885, + "min": 0.05001908168196678, + "max": 0.07216424122452736 + }, + "elapsed_seconds": 147.3313555046916, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380102, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 44 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 3, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 100 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 54 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 38 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl b/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..dafcc8a422c2b96096573905241ad071e1029eba --- /dev/null +++ b/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 44} +{"game": 2, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 3, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 4, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 100} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 54} +{"game": 8, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 38} +{"game": 10, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 11, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json b/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..45e6fc96ac370f80aee9ed905f286a3a7617dab3 --- /dev/null +++ b/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt", + "hashes": [ + "9ca408753dfbe4c2103016e9694be0e5033ed517459bf8a7cef2cb56164fa9a7", + "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c" + ], + "models": [ + { + "games": 8122, + "iteration": 251, + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 476, + "iteration": 14, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 3, + "losses": 1, + "cutoffs": 8, + "unfinished": 0, + "win_rate_decisive": 0.75, + "wilson_95_decisive": [ + 0.300641842582402, + 0.9544127391902995 + ], + "candidate_root_simulations": { + "count": 1607, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 1598, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 371, + "mean": 0.014295252356326804, + "min": 0.010564547032117844, + "max": 0.14097487926483154 + }, + "champion_search_seconds": { + "count": 370, + "mean": 0.09938162548115125, + "min": 0.06747699156403542, + "max": 0.20460616424679756 + }, + "elapsed_seconds": 42.1833250336349, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380102, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 34 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 42 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 43 + }, + { + "game": 10, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 104 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 3, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.jsonl b/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..8c5f1ade0b0bf47110cf49e1b32dea812d736f62 --- /dev/null +++ b/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 34} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 42} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 43} +{"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 104} +{"game": 1, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 2, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 0, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 7, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 4, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 5, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 8, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 3, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json b/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..1637309b5ae7566cd86e777e70ddd665c1e19edb --- /dev/null +++ b/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt", + "hashes": [ + "9ca408753dfbe4c2103016e9694be0e5033ed517459bf8a7cef2cb56164fa9a7", + "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486" + ], + "models": [ + { + "games": 8122, + "iteration": 251, + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 717, + "iteration": 22, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + } + ], + "wins": 9, + "losses": 0, + "cutoffs": 3, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.7008549515804562, + 1.0 + ], + "candidate_root_simulations": { + "count": 871, + "mean": 325.7910447761194, + "min": 119, + "max": 10000 + }, + "champion_root_simulations": { + "count": 848, + "mean": 20.55424528301887, + "min": 6, + "max": 28 + }, + "candidate_search_seconds": { + "count": 871, + "mean": 0.050161529062946686, + "min": 0.029994037002325058, + "max": 0.09366054832935333 + }, + "champion_search_seconds": { + "count": 848, + "mean": 0.051409703450945185, + "min": 0.050026945769786835, + "max": 0.06496056169271469 + }, + "elapsed_seconds": 87.46665747836232, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380102, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 47 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 33 + }, + { + "game": 3, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 124 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 74 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 31 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 59 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 51 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 102 + } + ] +} diff --git a/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl b/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..717ee6d6afdb1b4272aa9e5f7799bea4e2c4f851 --- /dev/null +++ b/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 47} +{"game": 1, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 33} +{"game": 3, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 124} +{"game": 6, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 74} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 31} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 59} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 51} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 102} diff --git a/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json b/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..1f59b9c70712b2f7c4f19b5a1c0f938e1c759049 --- /dev/null +++ b/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt", + "hashes": [ + "9ca408753dfbe4c2103016e9694be0e5033ed517459bf8a7cef2cb56164fa9a7", + "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486" + ], + "models": [ + { + "games": 8122, + "iteration": 251, + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 717, + "iteration": 22, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + } + ], + "wins": 8, + "losses": 0, + "cutoffs": 4, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.6755924351161198, + 1.0 + ], + "candidate_root_simulations": { + "count": 950, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 937, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 368, + "mean": 0.015138583085702165, + "min": 0.01193181797862053, + "max": 0.1276196613907814 + }, + "champion_search_seconds": { + "count": 369, + "mean": 0.06623505816105904, + "min": 0.04075057804584503, + "max": 0.19303397834300995 + }, + "elapsed_seconds": 30.092273447662592, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380102, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 19 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 28 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 32 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 52 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 56 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 76 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 87 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.jsonl b/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..e389ac12df374ffbfa809ebba275062e04a3bb26 --- /dev/null +++ b/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 19} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 28} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 32} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 52} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 56} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 76} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 87} +{"game": 5, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 10, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 0, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 6, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/mlp256-seed410102-vs-squarecnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json b/mlp256-seed410102-vs-squarecnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..e606cb1b0626d9c41e870aa840cf0133d9a906cf --- /dev/null +++ b/mlp256-seed410102-vs-squarecnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410102/latest.pt", + "hashes": [ + "9ca408753dfbe4c2103016e9694be0e5033ed517459bf8a7cef2cb56164fa9a7", + "d3e8eb4f1e75d18bbeb28bef1b46c6c3f84b1ab6f009e63cc950a88f647c8025" + ], + "models": [ + { + "games": 8122, + "iteration": 251, + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 8449, + "iteration": 260, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 1, + "losses": 11, + "cutoffs": 0, + "unfinished": 0, + "win_rate_decisive": 0.08333333333333333, + "wilson_95_decisive": [ + 0.014865094404917123, + 0.3538799111411169 + ], + "candidate_root_simulations": { + "count": 336, + "mean": 264.05357142857144, + "min": 117, + "max": 10000 + }, + "champion_root_simulations": { + "count": 347, + "mean": 555.4438040345822, + "min": 53, + "max": 10000 + }, + "candidate_search_seconds": { + "count": 336, + "mean": 0.050196016961265175, + "min": 0.03129367530345917, + "max": 0.06348027661442757 + }, + "champion_search_seconds": { + "count": 347, + "mean": 0.04999155934459881, + "min": 0.03198779746890068, + "max": 0.06331682950258255 + }, + "elapsed_seconds": 34.30227893218398, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410102/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380102, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410102-vs-squarecnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 28 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 73 + }, + { + "game": 2, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 30 + }, + { + "game": 3, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 81 + }, + { + "game": 4, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 40 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 31 + }, + { + "game": 6, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 44 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 33 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 55 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 39 + }, + { + "game": 10, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 110 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 41 + } + ] +} diff --git a/mlp256-seed410102-vs-squarecnn32-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl b/mlp256-seed410102-vs-squarecnn32-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..ce781fc467922b6b6e6a7f0a370805a9cb848ec9 --- /dev/null +++ b/mlp256-seed410102-vs-squarecnn32-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 28} +{"game": 1, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 73} +{"game": 2, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 30} +{"game": 3, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 81} +{"game": 4, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 40} +{"game": 5, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 31} +{"game": 6, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 44} +{"game": 7, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 33} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 55} +{"game": 9, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 39} +{"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 110} +{"game": 11, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 41} diff --git a/mlp256-seed410102-vs-squarecnn32-seed410102-equal_simulations-seed1380102-attempt1.json b/mlp256-seed410102-vs-squarecnn32-seed410102-equal_simulations-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..e8785c3c761c2c4a3cc5b888a2d2920eaea942e6 --- /dev/null +++ b/mlp256-seed410102-vs-squarecnn32-seed410102-equal_simulations-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410102/latest.pt", + "hashes": [ + "9ca408753dfbe4c2103016e9694be0e5033ed517459bf8a7cef2cb56164fa9a7", + "d3e8eb4f1e75d18bbeb28bef1b46c6c3f84b1ab6f009e63cc950a88f647c8025" + ], + "models": [ + { + "games": 8122, + "iteration": 251, + "config": { + "kind": "mlp", + "width": 256, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 8449, + "iteration": 260, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 1, + "losses": 8, + "cutoffs": 3, + "unfinished": 0, + "win_rate_decisive": 0.1111111111111111, + "wilson_95_decisive": [ + 0.019890887638544935, + 0.4349997055766559 + ], + "candidate_root_simulations": { + "count": 821, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 821, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 374, + "mean": 0.013619239516755477, + "min": 0.011019594967365265, + "max": 0.1229017823934555 + }, + "champion_search_seconds": { + "count": 375, + "mean": 0.03153006933132808, + "min": 0.001027137041091919, + "max": 0.2048126608133316 + }, + "elapsed_seconds": 16.987353712320328, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/mlp256-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410102/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380102, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/mlp256-seed410102-vs-squarecnn32-seed410102-equal_simulations-seed1380102-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 5, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 29 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 33 + }, + { + "game": 2, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 40 + }, + { + "game": 10, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 40 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 47 + }, + { + "game": 4, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 46 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 59 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 67 + }, + { + "game": 3, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 89 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/mlp256-seed410102-vs-squarecnn32-seed410102-equal_simulations-seed1380102-attempt1.jsonl b/mlp256-seed410102-vs-squarecnn32-seed410102-equal_simulations-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..f0b77498fa5c7c6cd9659570590b6919d742d460 --- /dev/null +++ b/mlp256-seed410102-vs-squarecnn32-seed410102-equal_simulations-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 5, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 29} +{"game": 7, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 33} +{"game": 2, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 40} +{"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 40} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 47} +{"game": 4, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 46} +{"game": 1, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 59} +{"game": 9, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 67} +{"game": 3, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 89} +{"game": 0, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 11, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 8, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json b/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..5f20b90c324095cb02d0e052c41e9a45e2298827 --- /dev/null +++ b/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt", + "hashes": [ + "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4", + "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99" + ], + "models": [ + { + "games": 524, + "iteration": 15, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + }, + { + "games": 582, + "iteration": 18, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 3, + "losses": 7, + "cutoffs": 2, + "unfinished": 0, + "win_rate_decisive": 0.3, + "wilson_95_decisive": [ + 0.10779126740630102, + 0.6032218525388546 + ], + "candidate_root_simulations": { + "count": 768, + "mean": 73.85807291666667, + "min": 14, + "max": 10000 + }, + "champion_root_simulations": { + "count": 779, + "mean": 81.48395378690628, + "min": 6, + "max": 10000 + }, + "candidate_search_seconds": { + "count": 768, + "mean": 0.05112794004768754, + "min": 0.03248418867588043, + "max": 0.060117293149232864 + }, + "champion_search_seconds": { + "count": 779, + "mean": 0.05126317967308042, + "min": 0.03482479602098465, + "max": 0.06131729856133461 + }, + "elapsed_seconds": 79.2865728251636, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380101, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 113 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 3, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 4, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 74 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 51 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 119 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 111 + }, + { + "game": 10, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 68 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 63 + } + ] +} diff --git a/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl b/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..f36435bd4c60441552ff50c725a82479b250a04a --- /dev/null +++ b/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 113} +{"game": 1, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 49} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41} +{"game": 3, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 4, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 74} +{"game": 5, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 51} +{"game": 6, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 7, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 119} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 9, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 111} +{"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 68} +{"game": 11, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 63} diff --git a/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json b/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..4be58dce6b4c536bbd40981cc18340948193de01 --- /dev/null +++ b/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json @@ -0,0 +1,136 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt", + "hashes": [ + "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4", + "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99" + ], + "models": [ + { + "games": 524, + "iteration": 15, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + }, + { + "games": 582, + "iteration": 18, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 5, + "losses": 2, + "cutoffs": 0, + "unfinished": 5, + "win_rate_decisive": 0.7142857142857143, + "wilson_95_decisive": [ + 0.3589344518326194, + 0.9177810759959433 + ], + "candidate_root_simulations": { + "count": 825, + "mean": 31.92969696969697, + "min": 3, + "max": 32 + }, + "champion_root_simulations": { + "count": 817, + "mean": 31.88249694002448, + "min": 0, + "max": 32 + }, + "candidate_search_seconds": { + "count": 229, + "mean": 0.08270542749801578, + "min": 0.007983498275279999, + "max": 0.32279587537050247 + }, + "champion_search_seconds": { + "count": 220, + "mean": 0.11818442859432914, + "min": 0.002604551613330841, + "max": 0.24893256649374962 + }, + "elapsed_seconds": 45.01017473638058, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380101, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 35 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 37 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 44 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 2, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 56 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 60 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 159 + } + ] +} diff --git a/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.jsonl b/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..7b9119d2452bf47cbc52093482645b7a852330c0 --- /dev/null +++ b/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.jsonl @@ -0,0 +1,7 @@ +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 35} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 37} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 44} +{"game": 9, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 49} +{"game": 2, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 56} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 60} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 159} diff --git a/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json b/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json new file mode 100644 index 0000000000000000000000000000000000000000..5556165fdd5591dd584b84b898e67ef2b364939a --- /dev/null +++ b/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json @@ -0,0 +1,176 @@ +{ + "candidate": "checkpoints/overnight/20260909T235853Z/querytransformer256x2-seed410101-454d54089b4a.pt", + "champion": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt", + "hashes": [ + "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4", + "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99" + ], + "models": [ + { + "games": 524, + "iteration": 15, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + }, + { + "games": 582, + "iteration": 18, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 5, + "losses": 2, + "cutoffs": 5, + "unfinished": 0, + "win_rate_decisive": 0.7142857142857143, + "wilson_95_decisive": [ + 0.3589344518326194, + 0.9177810759959433 + ], + "candidate_root_simulations": { + "count": 1169, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 1161, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 367, + "mean": 0.07162341103972791, + "min": 0.03635891526937485, + "max": 0.29671428352594376 + }, + "champion_search_seconds": { + "count": 344, + "mean": 0.10342062305832325, + "min": 0.06300070509314537, + "max": 0.2168196216225624 + }, + "elapsed_seconds": 61.950876381248236, + "config": { + "candidate": "checkpoints/overnight/20260909T235853Z/querytransformer256x2-seed410101-454d54089b4a.pt", + "champion": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380101, + "max_ply": 300, + "seconds": 180, + "output": "reports/overnight/20260909T235853Z/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json", + "device": "cuda", + "threads": 1, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 35 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 37 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 44 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 2, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 56 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 60 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 159 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.jsonl b/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..edc537efb143a2ba2660655f87129cf37cfd9dec --- /dev/null +++ b/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.jsonl @@ -0,0 +1,12 @@ +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 35} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 37} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 44} +{"game": 9, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 49} +{"game": 2, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 56} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 60} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 159} +{"game": 8, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 0, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 7, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 11, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 1, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json b/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..928924574cdb5b72a4b650518995469f4718d7b9 --- /dev/null +++ b/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt", + "hashes": [ + "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486", + "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c" + ], + "models": [ + { + "games": 717, + "iteration": 22, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + }, + { + "games": 476, + "iteration": 14, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 6, + "losses": 0, + "cutoffs": 6, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.6096657120978348, + 1.0 + ], + "candidate_root_simulations": { + "count": 1350, + "mean": 57.968148148148146, + "min": 12, + "max": 10000 + }, + "champion_root_simulations": { + "count": 1330, + "mean": 16.8203007518797, + "min": 9, + "max": 22 + }, + "candidate_search_seconds": { + "count": 1350, + "mean": 0.05120152745257925, + "min": 0.035334262996912, + "max": 0.06653415039181709 + }, + "champion_search_seconds": { + "count": 1330, + "mean": 0.05127785819813721, + "min": 0.050020262598991394, + "max": 0.06667482107877731 + }, + "elapsed_seconds": 137.43875171989202, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380102, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 187 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 66 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 83 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 40 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 29 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl b/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..6f28d8e1027e430d9cba116c23455bf376e8af4d --- /dev/null +++ b/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 1, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 187} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 66} +{"game": 4, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 5, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 83} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 40} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 29} +{"game": 9, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 10, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 11, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json b/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..6976fbf6895fc24c54e3fd65ec49a89e1d7ad3bf --- /dev/null +++ b/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json @@ -0,0 +1,152 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt", + "hashes": [ + "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486", + "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c" + ], + "models": [ + { + "games": 717, + "iteration": 22, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + }, + { + "games": 476, + "iteration": 14, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 9, + "losses": 0, + "cutoffs": 0, + "unfinished": 3, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.7008549515804562, + 1.0 + ], + "candidate_root_simulations": { + "count": 684, + "mean": 31.941520467836256, + "min": 12, + "max": 32 + }, + "champion_root_simulations": { + "count": 676, + "mean": 31.952662721893493, + "min": 0, + "max": 32 + }, + "candidate_search_seconds": { + "count": 316, + "mean": 0.05658120499287225, + "min": 0.024914517998695374, + "max": 0.29415473341941833 + }, + "champion_search_seconds": { + "count": 316, + "mean": 0.08566429491944705, + "min": 0.001996941864490509, + "max": 0.17104847729206085 + }, + "elapsed_seconds": 45.008179280906916, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380102, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 36 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 38 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 42 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 43 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 57 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 68 + } + ] +} diff --git a/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.jsonl b/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..98f49f12d1331a45bedcbdc9018f4a0b15e06c11 --- /dev/null +++ b/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.jsonl @@ -0,0 +1,9 @@ +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 36} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 38} +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 42} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 43} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 57} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 68} diff --git a/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-retry-complete.json b/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-retry-complete.json new file mode 100644 index 0000000000000000000000000000000000000000..8082a360eb33b8e9c3f286b7e6a7478f86c416cd --- /dev/null +++ b/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-retry-complete.json @@ -0,0 +1,176 @@ +{ + "candidate": "checkpoints/overnight/20260909T235853Z/querytransformer256x2-seed410102-a0a1d4c39231.pt", + "champion": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410102-02b9f1909248.pt", + "hashes": [ + "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486", + "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c" + ], + "models": [ + { + "games": 717, + "iteration": 22, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + }, + { + "games": 476, + "iteration": 14, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 9, + "losses": 0, + "cutoffs": 3, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.7008549515804562, + 1.0 + ], + "candidate_root_simulations": { + "count": 762, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 754, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 368, + "mean": 0.05716586215219096, + "min": 0.03633255511522293, + "max": 0.37113621830940247 + }, + "champion_search_seconds": { + "count": 369, + "mean": 0.08689368917003557, + "min": 0.0574614480137825, + "max": 0.1723187081515789 + }, + "elapsed_seconds": 53.1677360907197, + "config": { + "candidate": "checkpoints/overnight/20260909T235853Z/querytransformer256x2-seed410102-a0a1d4c39231.pt", + "champion": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410102-02b9f1909248.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380102, + "max_ply": 300, + "seconds": 180, + "output": "reports/overnight/20260909T235853Z/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-retry-complete.json", + "device": "cuda", + "threads": 1, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 36 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 38 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 42 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 43 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 57 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 68 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-retry-complete.jsonl b/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-retry-complete.