{ "name": "root", "gauges": { "SoccerTwos.Policy.Entropy.mean": { "value": 0.657804012298584, "min": 0.6439549922943115, "max": 0.9220815896987915, "count": 4000 }, "SoccerTwos.Policy.Entropy.sum": { "value": 13324.4775390625, "min": 9013.736328125, "max": 31204.041015625, "count": 4000 }, "SoccerTwos.Environment.EpisodeLength.mean": { "value": 70.84057971014492, "min": 37.41984732824427, "max": 118.0952380952381, "count": 4000 }, "SoccerTwos.Environment.EpisodeLength.sum": { "value": 19552.0, "min": 16444.0, "max": 22664.0, "count": 4000 }, "SoccerTwos.Self-play.ELO.mean": { "value": 1785.7870992616793, "min": 1665.5704364111905, "max": 1848.444050226308, "count": 4000 }, "SoccerTwos.Self-play.ELO.sum": { "value": 246438.61969811172, "min": 149123.18746302405, "max": 451733.0875595351, "count": 4000 }, "SoccerTwos.Step.mean": { "value": 99999950.0, "min": 60009926.0, "max": 99999950.0, "count": 4000 }, "SoccerTwos.Step.sum": { "value": 99999950.0, "min": 60009926.0, "max": 99999950.0, "count": 4000 }, "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": { "value": -0.02403123490512371, "min": -0.12348798662424088, "max": 0.05139973759651184, "count": 4000 }, "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": { "value": -3.316310405731201, "min": -20.992958068847656, "max": 7.915559768676758, "count": 4000 }, "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": { "value": -0.023414287716150284, "min": -0.12312929332256317, "max": 0.05094476044178009, "count": 4000 }, "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": { "value": -3.2311716079711914, "min": -20.93198013305664, "max": 7.845493316650391, "count": 4000 }, "SoccerTwos.Environment.CumulativeReward.mean": { "value": 0.0, "min": 0.0, "max": 0.0, "count": 4000 }, "SoccerTwos.Environment.CumulativeReward.sum": { "value": 0.0, "min": 0.0, "max": 0.0, "count": 4000 }, "SoccerTwos.Policy.ExtrinsicReward.mean": { "value": -0.04548695899438167, "min": -0.4755757586522536, "max": 0.3431596885355868, "count": 4000 }, "SoccerTwos.Policy.ExtrinsicReward.sum": { "value": -6.27720034122467, "min": -62.77600014209747, "max": 51.101200103759766, "count": 4000 }, "SoccerTwos.Environment.GroupCumulativeReward.mean": { "value": -0.04548695899438167, "min": -0.4755757586522536, "max": 0.3431596885355868, "count": 4000 }, "SoccerTwos.Environment.GroupCumulativeReward.sum": { "value": -6.27720034122467, "min": -62.77600014209747, "max": 51.101200103759766, "count": 4000 }, "SoccerTwos.IsTraining.mean": { "value": 1.0, "min": 1.0, "max": 1.0, "count": 4000 }, "SoccerTwos.IsTraining.sum": { "value": 1.0, "min": 1.0, "max": 1.0, "count": 4000 }, "SoccerTwos.Losses.PolicyLoss.mean": { "value": 0.012369273196478995, "min": 0.00639950543481973, "max": 0.018580751745806385, "count": 973 }, "SoccerTwos.Losses.PolicyLoss.sum": { "value": 0.012369273196478995, "min": 0.00639950543481973, "max": 0.018580751745806385, "count": 973 }, "SoccerTwos.Losses.ValueLoss.mean": { "value": 0.09552693540851274, "min": 0.07807423248887062, "max": 0.11450178325176238, "count": 973 }, "SoccerTwos.Losses.ValueLoss.sum": { "value": 0.09552693540851274, "min": 0.07807423248887062, "max": 0.11450178325176238, "count": 973 }, "SoccerTwos.Losses.BaselineLoss.mean": { "value": 0.09561899180213611, "min": 0.07871665557225545, "max": 0.11573187485337258, "count": 973 }, "SoccerTwos.Losses.BaselineLoss.sum": { "value": 0.09561899180213611, "min": 0.07871665557225545, "max": 0.11573187485337258, "count": 