| { |
| "iteration": 48, |
| "timestamp": "2026-05-11T04:48:49.392550+00:00", |
| "buffer_size": 17625, |
| "config": { |
| "iterations": 180, |
| "episodes_per_iter": 20, |
| "mcts_sims": 160, |
| "c_puct": 1.5, |
| "temp_threshold": 28, |
| "add_noise": true, |
| "seed": 42, |
| "verbose_logs": false, |
| "episode_log_every": 25, |
| "epochs": 1, |
| "batch_size": 224, |
| "learning_rate": 0.0003, |
| "weight_decay": 0.0001, |
| "value_loss_coeff": 0.5, |
| "reward_shaping_enabled": true, |
| "reward_shaping_scale": 0.1, |
| "reward_shaping_gamma": 0.99, |
| "reward_shaping_material_weight": 0.6, |
| "reward_shaping_mobility_weight": 0.4, |
| "reward_shaping_draw_penalty": 0.1, |
| "league_enabled": true, |
| "league_path": "checkpoints/league_ratings.json", |
| "league_checkpoint_pool_top_k": 3, |
| "league_checkpoint_pool_min_games": 4, |
| "league_selfplay_checkpoint_prob": 0.55, |
| "league_champion_games": 6, |
| "buffer_size": 50000, |
| "val_split": 0.1, |
| "shuffle_train_val_split": true, |
| "train_recent_fraction": 0.7, |
| "train_recent_window_fraction": 0.4, |
| "d_model": 128, |
| "nhead": 8, |
| "num_layers": 6, |
| "dim_feedforward": 512, |
| "dropout": 0.1, |
| "log_dir": "/kaggle/working/logs", |
| "checkpoint_dir": "/kaggle/working/checkpoints", |
| "save_every": 6, |
| "keep_last_n_local_checkpoints": 3, |
| "keep_last_n_log_versions": 2, |
| "keep_last_n_hf_checkpoints": 3, |
| "onnx_path": "ataxx_model.onnx", |
| "export_onnx": true, |
| "hf_enabled": true, |
| "hf_repo_id": "dieg0code/ataxx-zero", |
| "hf_run_id": "policy_spatial_v9", |
| "hf_bootstrap_run_id": "policy_spatial_v8", |
| "hf_reset_iteration": true, |
| "hf_token_env": "HF_TOKEN", |
| "hf_local_dir": "/kaggle/working/hf_checkpoints", |
| "max_pending_hf_uploads": 2, |
| "hf_upload_future_timeout_s": 120.0, |
| "show_progress_bar": false, |
| "trainer_log_every_n_steps": 99999, |
| "monitor_log_every": 5, |
| "epoch_pulse_every": 2, |
| "num_workers": 4, |
| "persistent_workers": true, |
| "strict_probs": false, |
| "trainer_devices": 2, |
| "trainer_strategy": "ddp_spawn", |
| "trainer_precision": "16-mixed", |
| "trainer_benchmark": true, |
| "ddp_timeout_seconds": 180, |
| "mcts_use_amp": true, |
| "mcts_cache_size": 100000, |
| "mcts_leaf_batch_size": 32, |
| "fail_on_selfplay_parallel_error": true, |
| "fail_on_hf_upload_error": true, |
| "opponent_self_prob": 0.6, |
| "opponent_heuristic_prob": 0.25, |
| "opponent_random_prob": 0.15, |
| "opponent_heuristic_level": "normal", |
| "opponent_heuristic_epsilon": 0.1, |
| "opponent_heuristic_easy_prob": 0.2, |
| "opponent_heuristic_normal_prob": 0.5, |
| "opponent_heuristic_hard_prob": 0.3, |
| "model_side_swap_prob": 0.5, |
| "eval_enabled": true, |
| "eval_every": 6, |
| "eval_games": 64, |
| "eval_sims": 160, |
| "eval_heuristic_level": "hard", |
| "eval_heuristic_levels": "easy,normal,hard,apex,gambit,sentinel", |
| "restore_best_on_regression": true, |
| "eval_regression_delta": 0.06, |
| "eval_regression_patience": 2, |
| "selfplay_workers": 2, |
| "selfplay_progress_every_s": 120.0, |
| "selfplay_episode_timeout_s": 1800.0, |
| "compile_model": true, |
| "quiet_mode": false, |
| "warmup_games": 320, |
| "warmup_epochs": 4, |
| "warmup_heuristic_level": "sentinel", |
| "warmup_heuristic_levels": "hard,apex,sentinel" |
| }, |
| "stats": { |
| "replay_size": 17625, |
| "best_eval_score": 0.6041666666666666, |
| "train_loss_total": 2.1545732021331787, |
| "train_loss_value": 0.5328640937805176, |
| "train_loss_policy": 1.8881415128707886, |
| "train_value_mae": 0.5306381583213806, |
| "train_policy_accuracy": 0.3781518042087555, |
| "train_lr": 0.00029997737146914005, |
| "score": 0.5013020833333334, |
| "eval_total_wins": 192, |
| "eval_total_losses": 191, |
| "eval_total_draws": 1, |
| "eval_levels": "easy,normal,hard,apex,gambit,sentinel", |
| "eval_score_easy": 0.5625, |
| "eval_score_normal": 0.296875, |
| "eval_score_hard": 0.6796875, |
| "eval_score_apex": 0.703125, |
| "eval_score_gambit": 0.125, |
| "eval_score_sentinel": 0.640625 |
| } |
| } |