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..ea44d73e5c1e4942b629c4d86c81bb1f5e68bdcf --- /dev/null +++ b/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-retry-complete.jsonl @@ -0,0 +1,12 @@ +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 36} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 38} +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 41} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 42} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 43} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 57} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 68} +{"game": 9, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 1, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 2, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/results.json b/results.json new file mode 100644 index 0000000000000000000000000000000000000000..8c0eeb78ac8b459556c686b419d74888a0195f57 --- /dev/null +++ b/results.json @@ -0,0 +1,1319 @@ +{ + "backend_revision": { + "gipf_engine.cpython-312-x86_64-linux-gnu.so": "c0c66af56038ea3277ff93d3511adaedf38d5927b832b675d8ae429360e6979b", + "training/model.py": "1130a6f7e23524c04790adc702b8ecc7a9d92008bb354a398f377735b379dd71", + "training/native_search.py": "88beacc193f1fcfae082e46d93b3e1c8ca5edf76bf99c2c6298bd3d6cbe359f0", + "training/query_transformer.py": "9ff230547d8a0f9fd393fe05414a496724b3c514f76839a8f4c956e941f70385", + "training/search.py": "54587d2fde6bc62a671d9ff22ca2681f6fff521bbc5ed439c9cdd2b9c1786664", + "training/train.py": "8cbb2ed0b29d35c5101b6d9882cc2e83dc919277a32816b322a059abf97bd648" + }, + "continuation": { + "completed": true, + "league_deadline": 1789017000.0, + "run": "runs/overnight/20260909T235853Z/hexcnn32-seed410102", + "seconds": 11544 + }, + "deadline": 1789020000.0, + "duels": [ + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 975, + "max": 10000, + "mean": 268.34871794871793, + "min": 119 + }, + "champion_root_simulations": { + "count": 981, + "max": 10000, + "mean": 199.56574923547402, + "min": 42 + }, + "cutoffs": 0, + "losses": 9, + "unfinished": 0, + "wins": 3 + }, + "right": "squarecnn32-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410102-vs-squarecnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 336, + "max": 10000, + "mean": 264.05357142857144, + "min": 117 + }, + "champion_root_simulations": { + "count": 347, + "max": 10000, + "mean": 555.4438040345822, + "min": 53 + }, + "cutoffs": 0, + "losses": 11, + "unfinished": 0, + "wins": 1 + }, + "right": "squarecnn32-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 801, + "max": 10000, + "mean": 228.27715355805245, + "min": 100 + }, + "champion_root_simulations": { + "count": 801, + "max": 10000, + "mean": 201.11610486891385, + "min": 41 + }, + "cutoffs": 2, + "losses": 8, + "unfinished": 0, + "wins": 2 + }, + "right": "hexcnn32-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 413, + "max": 5467, + "mean": 201.4721549636804, + "min": 139 + }, + "champion_root_simulations": { + "count": 414, + "max": 10000, + "mean": 524.7777777777778, + "min": 44 + }, + "cutoffs": 0, + "losses": 12, + "unfinished": 0, + "wins": 0 + }, + "right": "hexcnn32-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 812, + "max": 10000, + "mean": 358.23522167487687, + "min": 123 + }, + "champion_root_simulations": { + "count": 790, + "max": 28, + "mean": 21.61139240506329, + "min": 10 + }, + "cutoffs": 2, + "losses": 0, + "unfinished": 0, + "wins": 10 + }, + "right": "querytransformer256x2-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 871, + "max": 10000, + "mean": 325.7910447761194, + "min": 119 + }, + "champion_root_simulations": { + "count": 848, + "max": 28, + "mean": 20.55424528301887, + "min": 6 + }, + "cutoffs": 3, + "losses": 0, + "unfinished": 0, + "wins": 9 + }, + "right": "querytransformer256x2-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 449, + "max": 10000, + "mean": 538.0890868596882, + "min": 89 + }, + "champion_root_simulations": { + "count": 426, + "max": 21, + "mean": 15.556338028169014, + "min": 7 + }, + "cutoffs": 1, + "losses": 0, + "unfinished": 0, + "wins": 11 + }, + "right": "querytransformer164x5-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 1456, + "max": 10000, + "mean": 245.4114010989011, + "min": 32 + }, + "champion_root_simulations": { + "count": 1439, + "max": 22, + "mean": 16.82209867963864, + "min": 6 + }, + "cutoffs": 7, + "losses": 0, + "unfinished": 0, + "wins": 5 + }, + "right": "querytransformer164x5-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 550, + "max": 10000, + "mean": 202.44181818181818, + "min": 54 + }, + "champion_root_simulations": { + "count": 529, + "max": 10000, + "mean": 244.84120982986767, + "min": 46 + }, + "cutoffs": 0, + "losses": 5, + "unfinished": 0, + "wins": 7 + }, + "right": "hexcnn32-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 393, + "max": 10000, + "mean": 304.5394402035623, + "min": 53 + }, + "champion_root_simulations": { + "count": 393, + "max": 10000, + "mean": 319.1501272264631, + "min": 61 + }, + "cutoffs": 0, + "losses": 6, + "unfinished": 0, + "wins": 6 + }, + "right": "hexcnn32-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 493, + "max": 10000, + "mean": 379.5496957403651, + "min": 63 + }, + "champion_root_simulations": { + "count": 464, + "max": 38, + "mean": 22.601293103448278, + "min": 12 + }, + "cutoffs": 0, + "losses": 0, + "unfinished": 0, + "wins": 12 + }, + "right": "querytransformer256x2-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 142, + "max": 10000, + "mean": 1453.8661971830986, + "min": 64 + }, + "champion_root_simulations": { + "count": 127, + "max": 641, + "mean": 26.62992125984252, + "min": 11 + }, + "cutoffs": 0, + "losses": 0, + "unfinished": 0, + "wins": 12 + }, + "right": "querytransformer256x2-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 383, + "max": 10000, + "mean": 459.7389033942559, + "min": 53 + }, + "champion_root_simulations": { + "count": 360, + "max": 21, + "mean": 16.216666666666665, + "min": 8 + }, + "cutoffs": 0, + "losses": 0, + "unfinished": 0, + "wins": 12 + }, + "right": "querytransformer164x5-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 334, + "max": 10000, + "mean": 640.0688622754491, + "min": 60 + }, + "champion_root_simulations": { + "count": 321, + "max": 21, + "mean": 17.14018691588785, + "min": 11 + }, + "cutoffs": 1, + "losses": 0, + "unfinished": 0, + "wins": 11 + }, + "right": "querytransformer164x5-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "hexcnn32-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 213, + "max": 10000, + "mean": 1017.1032863849765, + "min": 61 + }, + "champion_root_simulations": { + "count": 196, + "max": 27, + "mean": 22.714285714285715, + "min": 14 + }, + "cutoffs": 0, + "losses": 0, + "unfinished": 0, + "wins": 12 + }, + "right": "querytransformer256x2-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "hexcnn32-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 117, + "max": 10000, + "mean": 1548.6581196581196, + "min": 64 + }, + "champion_root_simulations": { + "count": 110, + "max": 25, + "mean": 21.654545454545456, + "min": 15 + }, + "cutoffs": 0, + "losses": 0, + "unfinished": 0, + "wins": 12 + }, + "right": "querytransformer256x2-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "hexcnn32-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 925, + "max": 10000, + "mean": 197.71243243243242, + "min": 49 + }, + "champion_root_simulations": { + "count": 911, + "max": 21, + "mean": 16.644346871569702, + "min": 6 + }, + "cutoffs": 4, + "losses": 0, + "unfinished": 0, + "wins": 8 + }, + "right": "querytransformer164x5-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "hexcnn32-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 368, + "max": 10000, + "mean": 533.054347826087, + "min": 58 + }, + "champion_root_simulations": { + "count": 354, + "max": 21, + "mean": 16.86723163841808, + "min": 8 + }, + "cutoffs": 1, + "losses": 0, + "unfinished": 0, + "wins": 11 + }, + "right": "querytransformer164x5-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "querytransformer256x2-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 768, + "max": 10000, + "mean": 73.85807291666667, + "min": 14 + }, + "champion_root_simulations": { + "count": 779, + "max": 10000, + "mean": 81.48395378690628, + "min": 6 + }, + "cutoffs": 2, + "losses": 7, + "unfinished": 0, + "wins": 3 + }, + "right": "querytransformer164x5-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "querytransformer256x2-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 1350, + "max": 10000, + "mean": 57.968148148148146, + "min": 12 + }, + "champion_root_simulations": { + "count": 1330, + "max": 22, + "mean": 16.8203007518797, + "min": 9 + }, + "cutoffs": 6, + "losses": 0, + "unfinished": 0, + "wins": 6 + }, + "right": "querytransformer164x5-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 808, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 819, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 2, + "losses": 7, + "unfinished": 0, + "wins": 3 + }, + "right": "squarecnn32-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410102-vs-squarecnn32-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 821, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 821, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 3, + "losses": 8, + "unfinished": 0, + "wins": 1 + }, + "right": "squarecnn32-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 731, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 739, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 2, + "losses": 9, + "unfinished": 0, + "wins": 1 + }, + "right": "hexcnn32-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 726, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 722, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 3, + "losses": 9, + "unfinished": 0, + "wins": 0 + }, + "right": "hexcnn32-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 1245, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 1217, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 5, + "losses": 2, + "unfinished": 0, + "wins": 5 + }, + "right": "querytransformer256x2-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 950, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 937, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 4, + "losses": 0, + "unfinished": 0, + "wins": 8 + }, + "right": "querytransformer256x2-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 473, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 454, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 0, + "losses": 1, + "unfinished": 0, + "wins": 11 + }, + "right": "querytransformer164x5-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 1607, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 1598, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 8, + "losses": 1, + "unfinished": 0, + "wins": 3 + }, + "right": "querytransformer164x5-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 517, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 494, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 1, + "losses": 1, + "unfinished": 0, + "wins": 10 + }, + "right": "hexcnn32-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 585, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 677, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 2, + "losses": 6, + "unfinished": 0, + "wins": 4 + }, + "right": "hexcnn32-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 649, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 617, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 1, + "losses": 0, + "unfinished": 0, + "wins": 11 + }, + "right": "querytransformer256x2-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 546, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 511, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 2, + "losses": 0, + "unfinished": 0, + "wins": 10 + }, + "right": "querytransformer256x2-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 599, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 570, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 1, + "losses": 1, + "unfinished": 0, + "wins": 10 + }, + "right": "querytransformer164x5-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 344, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 328, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 1, + "losses": 0, + "unfinished": 0, + "wins": 11 + }, + "right": "querytransformer164x5-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "hexcnn32-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 359, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 352, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 1, + "losses": 2, + "unfinished": 0, + "wins": 9 + }, + "right": "querytransformer256x2-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "hexcnn32-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 339, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 324, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 1, + "losses": 0, + "unfinished": 0, + "wins": 11 + }, + "right": "querytransformer256x2-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": false, + "left": "hexcnn32-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 1184, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 1172, + "max": 32, + "mean": 31.907849829351537, + "min": 5 + }, + "cutoffs": 0, + "losses": 0, + "unfinished": 6, + "wins": 6 + }, + "right": "querytransformer164x5-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "hexcnn32-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 367, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 353, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 1, + "losses": 0, + "unfinished": 0, + "wins": 11 + }, + "right": "querytransformer164x5-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": false, + "left": "querytransformer256x2-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 825, + "max": 32, + "mean": 31.92969696969697, + "min": 3 + }, + "champion_root_simulations": { + "count": 817, + "max": 32, + "mean": 31.88249694002448, + "min": 0 + }, + "cutoffs": 0, + "losses": 2, + "unfinished": 5, + "wins": 5 + }, + "right": "querytransformer164x5-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": false, + "left": "querytransformer256x2-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 684, + "max": 32, + "mean": 31.941520467836256, + "min": 12 + }, + "champion_root_simulations": { + "count": 676, + "max": 32, + "mean": 31.952662721893493, + "min": 0 + }, + "cutoffs": 0, + "losses": 0, + "unfinished": 3, + "wins": 9 + }, + "right": "querytransformer164x5-seed410102", + "seed": 1380102 + }, + { + "attempt": 3, + "complete": true, + "left": "squarecnn32-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-hexcnn32-seed410102-equal_cpu_time-seed989001-attempt3.json", + "result": { + "candidate_root_simulations": { + "count": 3382, + "max": 10000, + "mean": 234.5408042578356, + "min": 4 + }, + "champion_root_simulations": { + "count": 3326, + "max": 10000, + "mean": 271.61996392062537, + "min": 37 + }, + "cutoffs": 0, + "losses": 38, + "unfinished": 0, + "wins": 42 + }, + "right": "hexcnn32-seed410102", + "seed": 989001 + } + ], + "errors": [ + { + "label": "huggingface-upload", + "message": "huggingface-upload exited 1", + "type": "RuntimeError" + }, + { + "label": "github-push", + "message": "github-push exited 128", + "type": "RuntimeError" + } + ], + "promotion": { + "candidate": "checkpoints/overnight/20260909T235853Z/selected-final-cb19bc298969.pt", + "duel_report": "reports/overnight/20260909T235853Z/final-vs-served.json", + "greedy_report": null, + "passed": false + }, + "publication": { + "github_commit": true, + "github_push": false, + "github_release": false, + "hf_repo": "wuschelschulz/gipf-zero", + "source_hash": "aa533bb8922231a3e0dc53dd2ff947cab6899ef09a9245463c89ac99ea776ef0", + "uploaded": false + }, + "ranking": { + "alternative": "hexcnn32", + "direct_wilson_95": [ + 0.3507486535891948, + 0.72108666268789 + ], + "inconclusive": true, + "rows": [ + { + "mean_root_simulations": 524.4018087810757, + "name": "squarecnn32", + "rate": 0.8333333333333334, + "total": 96, + "unplayed": 0, + "wilson_95": [ + 0.746290098481051, + 0.8947261762804988 + ], + "wins": 80 + }, + { + "mean_root_simulations": 573.3016757505798, + "name": "hexcnn32", + "rate": 0.7708333333333334, + "total": 96, + "unplayed": 0, + "wilson_95": [ + 