Cccccz/HY

1{2  "output": "/mnt/local_nvme/cz/projects/HY-WorldPlay-DEV-Predictor/datasets/hyworldplay_predictor_confidence_validation25_prefeature/cases.jsonl",3  "cases": 254}5{6  "records": 800,7  "train_records": 7008}9WARNING:__main__:10*****************************************11Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. 12*****************************************13rank0 extracted 25/133 batches14rank0 extracted 50/133 batches15rank0 extracted 75/133 batches16rank0 extracted 100/133 batches17rank0 extracted 125/133 batches18Saved 8 feature shards to /mnt/local_nvme/cz/projects/HY-WorldPlay-DEV-Predictor/confidence_validation/features/train19WARNING:__main__:20*****************************************21Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. 22*****************************************23rank0 extracted 25/35 batches24Saved 8 feature shards to /mnt/local_nvme/cz/projects/HY-WorldPlay-DEV-Predictor/confidence_validation/features/validation25WARNING:__main__:26*****************************************27Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. 28*****************************************29[rank0]:[W819 05:29:45.641755152 ProcessGroupNCCL.cpp:4115] [PG ID 0 PG GUID 0 Rank 0]  using GPU 0 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect.Specify device_ids in barrier() to force use of a particular device,or call init_process_group() with a device_id.30[rank6]:[W819 05:29:45.820996456 ProcessGroupNCCL.cpp:4115] [PG ID 0 PG GUID 0 Rank 6]  using GPU 6 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect.Specify device_ids in barrier() to force use of a particular device,or call init_process_group() with a device_id.31[rank1]:[W819 05:29:45.868540764 ProcessGroupNCCL.cpp:4115] [PG ID 0 PG GUID 0 Rank 1]  using GPU 1 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect.Specify device_ids in barrier() to force use of a particular device,or call init_process_group() with a device_id.32[rank2]:[W819 05:29:46.322682030 ProcessGroupNCCL.cpp:4115] [PG ID 0 PG GUID 0 Rank 2]  using GPU 2 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect.Specify device_ids in barrier() to force use of a particular device,or call init_process_group() with a device_id.33[rank7]:[W819 05:29:46.386567270 ProcessGroupNCCL.cpp:4115] [PG ID 0 PG GUID 0 Rank 7]  using GPU 7 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect.Specify device_ids in barrier() to force use of a particular device,or call init_process_group() with a device_id.34[rank5]:[W819 05:29:46.398470481 ProcessGroupNCCL.cpp:4115] [PG ID 0 PG GUID 0 Rank 5]  using GPU 5 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect.Specify device_ids in barrier() to force use of a particular device,or call init_process_group() with a device_id.35[rank4]:[W819 05:29:46.426298692 ProcessGroupNCCL.cpp:4115] [PG ID 0 PG GUID 0 Rank 4]  using GPU 4 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect.Specify device_ids in barrier() to force use of a particular device,or call init_process_group() with a device_id.36[rank3]:[W819 05:29:46.471978908 ProcessGroupNCCL.cpp:4115] [PG ID 0 PG GUID 0 Rank 3]  using GPU 3 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect.Specify device_ids in barrier() to force use of a particular device,or call init_process_group() with a device_id.37c0 epoch 1/40 loss=1.95157338c0 epoch 10/40 loss=0.07542239c0 epoch 20/40 loss=0.03532340c0 epoch 30/40 loss=0.02595441c0 epoch 40/40 loss=0.02048842c1 epoch 1/40 loss=3.38329943c1 epoch 10/40 loss=0.14925344c1 epoch 20/40 loss=0.04659145c1 epoch 30/40 loss=0.02969946c1 epoch 40/40 loss=0.02291247c2 epoch 1/40 loss=2.85527448c2 epoch 10/40 loss=0.10903449c2 epoch 20/40 loss=0.04001450c2 epoch 30/40 loss=0.02911951c2 epoch 40/40 loss=0.02253652{53  "protocol": "WorldPlay fulltrain100 (8400 transitions); independent validation25 (2100 transitions); no validation tuning",54  "train_samples": 8400,55  "validation_samples": 2100,56  "c1_validation_metrics": {57    "overall": {58      "model": "Confidence C1",59      "split": "validation",60      "step": "overall",61      "spearman": 0.9529179491248462,62      "pearson": 0.96539306640625,63      "log_mae": 0.1599988937377929764    },65    "P1": {66      "model": "Confidence C1",67      "split": "validation",68      "step": "P1",69      "spearman": 0.8590718873187204,70      "pearson": 0.8829622268676758,71      "log_mae": 0.1769156455993652372    },73    "P2": {74      "model": "Confidence C1",75      "split": "validation",76      "step": "P2",77      "spearman": 0.8631690385373965,78      "pearson": 0.9074602127075195,79      "log_mae": 0.1450339406728744580    },81    "P3": {82      "model": "Confidence C1",83      "split": "validation",84      "step": "P3",85      "spearman": 0.8700319930976419,86      "pearson": 0.8872363567352295,87      "log_mae": 0.1580470949411392288    }89  },90  "remaining_risk_at_33pct": {91    "Random": 0.009558672141283751,92    "Timestep-only": 0.005005753133445978,93    "Confidence C0": 0.008430036716163158,94    "Confidence C1": 0.004878982435911894,95    "Confidence C2": 0.004925357177853584,96    "Oracle": 0.00484056910499930497  },98  "gain_vs_random": 0.48957529207015615,99  "oracle_gap": 0.007935705508866102,100  "routing_at_33pct": {101    "P1": {102      "selection_fraction": 0.012857142857142857,103      "within_step_fallback_rate": 0.012857142857142857104    },105    "P2": {106      "selection_fraction": 0.05714285714285714,107      "within_step_fallback_rate": 0.05714285714285714108    },109    "P3": {110      "selection_fraction": 0.93,111      "within_step_fallback_rate": 0.93112    }113  }114}115