0.6773148946997618, + 0.8435108285439974 + ], + "wins": 74 + }, + { + "mean_root_simulations": 303.709794038576, + "name": "mlp256", + "rate": 0.4270833333333333, + "total": 96, + "unplayed": 0, + "wilson_95": [ + 0.33282083114008043, + 0.5269568587558306 + ], + "wins": 41 + }, + { + "mean_root_simulations": 33.448988035627366, + "name": "querytransformer256x2", + "rate": 0.09375, + "total": 96, + "unplayed": 0, + "wilson_95": [ + 0.050108108212146014, + 0.16865330692221547 + ], + "wins": 9 + }, + { + "mean_root_simulations": 24.69389041739199, + "name": "querytransformer164x5", + "rate": 0.07291666666666667, + "total": 96, + "unplayed": 0, + "wilson_95": [ + 0.03576758131036273, + 0.1429303166514019 + ], + "wins": 7 + } + ], + "seed_scores": { + "410101": [ + 40, + 48 + ], + "410102": [ + 40, + 48 + ] + }, + "tie_break": { + "alternative_seed_scores": { + "410101": [ + 33, + 48 + ], + "410102": [ + 41, + 48 + ] + }, + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-hexcnn32-seed410102-equal_cpu_time-seed989001-attempt3.json", + "selected": "hexcnn32", + "wilson_95": [ + 0.41695642649729947, + 0.6307526670259244 + ] + }, + "tie_rule": "higher achieved CPU root simulations within the fixed 50 ms cap only after an extra head-to-head remains inconclusive", + "winner": "hexcnn32", + "winner_seed": 410102 + }, + "status": "partial", + "tag": "20260909T235853Z", + "training": [ + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/mlp256-seed410101-9acc8f0f15e5.pt", + "name": "mlp256-seed410101", + "run": "runs/overnight/20260909T235853Z/mlp256-seed410101" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/mlp256-seed410102-9ca408753dfb.pt", + "name": "mlp256-seed410102", + "run": "runs/overnight/20260909T235853Z/mlp256-seed410102" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/squarecnn32-seed410101-e2a8946b1405.pt", + "name": "squarecnn32-seed410101", + "run": "runs/overnight/20260909T235853Z/squarecnn32-seed410101" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/squarecnn32-seed410102-d3e8eb4f1e75.pt", + "name": "squarecnn32-seed410102", + "run": "runs/overnight/20260909T235853Z/squarecnn32-seed410102" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/hexcnn32-seed410101-d6f1c7081cdb.pt", + "name": "hexcnn32-seed410101", + "run": "runs/overnight/20260909T235853Z/hexcnn32-seed410101" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/hexcnn32-seed410102-c7d5bd9e9ee0.pt", + "name": "hexcnn32-seed410102", + "run": "runs/overnight/20260909T235853Z/hexcnn32-seed410102" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/querytransformer256x2-seed410101-454d54089b4a.pt", + "name": "querytransformer256x2-seed410101", + "run": "runs/overnight/20260909T235853Z/querytransformer256x2-seed410101" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/querytransformer256x2-seed410102-a0a1d4c39231.pt", + "name": "querytransformer256x2-seed410102", + "run": "runs/overnight/20260909T235853Z/querytransformer256x2-seed410102" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt", + "name": "querytransformer164x5-seed410101", + "run": "runs/overnight/20260909T235853Z/querytransformer164x5-seed410101" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410102-02b9f1909248.pt", + "name": "querytransformer164x5-seed410102", + "run": "runs/overnight/20260909T235853Z/querytransformer164x5-seed410102" + } + ], + "training_stop": 1789018200.0, + "version": "overnight-v1" +} diff --git a/secondary-retries.json b/secondary-retries.json new file mode 100644 index 0000000000000000000000000000000000000000..bbd74f6b969599942cf70cd1ceb63a3c9a55ff61 --- /dev/null +++ b/secondary-retries.json @@ -0,0 +1,153 @@ +{ + "purpose": "Retries of only the three timed-out equal-simulations duels. The primary summary.json is intentionally unchanged.", + "all_complete": true, + "entries": [ + { + "original_report": "reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json", + "retry_report": "reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json", + "retry_log": "reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.log", + "left": "hexcnn32-seed410101", + "right": "querytransformer164x5-seed410101", + "seed": 1380101, + "frozen": { + "candidate": "checkpoints/overnight/20260909T235853Z/hexcnn32-seed410101-d6f1c7081cdb.pt", + "champion": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt" + }, + "hashes": [ + "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5", + "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99" + ], + "frozen_hashes_match_original": true, + "command": "PYTHONPATH=/workspace/gipf timeout 200s /venv/main/bin/python -m training.duel --candidate checkpoints/overnight/20260909T235853Z/hexcnn32-seed410101-d6f1c7081cdb.pt --champion checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt --games 12 --batch 64 --simulations 32 --seed 1380101 --max-ply 300 --seconds 180 --output reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json --device cuda --threads 1 --budget-ms 0", + "protocol_verified": { + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380101, + "max_ply": 300, + "seconds": 180, + "device": "cuda", + "threads": 1, + "budget_ms": 0.0 + }, + "result": { + "wins": 6, + "losses": 0, + "cutoffs": 6, + "unfinished": 0, + "elapsed_seconds": 45.535510551184416, + "win_rate_decisive": 1.0, + "candidate_root_simulations": { + "count": 1259, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 1247, + "mean": 32.0, + "min": 32, + "max": 32 + } + } + }, + { + "original_report": "reports/overnight/20260909T235853Z/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json", + "retry_report": "reports/overnight/20260909T235853Z/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json", + "retry_log": "reports/overnight/20260909T235853Z/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.log", + "left": "querytransformer256x2-seed410101", + "right": "querytransformer164x5-seed410101", + "seed": 1380101, + "frozen": { + "candidate": "checkpoints/overnight/20260909T235853Z/querytransformer256x2-seed410101-454d54089b4a.pt", + "champion": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt" + }, + "hashes": [ + "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4", + "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99" + ], + "frozen_hashes_match_original": true, + "command": "PYTHONPATH=/workspace/gipf timeout 200s /venv/main/bin/python -m training.duel --candidate checkpoints/overnight/20260909T235853Z/querytransformer256x2-seed410101-454d54089b4a.pt --champion checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt --games 12 --batch 64 --simulations 32 --seed 1380101 --max-ply 300 --seconds 180 --output reports/overnight/20260909T235853Z/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-retry-complete.json --device cuda --threads 1 --budget-ms 0", + "protocol_verified": { + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380101, + "max_ply": 300, + "seconds": 180, + "device": "cuda", + "threads": 1, + "budget_ms": 0.0 + }, + "result": { + "wins": 5, + "losses": 2, + "cutoffs": 5, + "unfinished": 0, + "elapsed_seconds": 61.950876381248236, + "win_rate_decisive": 0.7142857142857143, + "candidate_root_simulations": { + "count": 1169, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 1161, + "mean": 32.0, + "min": 32, + "max": 32 + } + } + }, + { + "original_report": "reports/overnight/20260909T235853Z/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json", + "retry_report": "reports/overnight/20260909T235853Z/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-retry-complete.json", + "retry_log": "reports/overnight/20260909T235853Z/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-retry-complete.log", + "left": "querytransformer256x2-seed410102", + "right": "querytransformer164x5-seed410102", + "seed": 1380102, + "frozen": { + "candidate": "checkpoints/overnight/20260909T235853Z/querytransformer256x2-seed410102-a0a1d4c39231.pt", + "champion": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410102-02b9f1909248.pt" + }, + "hashes": [ + "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486", + "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c" + ], + "frozen_hashes_match_original": true, + "command": "PYTHONPATH=/workspace/gipf timeout 200s /venv/main/bin/python -m training.duel --candidate checkpoints/overnight/20260909T235853Z/querytransformer256x2-seed410102-a0a1d4c39231.pt --champion checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410102-02b9f1909248.pt --games 12 --batch 64 --simulations 32 --seed 1380102 --max-ply 300 --seconds 180 --output reports/overnight/20260909T235853Z/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-retry-complete.json --device cuda --threads 1 --budget-ms 0", + "protocol_verified": { + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380102, + "max_ply": 300, + "seconds": 180, + "device": "cuda", + "threads": 1, + "budget_ms": 0.0 + }, + "result": { + "wins": 9, + "losses": 0, + "cutoffs": 3, + "unfinished": 0, + "elapsed_seconds": 53.1677360907197, + "win_rate_decisive": 1.0, + "candidate_root_simulations": { + "count": 762, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 754, + "mean": 32.0, + "min": 32, + "max": 32 + } + } + } + ] +} diff --git a/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json b/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..b503c03803ad57562d2fc2f2718612f327eb0ea3 --- /dev/null +++ b/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt", + "hashes": [ + "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1", + "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5" + ], + "models": [ + { + "games": 6214, + "iteration": 192, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 7877, + "iteration": 242, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 7, + "losses": 5, + "cutoffs": 0, + "unfinished": 0, + "win_rate_decisive": 0.5833333333333334, + "wilson_95_decisive": [ + 0.3195113125495498, + 0.8067396863412434 + ], + "candidate_root_simulations": { + "count": 550, + "mean": 202.44181818181818, + "min": 54, + "max": 10000 + }, + "champion_root_simulations": { + "count": 529, + "mean": 244.84120982986767, + "min": 46, + "max": 10000 + }, + "candidate_search_seconds": { + "count": 550, + "mean": 0.050231892344626515, + "min": 0.031321536749601364, + "max": 0.07072949782013893 + }, + "champion_search_seconds": { + "count": 529, + "mean": 0.050269144870664764, + "min": 0.03204876184463501, + "max": 0.061883267015218735 + }, + "elapsed_seconds": 54.311922907829285, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380101, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 146 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 54 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 55 + }, + { + "game": 3, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 73 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 99 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 78 + }, + { + "game": 6, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 68 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 85 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 103 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 47 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 90 + } + ] +} diff --git a/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl b/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..cceb8b9d585363ef2fc7be55d947f83113e1b469 --- /dev/null +++ b/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 146} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 54} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 55} +{"game": 3, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 73} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 99} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 78} +{"game": 6, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 68} +{"game": 7, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 85} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 103} +{"game": 9, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 47} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 90} diff --git a/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json b/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..a0a7c61e6ef724394775946c30ed9d56b5ad2ed5 --- /dev/null +++ b/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt", + "hashes": [ + "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1", + "d6f1c7081cdb8f802f88322f65368dfb52cb0cd10e2fae155efae1973e0bc2e5" + ], + "models": [ + { + "games": 6214, + "iteration": 192, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 7877, + "iteration": 242, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 10, + "losses": 1, + "cutoffs": 1, + "unfinished": 0, + "win_rate_decisive": 0.9090909090909091, + "wilson_95_decisive": [ + 0.6226415635484045, + 0.9837678271141154 + ], + "candidate_root_simulations": { + "count": 517, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 494, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 231, + "mean": 0.032635227152150434, + "min": 0.0010794885456562042, + "max": 0.28818827867507935 + }, + "champion_search_seconds": { + "count": 231, + "mean": 0.033849176277458926, + "min": 0.02713080495595932, + "max": 0.05115046352148056 + }, + "elapsed_seconds": 15.402560856193304, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410101/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380101, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 24 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 31 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 33 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 44 + }, + { + "game": 10, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 44 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 72 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 73 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 76 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 80 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 82 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.jsonl b/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..dc9e0b2fdf0a1ecff0c7a6ca1794aa9b419f1614 --- /dev/null +++ b/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 24} +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 31} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 33} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 44} +{"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 44} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 72} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 73} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 76} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 80} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 82} +{"game": 6, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/squarecnn32-seed410101-vs-hexcnn32-seed410102-equal_cpu_time-seed989001-attempt3.json b/squarecnn32-seed410101-vs-hexcnn32-seed410102-equal_cpu_time-seed989001-attempt3.json new file mode 100644 index 0000000000000000000000000000000000000000..e93a9ad2fa05356652ec1da4f9c26186497a590d --- /dev/null +++ b/squarecnn32-seed410101-vs-hexcnn32-seed410102-equal_cpu_time-seed989001-attempt3.json @@ -0,0 +1,720 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "hashes": [ + "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1", + "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d" + ], + "models": [ + { + "games": 6214, + "iteration": 192, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 6813, + "iteration": 209, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 42, + "losses": 38, + "cutoffs": 0, + "unfinished": 0, + "win_rate_decisive": 0.525, + "wilson_95_decisive": [ + 0.41695642649729947, + 0.6307526670259244 + ], + "candidate_root_simulations": { + "count": 3382, + "mean": 234.5408042578356, + "min": 4, + "max": 10000 + }, + "champion_root_simulations": { + "count": 3326, + "mean": 271.61996392062537, + "min": 37, + "max": 10000 + }, + "candidate_search_seconds": { + "count": 3382, + "mean": 0.05015051359281006, + "min": 0.026926055550575256, + "max": 0.09654531255364418 + }, + "champion_search_seconds": { + "count": 3326, + "mean": 0.0502294227581693, + "min": 0.026342011988162994, + "max": 0.10072683170437813 + }, + "elapsed_seconds": 337.1946141421795, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "games": 80, + "batch": 1, + "simulations": 10000, + "seed": 989001, + "max_ply": 300, + "seconds": 600, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-hexcnn32-seed410102-equal_cpu_time-seed989001-attempt3.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 129 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 120 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 130 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 51 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 122 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 97 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 62 + }, + { + "game": 8, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 100 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 114 + }, + { + "game": 10, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 64 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 80 + }, + { + "game": 12, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 84 + }, + { + "game": 13, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 57 + }, + { + "game": 14, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 65 + }, + { + "game": 15, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 47 + }, + { + "game": 16, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 62 + }, + { + "game": 17, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 76 + }, + { + "game": 18, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 83 + }, + { + "game": 19, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 86 + }, + { + "game": 20, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 68 + }, + { + "game": 21, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 57 + }, + { + "game": 22, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 71 + }, + { + "game": 23, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 62 + }, + { + "game": 24, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 100 + }, + { + "game": 25, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 56 + }, + { + "game": 26, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 107 + }, + { + "game": 27, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 71 + }, + { + "game": 28, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 33 + }, + { + "game": 29, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 126 + }, + { + "game": 30, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 53 + }, + { + "game": 31, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 27 + }, + { + "game": 32, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 86 + }, + { + "game": 33, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 105 + }, + { + "game": 34, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 57 + }, + { + "game": 35, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 154 + }, + { + "game": 36, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 90 + }, + { + "game": 37, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 53 + }, + { + "game": 38, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 56 + }, + { + "game": 39, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 39 + }, + { + "game": 40, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 32 + }, + { + "game": 41, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 33 + }, + { + "game": 42, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 83 + }, + { + "game": 43, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 108 + }, + { + "game": 44, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 36 + }, + { + "game": 45, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 37 + }, + { + "game": 46, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 85 + }, + { + "game": 47, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 90 + }, + { + "game": 48, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 44 + }, + { + "game": 49, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 94 + }, + { + "game": 50, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 88 + }, + { + "game": 51, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 65 + }, + { + "game": 52, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 46 + }, + { + "game": 53, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 70 + }, + { + "game": 54, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 103 + }, + { + "game": 55, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 23 + }, + { + "game": 56, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 57, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 47 + }, + { + "game": 58, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 60 + }, + { + "game": 59, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 72 + }, + { + "game": 60, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 28 + }, + { + "game": 61, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 43 + }, + { + "game": 62, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 26 + }, + { + "game": 63, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 40 + }, + { + "game": 64, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 129 + }, + { + "game": 65, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 112 + }, + { + "game": 66, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 46 + }, + { + "game": 67, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 53 + }, + { + "game": 68, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 109 + }, + { + "game": 69, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 70, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 77 + }, + { + "game": 71, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 125 + }, + { + "game": 72, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 71 + }, + { + "game": 73, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 27 + }, + { + "game": 74, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 66 + }, + { + "game": 75, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 57 + }, + { + "game": 76, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 71 + }, + { + "game": 77, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 64 + }, + { + "game": 78, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 129 + }, + { + "game": 79, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 104 + } + ] +} diff --git a/squarecnn32-seed410101-vs-hexcnn32-seed410102-equal_cpu_time-seed989001-attempt3.jsonl b/squarecnn32-seed410101-vs-hexcnn32-seed410102-equal_cpu_time-seed989001-attempt3.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..dc847787d26c8c8cc817f25cbb8a1c23d95a6cc6 --- /dev/null +++ b/squarecnn32-seed410101-vs-hexcnn32-seed410102-equal_cpu_time-seed989001-attempt3.jsonl @@ -0,0 +1,80 @@ +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 129} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 120} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 49} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 130} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 51} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 122} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 97} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 62} +{"game": 8, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 100} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 114} +{"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 64} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 80} +{"game": 12, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 84} +{"game": 13, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 57} +{"game": 14, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 65} +{"game": 15, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 47} +{"game": 16, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 62} +{"game": 17, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 76} +{"game": 18, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 83} +{"game": 19, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 86} +{"game": 20, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 68} +{"game": 21, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 57} +{"game": 22, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 71} +{"game": 23, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 62} +{"game": 24, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 100} +{"game": 25, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 56} +{"game": 26, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 107} +{"game": 27, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 71} +{"game": 28, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 33} +{"game": 29, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 126} +{"game": 30, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 53} +{"game": 31, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 27} +{"game": 32, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 86} +{"game": 33, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 105} +{"game": 34, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 57} +{"game": 35, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 154} +{"game": 36, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 90} +{"game": 37, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 53} +{"game": 38, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 56} +{"game": 39, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 39} +{"game": 40, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 32} +{"game": 41, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 33} +{"game": 42, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 83} +{"game": 43, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 108} +{"game": 44, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 36} +{"game": 45, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 37} +{"game": 46, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 85} +{"game": 47, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 90} +{"game": 48, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 44} +{"game": 49, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 94} +{"game": 50, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 88} +{"game": 51, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 65} +{"game": 52, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 46} +{"game": 53, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 70} +{"game": 54, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 103} +{"game": 55, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 23} +{"game": 56, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 49} +{"game": 57, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 47} +{"game": 58, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 60} +{"game": 59, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 72} +{"game": 60, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 28} +{"game": 61, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 43} +{"game": 62, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 26} +{"game": 63, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 40} +{"game": 64, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 129} +{"game": 65, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 112} +{"game": 66, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 46} +{"game": 67, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 53} +{"game": 68, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 109} +{"game": 69, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 49} +{"game": 70, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 77} +{"game": 71, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 125} +{"game": 72, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 71} +{"game": 73, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 27} +{"game": 74, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 66} +{"game": 75, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 57} +{"game": 76, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 71} +{"game": 77, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 64} +{"game": 78, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 129} +{"game": 79, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 104} diff --git a/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json b/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..de4d073e903b09373d7c2480edb089199f317ff8 --- /dev/null +++ b/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt", + "hashes": [ + "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1", + "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99" + ], + "models": [ + { + "games": 6214, + "iteration": 192, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 582, + "iteration": 18, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 12, + "losses": 0, + "cutoffs": 0, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.7575059933447593, + 1.0 + ], + "candidate_root_simulations": { + "count": 383, + "mean": 459.7389033942559, + "min": 53, + "max": 10000 + }, + "champion_root_simulations": { + "count": 360, + "mean": 16.216666666666665, + "min": 8, + "max": 21 + }, + "candidate_search_seconds": { + "count": 383, + "mean": 0.0497943941435677, + "min": 0.02946627512574196, + "max": 0.0542784258723259 + }, + "champion_search_seconds": { + "count": 360, + "mean": 0.051412929159899555, + "min": 0.05004674568772316, + "max": 0.06146227940917015 + }, + "elapsed_seconds": 37.63307252526283, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380101, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 67 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 36 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 31 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 94 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 45 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 50 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 99 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 48 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 57 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 42 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 51 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 48 + } + ] +} diff --git a/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl b/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..d4112af69ff13cb61e40220b56df1dfce51dc803 --- /dev/null +++ b/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 67} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 36} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 31} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 94} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 45} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 50} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 99} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 48} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 57} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 42} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 51} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 48} diff --git a/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json b/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..1727893fcd1e7754b1c4dbdd650657fbcc9d9f81 --- /dev/null +++ b/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt", + "hashes": [ + "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1", + "8568880b8377b2fd9d2d0f10b1081838644f69e6e9be3df26bb5fc9c9188ed99" + ], + "models": [ + { + "games": 6214, + "iteration": 192, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 582, + "iteration": 18, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 10, + "losses": 1, + "cutoffs": 1, + "unfinished": 0, + "win_rate_decisive": 0.9090909090909091, + "wilson_95_decisive": [ + 0.6226415635484045, + 0.9837678271141154 + ], + "candidate_root_simulations": { + "count": 599, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 570, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 248, + "mean": 0.0331877606560386, + "min": 0.002014435827732086, + "max": 0.2971048541367054 + }, + "champion_search_seconds": { + "count": 239, + "mean": 0.09911586252278863, + "min": 0.06577104330062866, + "max": 0.19971605762839317 + }, + "elapsed_seconds": 31.971445746719837, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410101/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380101, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 28 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 30 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 33 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 63 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 64 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 73 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 77 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 78 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 88 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 121 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.jsonl b/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..91c5c8233454a3aa87a25b36efbf98d240ddee49 --- /dev/null +++ b/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 28} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 30} +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 33} +{"game": 1, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 41} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 63} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 64} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 73} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 77} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 78} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 88} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 121} +{"game": 2, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json b/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..a50e80f2bda647d07786caaf4ac12c89f822ab5f --- /dev/null +++ b/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt", + "hashes": [ + "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1", + "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4" + ], + "models": [ + { + "games": 6214, + "iteration": 192, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 524, + "iteration": 15, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + } + ], + "wins": 12, + "losses": 0, + "cutoffs": 0, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.7575059933447593, + 1.0 + ], + "candidate_root_simulations": { + "count": 493, + "mean": 379.5496957403651, + "min": 63, + "max": 10000 + }, + "champion_root_simulations": { + "count": 464, + "mean": 22.601293103448278, + "min": 12, + "max": 38 + }, + "candidate_search_seconds": { + "count": 493, + "mean": 0.05003967622774851, + "min": 0.030458752065896988, + "max": 0.06294817104935646 + }, + "champion_search_seconds": { + "count": 464, + "mean": 0.05108116690775957, + "min": 0.050021734088659286, + "max": 0.06095051020383835 + }, + "elapsed_seconds": 48.445274382829666, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380101, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 93 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 40 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 135 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 74 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 79 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 74 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 43 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 56 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 61 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 52 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 67 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 66 + } + ] +} diff --git a/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl b/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..622379a69d4aee175f95d0ef9e6a63be4e155f25 --- /dev/null +++ b/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 93} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 40} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 135} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 74} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 79} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 74} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 43} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 56} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 61} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 52} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 67} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 66} diff --git a/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json b/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..d545e6b2c0686b2593ecf320e0a0e81ec809a062 --- /dev/null +++ b/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt", + "hashes": [ + "e2a8946b1405174aaa0b0d08e8d88e10078f40e90aac72596631d1bd6757cbd1", + "454d54089b4a383e448282be0472c67d25a6be6f4e5cb510d6a37256a9111ea4" + ], + "models": [ + { + "games": 6214, + "iteration": 192, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 524, + "iteration": 15, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + } + ], + "wins": 11, + "losses": 0, + "cutoffs": 1, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.7411670330319684, + 1.0 + ], + "candidate_root_simulations": { + "count": 649, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 617, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 249, + "mean": 0.03357411241794686, + "min": 0.001592107117176056, + "max": 0.3104752488434315 + }, + "champion_search_seconds": { + "count": 240, + "mean": 0.05915195237224301, + "min": 0.0388629287481308, + "max": 0.1813908889889717 + }, + "elapsed_seconds": 22.62154208496213, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410101/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410101/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380101, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 32 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 44 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 63 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 62 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 64 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 76 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 79 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 79 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 84 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 91 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 107 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.jsonl b/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..d57ea02d5fbb29c340b01524850cd81df12ce962 --- /dev/null +++ b/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 32} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 44} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 63} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 62} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 64} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 76} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 79} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 79} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 84} +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 91} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 107} +{"game": 4, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json b/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..ba466d37e97068d5600af4011ffcb0fe6455227f --- /dev/null +++ b/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "hashes": [ + "d3e8eb4f1e75d18bbeb28bef1b46c6c3f84b1ab6f009e63cc950a88f647c8025", + "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d" + ], + "models": [ + { + "games": 8449, + "iteration": 260, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 6813, + "iteration": 209, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 6, + "losses": 6, + "cutoffs": 0, + "unfinished": 0, + "win_rate_decisive": 0.5, + "wilson_95_decisive": [ + 0.2537815976337061, + 0.7462184023662939 + ], + "candidate_root_simulations": { + "count": 393, + "mean": 304.5394402035623, + "min": 53, + "max": 10000 + }, + "champion_root_simulations": { + "count": 393, + "mean": 319.1501272264631, + "min": 61, + "max": 10000 + }, + "candidate_search_seconds": { + "count": 393, + "mean": 0.050203031199124025, + "min": 0.030896257609128952, + "max": 0.05780206248164177 + }, + "champion_search_seconds": { + "count": 393, + "mean": 0.050151107136074825, + "min": 0.03056873008608818, + "max": 0.058407895267009735 + }, + "elapsed_seconds": 39.519955571740866, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380102, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 99 + }, + { + "game": 1, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 2, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 76 + }, + { + "game": 3, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 39 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 31 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 61 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 59 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 68 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 59 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 32 + }, + { + "game": 10, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 58 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 71 + } + ] +} diff --git a/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl b/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..71c5e06cb7d802f173fb47c9620cd2640d5d25a5 --- /dev/null +++ b/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 99} +{"game": 1, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 41} +{"game": 2, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 76} +{"game": 3, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 39} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 31} +{"game": 5, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 61} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 59} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 68} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 59} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 32} +{"game": 10, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 58} +{"game": 11, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 71} diff --git a/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json b/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..6289caf80e104ec5ec86225550e441d2085555a7 --- /dev/null +++ b/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "hashes": [ + "d3e8eb4f1e75d18bbeb28bef1b46c6c3f84b1ab6f009e63cc950a88f647c8025", + "c7d5bd9e9ee03104059833d76fa22e3aad2919e2f321302454d0b1750d8ce62d" + ], + "models": [ + { + "games": 8449, + "iteration": 260, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 6813, + "iteration": 209, + "config": { + "kind": "hexresnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + } + ], + "wins": 4, + "losses": 6, + "cutoffs": 2, + "unfinished": 0, + "win_rate_decisive": 0.4, + "wilson_95_decisive": [ + 0.1681803297062362, + 0.6873262302663417 + ], + "candidate_root_simulations": { + "count": 585, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 677, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 318, + "mean": 0.030963851982411347, + "min": 0.0030286014080047607, + "max": 0.28987184166908264 + }, + "champion_search_seconds": { + "count": 372, + "mean": 0.032216635562719836, + "min": 0.025788579136133194, + "max": 0.04726593568921089 + }, + "elapsed_seconds": 21.885406505316496, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/hexcnn32-seed410102/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380102, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 1, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 28 + }, + { + "game": 7, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 35 + }, + { + "game": 4, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 34 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 35 + }, + { + "game": 8, + "candidate_color": 1, + "winner": -1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 36 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 41 + }, + { + "game": 9, + "candidate_color": -1, + "winner": 1, + "outcome": "loss", + "cutoff_reason": null, + "ply": 45 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 49 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.jsonl b/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..756b620f0763dad3a4857fe5f758d21c946fe705 --- /dev/null +++ b/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 1, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 25} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 28} +{"game": 7, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 35} +{"game": 4, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 34} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 35} +{"game": 8, "candidate_color": 1, "winner": -1, "outcome": "loss", "cutoff_reason": null, "ply": 36} +{"game": 11, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 41} +{"game": 9, "candidate_color": -1, "winner": 1, "outcome": "loss", "cutoff_reason": null, "ply": 45} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 49} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 49} +{"game": 0, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 5, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json b/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..e03d82a5d4167c92298ef985b411e7f581a8ee22 --- /dev/null +++ b/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt", + "hashes": [ + "d3e8eb4f1e75d18bbeb28bef1b46c6c3f84b1ab6f009e63cc950a88f647c8025", + "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c" + ], + "models": [ + { + "games": 8449, + "iteration": 260, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 476, + "iteration": 14, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 11, + "losses": 0, + "cutoffs": 1, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.7411670330319684, + 1.0 + ], + "candidate_root_simulations": { + "count": 334, + "mean": 640.0688622754491, + "min": 60, + "max": 10000 + }, + "champion_root_simulations": { + "count": 321, + "mean": 17.14018691588785, + "min": 11, + "max": 21 + }, + "candidate_search_seconds": { + "count": 334, + "mean": 0.049763106512363084, + "min": 0.02786746248602867, + "max": 0.05684010684490204 + }, + "champion_search_seconds": { + "count": 321, + "mean": 0.05116588806855344, + "min": 0.05002874508500099, + "max": 0.05709265545010567 + }, + "elapsed_seconds": 33.0934715308249, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380102, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 27 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 28 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 29 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 34 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 20 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 17 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 20 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 33 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 20 + } + ] +} diff --git a/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl b/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..a59789a155f1b6522563b4ae9ccc3e65c73c977b --- /dev/null +++ b/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 27} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 28} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 29} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 34} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 5, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 20} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 17} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 20} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 33} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 20} diff --git a/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json b/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..cb4b767b3bfe0683f5825df568ed529e821f200a --- /dev/null +++ b/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt", + "hashes": [ + "d3e8eb4f1e75d18bbeb28bef1b46c6c3f84b1ab6f009e63cc950a88f647c8025", + "02b9f1909248f78a8d814f1f611cb21e44e0b552d630b70d49df66cb04cc9a8c" + ], + "models": [ + { + "games": 8449, + "iteration": 260, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 476, + "iteration": 14, + "config": { + "kind": "transformer", + "width": 164, + "blocks": 5, + "head": "query" + } + } + ], + "wins": 11, + "losses": 0, + "cutoffs": 1, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.7411670330319684, + 1.0 + ], + "candidate_root_simulations": { + "count": 344, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 328, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 203, + "mean": 0.03357011074282853, + "min": 0.002152536064386368, + "max": 0.29505839943885803 + }, + "champion_search_seconds": { + "count": 201, + "mean": 0.08379500094263707, + "min": 0.06239038705825806, + "max": 0.19482364505529404 + }, + "elapsed_seconds": 23.69235148653388, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer164x5-seed410102/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380102, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 17 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 19 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 22 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 24 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 29 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 28 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 33 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 32 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 33 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 34 + }, + { + "game": 5, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.jsonl b/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..67a96a4d2dc39762a5eb5f0bd3372c8273567da6 --- /dev/null +++ b/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 17} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 19} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 22} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 24} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 29} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 28} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 33} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 32} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 33} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 34} +{"game": 5, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json b/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..91c6c73105604b55f5e476d76072dc602e7aeff6 --- /dev/null +++ b/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt", + "hashes": [ + "d3e8eb4f1e75d18bbeb28bef1b46c6c3f84b1ab6f009e63cc950a88f647c8025", + "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486" + ], + "models": [ + { + "games": 8449, + "iteration": 260, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 717, + "iteration": 22, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + } + ], + "wins": 12, + "losses": 0, + "cutoffs": 0, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.7575059933447593, + 1.0 + ], + "candidate_root_simulations": { + "count": 142, + "mean": 1453.8661971830986, + "min": 64, + "max": 10000 + }, + "champion_root_simulations": { + "count": 127, + "mean": 26.62992125984252, + "min": 11, + "max": 641 + }, + "candidate_search_seconds": { + "count": 142, + "mean": 0.048951520440234264, + "min": 0.02962544932961464, + "max": 0.05082029104232788 + }, + "champion_search_seconds": { + "count": 127, + "mean": 0.05104695779718752, + "min": 0.050039295107126236, + "max": 0.0547877661883831 + }, + "elapsed_seconds": 13.460505403578281, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt", + "games": 12, + "batch": 1, + "simulations": 10000, + "seed": 1380102, + "max_ply": 300, + "seconds": 180, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "device": "cpu", + "threads": 1, + "budget_ms": 50.0 + }, + "records": [ + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 19 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 20 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 30 + }, + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 19 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 16 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 17 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 20 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 26 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 19 + }, + { + "game": 11, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 30 + } + ] +} diff --git a/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl b/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..f75cae115afca27944269f87a6adba86ddad175f --- /dev/null +++ b/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 19} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 20} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 30} +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 19} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 16} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 17} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 20} +{"game": 8, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 26} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 19} +{"game": 11, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 30} diff --git a/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json b/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json new file mode 100644 index 0000000000000000000000000000000000000000..f19c12fd52a974679aa8ba0c61bd6f254b96fde0 --- /dev/null +++ b/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json @@ -0,0 +1,176 @@ +{ + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt", + "hashes": [ + "d3e8eb4f1e75d18bbeb28bef1b46c6c3f84b1ab6f009e63cc950a88f647c8025", + "a0a1d4c3923127d74b628540ae9fb3a117517ce6d8757629b73c55c4d9d3d486" + ], + "models": [ + { + "games": 8449, + "iteration": 260, + "config": { + "kind": "resnet", + "width": 32, + "blocks": 2, + "head": "flat" + } + }, + { + "games": 717, + "iteration": 22, + "config": { + "kind": "transformer", + "width": 256, + "blocks": 2, + "head": "query" + } + } + ], + "wins": 10, + "losses": 0, + "cutoffs": 2, + "unfinished": 0, + "win_rate_decisive": 1.0, + "wilson_95_decisive": [ + 0.7224672001371109, + 1.0 + ], + "candidate_root_simulations": { + "count": 546, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "champion_root_simulations": { + "count": 511, + "mean": 32.0, + "min": 32, + "max": 32 + }, + "candidate_search_seconds": { + "count": 302, + "mean": 0.03249530002030711, + "min": 0.02464822679758072, + "max": 0.30006716027855873 + }, + "champion_search_seconds": { + "count": 298, + "mean": 0.056531574427081434, + "min": 0.03716142848134041, + "max": 0.1819905824959278 + }, + "elapsed_seconds": 26.709463000297546, + "config": { + "candidate": "/workspace/gipf/runs/overnight/20260909T235853Z/squarecnn32-seed410102/latest.pt", + "champion": "/workspace/gipf/runs/overnight/20260909T235853Z/querytransformer256x2-seed410102/latest.pt", + "games": 12, + "batch": 64, + "simulations": 32, + "seed": 1380102, + "max_ply": 300, + "seconds": 45, + "output": "/workspace/gipf/reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json", + "device": "cuda", + "threads": 2, + "budget_ms": 0.0 + }, + "records": [ + { + "game": 4, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 17 + }, + { + "game": 5, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 24 + }, + { + "game": 9, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 24 + }, + { + "game": 1, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 24 + }, + { + "game": 3, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 22 + }, + { + "game": 6, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 23 + }, + { + "game": 10, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 25 + }, + { + "game": 2, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 29 + }, + { + "game": 7, + "candidate_color": -1, + "winner": -1, + "outcome": "win", + "cutoff_reason": null, + "ply": 32 + }, + { + "game": 0, + "candidate_color": 1, + "winner": 1, + "outcome": "win", + "cutoff_reason": null, + "ply": 81 + }, + { + "game": 11, + "candidate_color": -1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + }, + { + "game": 8, + "candidate_color": 1, + "winner": 0, + "outcome": "cutoff", + "cutoff_reason": "insertion", + "ply": 300 + } + ] +} diff --git a/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.jsonl b/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..1515e260cb7058c0db3d857c369df1711438bce2 --- /dev/null +++ b/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.jsonl @@ -0,0 +1,12 @@ +{"game": 4, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 17} +{"game": 5, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 24} +{"game": 9, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 24} +{"game": 1, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 24} +{"game": 3, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 22} +{"game": 6, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 23} +{"game": 10, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 25} +{"game": 2, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 29} +{"game": 7, "candidate_color": -1, "winner": -1, "outcome": "win", "cutoff_reason": null, "ply": 32} +{"game": 0, "candidate_color": 1, "winner": 1, "outcome": "win", "cutoff_reason": null, "ply": 81} +{"game": 11, "candidate_color": -1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} +{"game": 8, "candidate_color": 1, "winner": 0, "outcome": "cutoff", "cutoff_reason": "insertion", "ply": 300} diff --git a/summary.json b/summary.json new file mode 100644 index 0000000000000000000000000000000000000000..8c0eeb78ac8b459556c686b419d74888a0195f57 --- /dev/null +++ b/summary.json @@ -0,0 +1,1319 @@ +{ + "backend_revision": { + "gipf_engine.cpython-312-x86_64-linux-gnu.so": "c0c66af56038ea3277ff93d3511adaedf38d5927b832b675d8ae429360e6979b", + "training/model.py": "1130a6f7e23524c04790adc702b8ecc7a9d92008bb354a398f377735b379dd71", + "training/native_search.py": "88beacc193f1fcfae082e46d93b3e1c8ca5edf76bf99c2c6298bd3d6cbe359f0", + "training/query_transformer.py": "9ff230547d8a0f9fd393fe05414a496724b3c514f76839a8f4c956e941f70385", + "training/search.py": "54587d2fde6bc62a671d9ff22ca2681f6fff521bbc5ed439c9cdd2b9c1786664", + "training/train.py": "8cbb2ed0b29d35c5101b6d9882cc2e83dc919277a32816b322a059abf97bd648" + }, + "continuation": { + "completed": true, + "league_deadline": 1789017000.0, + "run": "runs/overnight/20260909T235853Z/hexcnn32-seed410102", + "seconds": 11544 + }, + "deadline": 1789020000.0, + "duels": [ + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410101-vs-squarecnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 975, + "max": 10000, + "mean": 268.34871794871793, + "min": 119 + }, + "champion_root_simulations": { + "count": 981, + "max": 10000, + "mean": 199.56574923547402, + "min": 42 + }, + "cutoffs": 0, + "losses": 9, + "unfinished": 0, + "wins": 3 + }, + "right": "squarecnn32-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410102-vs-squarecnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 336, + "max": 10000, + "mean": 264.05357142857144, + "min": 117 + }, + "champion_root_simulations": { + "count": 347, + "max": 10000, + "mean": 555.4438040345822, + "min": 53 + }, + "cutoffs": 0, + "losses": 11, + "unfinished": 0, + "wins": 1 + }, + "right": "squarecnn32-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 801, + "max": 10000, + "mean": 228.27715355805245, + "min": 100 + }, + "champion_root_simulations": { + "count": 801, + "max": 10000, + "mean": 201.11610486891385, + "min": 41 + }, + "cutoffs": 2, + "losses": 8, + "unfinished": 0, + "wins": 2 + }, + "right": "hexcnn32-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 413, + "max": 5467, + "mean": 201.4721549636804, + "min": 139 + }, + "champion_root_simulations": { + "count": 414, + "max": 10000, + "mean": 524.7777777777778, + "min": 44 + }, + "cutoffs": 0, + "losses": 12, + "unfinished": 0, + "wins": 0 + }, + "right": "hexcnn32-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 812, + "max": 10000, + "mean": 358.23522167487687, + "min": 123 + }, + "champion_root_simulations": { + "count": 790, + "max": 28, + "mean": 21.61139240506329, + "min": 10 + }, + "cutoffs": 2, + "losses": 0, + "unfinished": 0, + "wins": 10 + }, + "right": "querytransformer256x2-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 871, + "max": 10000, + "mean": 325.7910447761194, + "min": 119 + }, + "champion_root_simulations": { + "count": 848, + "max": 28, + "mean": 20.55424528301887, + "min": 6 + }, + "cutoffs": 3, + "losses": 0, + "unfinished": 0, + "wins": 9 + }, + "right": "querytransformer256x2-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 449, + "max": 10000, + "mean": 538.0890868596882, + "min": 89 + }, + "champion_root_simulations": { + "count": 426, + "max": 21, + "mean": 15.556338028169014, + "min": 7 + }, + "cutoffs": 1, + "losses": 0, + "unfinished": 0, + "wins": 11 + }, + "right": "querytransformer164x5-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 1456, + "max": 10000, + "mean": 245.4114010989011, + "min": 32 + }, + "champion_root_simulations": { + "count": 1439, + "max": 22, + "mean": 16.82209867963864, + "min": 6 + }, + "cutoffs": 7, + "losses": 0, + "unfinished": 0, + "wins": 5 + }, + "right": "querytransformer164x5-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 550, + "max": 10000, + "mean": 202.44181818181818, + "min": 54 + }, + "champion_root_simulations": { + "count": 529, + "max": 10000, + "mean": 244.84120982986767, + "min": 46 + }, + "cutoffs": 0, + "losses": 5, + "unfinished": 0, + "wins": 7 + }, + "right": "hexcnn32-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 393, + "max": 10000, + "mean": 304.5394402035623, + "min": 53 + }, + "champion_root_simulations": { + "count": 393, + "max": 10000, + "mean": 319.1501272264631, + "min": 61 + }, + "cutoffs": 0, + "losses": 6, + "unfinished": 0, + "wins": 6 + }, + "right": "hexcnn32-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 493, + "max": 10000, + "mean": 379.5496957403651, + "min": 63 + }, + "champion_root_simulations": { + "count": 464, + "max": 38, + "mean": 22.601293103448278, + "min": 12 + }, + "cutoffs": 0, + "losses": 0, + "unfinished": 0, + "wins": 12 + }, + "right": "querytransformer256x2-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 142, + "max": 10000, + "mean": 1453.8661971830986, + "min": 64 + }, + "champion_root_simulations": { + "count": 127, + "max": 641, + "mean": 26.62992125984252, + "min": 11 + }, + "cutoffs": 0, + "losses": 0, + "unfinished": 0, + "wins": 12 + }, + "right": "querytransformer256x2-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 383, + "max": 10000, + "mean": 459.7389033942559, + "min": 53 + }, + "champion_root_simulations": { + "count": 360, + "max": 21, + "mean": 16.216666666666665, + "min": 8 + }, + "cutoffs": 0, + "losses": 0, + "unfinished": 0, + "wins": 12 + }, + "right": "querytransformer164x5-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 334, + "max": 10000, + "mean": 640.0688622754491, + "min": 60 + }, + "champion_root_simulations": { + "count": 321, + "max": 21, + "mean": 17.14018691588785, + "min": 11 + }, + "cutoffs": 1, + "losses": 0, + "unfinished": 0, + "wins": 11 + }, + "right": "querytransformer164x5-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "hexcnn32-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 213, + "max": 10000, + "mean": 1017.1032863849765, + "min": 61 + }, + "champion_root_simulations": { + "count": 196, + "max": 27, + "mean": 22.714285714285715, + "min": 14 + }, + "cutoffs": 0, + "losses": 0, + "unfinished": 0, + "wins": 12 + }, + "right": "querytransformer256x2-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "hexcnn32-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 117, + "max": 10000, + "mean": 1548.6581196581196, + "min": 64 + }, + "champion_root_simulations": { + "count": 110, + "max": 25, + "mean": 21.654545454545456, + "min": 15 + }, + "cutoffs": 0, + "losses": 0, + "unfinished": 0, + "wins": 12 + }, + "right": "querytransformer256x2-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "hexcnn32-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 925, + "max": 10000, + "mean": 197.71243243243242, + "min": 49 + }, + "champion_root_simulations": { + "count": 911, + "max": 21, + "mean": 16.644346871569702, + "min": 6 + }, + "cutoffs": 4, + "losses": 0, + "unfinished": 0, + "wins": 8 + }, + "right": "querytransformer164x5-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "hexcnn32-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 368, + "max": 10000, + "mean": 533.054347826087, + "min": 58 + }, + "champion_root_simulations": { + "count": 354, + "max": 21, + "mean": 16.86723163841808, + "min": 8 + }, + "cutoffs": 1, + "losses": 0, + "unfinished": 0, + "wins": 11 + }, + "right": "querytransformer164x5-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "querytransformer256x2-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_cpu_time-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 768, + "max": 10000, + "mean": 73.85807291666667, + "min": 14 + }, + "champion_root_simulations": { + "count": 779, + "max": 10000, + "mean": 81.48395378690628, + "min": 6 + }, + "cutoffs": 2, + "losses": 7, + "unfinished": 0, + "wins": 3 + }, + "right": "querytransformer164x5-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "querytransformer256x2-seed410102", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_cpu_time-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 1350, + "max": 10000, + "mean": 57.968148148148146, + "min": 12 + }, + "champion_root_simulations": { + "count": 1330, + "max": 22, + "mean": 16.8203007518797, + "min": 9 + }, + "cutoffs": 6, + "losses": 0, + "unfinished": 0, + "wins": 6 + }, + "right": "querytransformer164x5-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410101-vs-squarecnn32-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 808, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 819, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 2, + "losses": 7, + "unfinished": 0, + "wins": 3 + }, + "right": "squarecnn32-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410102-vs-squarecnn32-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 821, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 821, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 3, + "losses": 8, + "unfinished": 0, + "wins": 1 + }, + "right": "squarecnn32-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 731, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 739, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 2, + "losses": 9, + "unfinished": 0, + "wins": 1 + }, + "right": "hexcnn32-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 726, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 722, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 3, + "losses": 9, + "unfinished": 0, + "wins": 0 + }, + "right": "hexcnn32-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 1245, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 1217, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 5, + "losses": 2, + "unfinished": 0, + "wins": 5 + }, + "right": "querytransformer256x2-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 950, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 937, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 4, + "losses": 0, + "unfinished": 0, + "wins": 8 + }, + "right": "querytransformer256x2-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 473, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 454, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 0, + "losses": 1, + "unfinished": 0, + "wins": 11 + }, + "right": "querytransformer164x5-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "mlp256-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/mlp256-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 1607, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 1598, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 8, + "losses": 1, + "unfinished": 0, + "wins": 3 + }, + "right": "querytransformer164x5-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-hexcnn32-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 517, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 494, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 1, + "losses": 1, + "unfinished": 0, + "wins": 10 + }, + "right": "hexcnn32-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-hexcnn32-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 585, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 677, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 2, + "losses": 6, + "unfinished": 0, + "wins": 4 + }, + "right": "hexcnn32-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 649, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 617, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 1, + "losses": 0, + "unfinished": 0, + "wins": 11 + }, + "right": "querytransformer256x2-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 546, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 511, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 2, + "losses": 0, + "unfinished": 0, + "wins": 10 + }, + "right": "querytransformer256x2-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 599, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 570, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 1, + "losses": 1, + "unfinished": 0, + "wins": 10 + }, + "right": "querytransformer164x5-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "squarecnn32-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 344, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 328, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 1, + "losses": 0, + "unfinished": 0, + "wins": 11 + }, + "right": "querytransformer164x5-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": true, + "left": "hexcnn32-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer256x2-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 359, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 352, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 1, + "losses": 2, + "unfinished": 0, + "wins": 9 + }, + "right": "querytransformer256x2-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "hexcnn32-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/hexcnn32-seed410102-vs-querytransformer256x2-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 339, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 324, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 1, + "losses": 0, + "unfinished": 0, + "wins": 11 + }, + "right": "querytransformer256x2-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": false, + "left": "hexcnn32-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/hexcnn32-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 1184, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 1172, + "max": 32, + "mean": 31.907849829351537, + "min": 5 + }, + "cutoffs": 0, + "losses": 0, + "unfinished": 6, + "wins": 6 + }, + "right": "querytransformer164x5-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": true, + "left": "hexcnn32-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/hexcnn32-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 367, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "champion_root_simulations": { + "count": 353, + "max": 32, + "mean": 32.0, + "min": 32 + }, + "cutoffs": 1, + "losses": 0, + "unfinished": 0, + "wins": 11 + }, + "right": "querytransformer164x5-seed410102", + "seed": 1380102 + }, + { + "attempt": 1, + "complete": false, + "left": "querytransformer256x2-seed410101", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/querytransformer256x2-seed410101-vs-querytransformer164x5-seed410101-equal_simulations-seed1380101-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 825, + "max": 32, + "mean": 31.92969696969697, + "min": 3 + }, + "champion_root_simulations": { + "count": 817, + "max": 32, + "mean": 31.88249694002448, + "min": 0 + }, + "cutoffs": 0, + "losses": 2, + "unfinished": 5, + "wins": 5 + }, + "right": "querytransformer164x5-seed410101", + "seed": 1380101 + }, + { + "attempt": 1, + "complete": false, + "left": "querytransformer256x2-seed410102", + "mode": "equal_simulations", + "report": "reports/overnight/20260909T235853Z/querytransformer256x2-seed410102-vs-querytransformer164x5-seed410102-equal_simulations-seed1380102-attempt1.json", + "result": { + "candidate_root_simulations": { + "count": 684, + "max": 32, + "mean": 31.941520467836256, + "min": 12 + }, + "champion_root_simulations": { + "count": 676, + "max": 32, + "mean": 31.952662721893493, + "min": 0 + }, + "cutoffs": 0, + "losses": 0, + "unfinished": 3, + "wins": 9 + }, + "right": "querytransformer164x5-seed410102", + "seed": 1380102 + }, + { + "attempt": 3, + "complete": true, + "left": "squarecnn32-seed410101", + "mode": "equal_cpu_time", + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-hexcnn32-seed410102-equal_cpu_time-seed989001-attempt3.json", + "result": { + "candidate_root_simulations": { + "count": 3382, + "max": 10000, + "mean": 234.5408042578356, + "min": 4 + }, + "champion_root_simulations": { + "count": 3326, + "max": 10000, + "mean": 271.61996392062537, + "min": 37 + }, + "cutoffs": 0, + "losses": 38, + "unfinished": 0, + "wins": 42 + }, + "right": "hexcnn32-seed410102", + "seed": 989001 + } + ], + "errors": [ + { + "label": "huggingface-upload", + "message": "huggingface-upload exited 1", + "type": "RuntimeError" + }, + { + "label": "github-push", + "message": "github-push exited 128", + "type": "RuntimeError" + } + ], + "promotion": { + "candidate": "checkpoints/overnight/20260909T235853Z/selected-final-cb19bc298969.pt", + "duel_report": "reports/overnight/20260909T235853Z/final-vs-served.json", + "greedy_report": null, + "passed": false + }, + "publication": { + "github_commit": true, + "github_push": false, + "github_release": false, + "hf_repo": "wuschelschulz/gipf-zero", + "source_hash": "aa533bb8922231a3e0dc53dd2ff947cab6899ef09a9245463c89ac99ea776ef0", + "uploaded": false + }, + "ranking": { + "alternative": "hexcnn32", + "direct_wilson_95": [ + 0.3507486535891948, + 0.72108666268789 + ], + "inconclusive": true, + "rows": [ + { + "mean_root_simulations": 524.4018087810757, + "name": "squarecnn32", + "rate": 0.8333333333333334, + "total": 96, + "unplayed": 0, + "wilson_95": [ + 0.746290098481051, + 0.8947261762804988 + ], + "wins": 80 + }, + { + "mean_root_simulations": 573.3016757505798, + "name": "hexcnn32", + "rate": 0.7708333333333334, + "total": 96, + "unplayed": 0, + "wilson_95": [ + 0.6773148946997618, + 0.8435108285439974 + ], + "wins": 74 + }, + { + "mean_root_simulations": 303.709794038576, + "name": "mlp256", + "rate": 0.4270833333333333, + "total": 96, + "unplayed": 0, + "wilson_95": [ + 0.33282083114008043, + 0.5269568587558306 + ], + "wins": 41 + }, + { + "mean_root_simulations": 33.448988035627366, + "name": "querytransformer256x2", + "rate": 0.09375, + "total": 96, + "unplayed": 0, + "wilson_95": [ + 0.050108108212146014, + 0.16865330692221547 + ], + "wins": 9 + }, + { + "mean_root_simulations": 24.69389041739199, + "name": "querytransformer164x5", + "rate": 0.07291666666666667, + "total": 96, + "unplayed": 0, + "wilson_95": [ + 0.03576758131036273, + 0.1429303166514019 + ], + "wins": 7 + } + ], + "seed_scores": { + "410101": [ + 40, + 48 + ], + "410102": [ + 40, + 48 + ] + }, + "tie_break": { + "alternative_seed_scores": { + "410101": [ + 33, + 48 + ], + "410102": [ + 41, + 48 + ] + }, + "report": "reports/overnight/20260909T235853Z/squarecnn32-seed410101-vs-hexcnn32-seed410102-equal_cpu_time-seed989001-attempt3.json", + "selected": "hexcnn32", + "wilson_95": [ + 0.41695642649729947, + 0.6307526670259244 + ] + }, + "tie_rule": "higher achieved CPU root simulations within the fixed 50 ms cap only after an extra head-to-head remains inconclusive", + "winner": "hexcnn32", + "winner_seed": 410102 + }, + "status": "partial", + "tag": "20260909T235853Z", + "training": [ + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/mlp256-seed410101-9acc8f0f15e5.pt", + "name": "mlp256-seed410101", + "run": "runs/overnight/20260909T235853Z/mlp256-seed410101" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/mlp256-seed410102-9ca408753dfb.pt", + "name": "mlp256-seed410102", + "run": "runs/overnight/20260909T235853Z/mlp256-seed410102" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/squarecnn32-seed410101-e2a8946b1405.pt", + "name": "squarecnn32-seed410101", + "run": "runs/overnight/20260909T235853Z/squarecnn32-seed410101" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/squarecnn32-seed410102-d3e8eb4f1e75.pt", + "name": "squarecnn32-seed410102", + "run": "runs/overnight/20260909T235853Z/squarecnn32-seed410102" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/hexcnn32-seed410101-d6f1c7081cdb.pt", + "name": "hexcnn32-seed410101", + "run": "runs/overnight/20260909T235853Z/hexcnn32-seed410101" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/hexcnn32-seed410102-c7d5bd9e9ee0.pt", + "name": "hexcnn32-seed410102", + "run": "runs/overnight/20260909T235853Z/hexcnn32-seed410102" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/querytransformer256x2-seed410101-454d54089b4a.pt", + "name": "querytransformer256x2-seed410101", + "run": "runs/overnight/20260909T235853Z/querytransformer256x2-seed410101" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/querytransformer256x2-seed410102-a0a1d4c39231.pt", + "name": "querytransformer256x2-seed410102", + "run": "runs/overnight/20260909T235853Z/querytransformer256x2-seed410102" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410101-8568880b8377.pt", + "name": "querytransformer164x5-seed410101", + "run": "runs/overnight/20260909T235853Z/querytransformer164x5-seed410101" + }, + { + "completed": true, + "frozen": "checkpoints/overnight/20260909T235853Z/querytransformer164x5-seed410102-02b9f1909248.pt", + "name": "querytransformer164x5-seed410102", + "run": "runs/overnight/20260909T235853Z/querytransformer164x5-seed410102" + } + ], + "training_stop": 1789018200.0, + "version": "overnight-v1" +}