973 }, "SoccerTwos.Policy.LearningRate.mean": { "value": 1.0323999147947604e-08, "min": 1.0323999147947604e-08, "max": 0.00047950611604116196, "count": 973 }, "SoccerTwos.Policy.LearningRate.sum": { "value": 1.0323999147947604e-08, "min": 1.0323999147947604e-08, "max": 0.00047950611604116196, "count": 973 }, "SoccerTwos.Policy.Epsilon.mean": { "value": 0.100000852, "min": 0.100000852, "max": 0.13995883800000003, "count": 973 }, "SoccerTwos.Policy.Epsilon.sum": { "value": 0.100000852, "min": 0.100000852, "max": 0.13995883800000003, "count": 973 }, "SoccerTwos.Policy.Beta.mean": { "value": 1.0042514799999778e-05, "min": 1.0042514799999778e-05, "max": 0.0020039460162, "count": 973 }, "SoccerTwos.Policy.Beta.sum": { "value": 1.0042514799999778e-05, "min": 1.0042514799999778e-05, "max": 0.0020039460162, "count": 973 } }, "metadata": { "timer_format_version": "0.1.0", "start_time_seconds": "1676307162", "python_version": "3.9.16 (main, Jan 11 2023, 16:05:54) \n[GCC 11.2.0]", "command_line_arguments": "/home/dfm/anaconda3/envs/hf-rl-scr2s/bin/mlagents-learn ./config/poca/SoccerTwos-2x-12-resnet-l2-linear.yaml --env=./training-envs-executables/SoccerTwos/SoccerTwos.x86_64 --run-id=SoccerTwos --no-graphics --results-dir results-2x-12-resnet-l2-linear --num-envs=7 --resume --run-id=SoccerTwos", "mlagents_version": "0.31.0.dev0", "mlagents_envs_version": "0.31.0.dev0", "communication_protocol_version": "1.5.0", "pytorch_version": "1.11.0+cu102", "numpy_version": "1.21.2", "end_time_seconds": "1676346654" }, "total": 39491.56532316399, "count": 1, "self": 0.37074283976107836, "children": { "run_training.setup": { "total": 0.02767869713716209, "count": 1, "self": 0.02767869713716209 }, "TrainerController.start_learning": { "total": 39491.16690162709, "count": 1, "self": 33.59568607318215, "children": { "TrainerController._reset_env": { "total": 27.265360637335107, "count": 201, "self": 27.265360637335107 }, "TrainerController.advance": { "total": 39430.04803913669, "count": 930879, "self": 27.246289231115952, "children": { "env_step": { "total": 20429.227719880873, "count": 930879, "self": 7700.056544515071, "children": { "SubprocessEnvManager._take_step": { "total": 12695.632731482387, "count": 2814983, "self": 386.68545385845937, "children": { "TorchPolicy.evaluate": { "total": 12308.947277623927, "count": 5174268, "self": 12308.947277623927 } } }, "workers": { "total": 33.53844388341531, "count": 930879, "self": 0.0, "children": { "worker_root": { "total": 276178.97632250306, "count": 2814280, "is_parallel": true, "self": 212880.03933615354, "children": { "run_training.setup": { "total": 0.0, "count": 0, "is_parallel": true, "self": 0.0, "children": { "steps_from_proto": { "total": 0.003266595071181655, "count": 2, "is_parallel": true, "self": 0.0009026317857205868, "children": { "_process_rank_one_or_two_observation": { "total": 0.002363963285461068, "count": 8, "is_parallel": true, "self": 0.002363963285461068 } } }, "UnityEnvironment.step": { "total": 0.04986436199396849, "count": 1, "is_parallel": true, "self": 0.001236666925251484, "children": { "UnityEnvironment._generate_step_input": { "total": 0.0010303319431841373, "count": 1, "is_parallel": true, "self": 0.0010303319431841373 }, "communicator.exchange": { "total": 0.0449853150639683, "count": 1, "is_parallel": true, "self": 0.0449853150639683 }, "steps_from_proto": { "total": 0.0026120480615645647, "count": 2, "is_parallel": true, "self": 0.0004060741048306227, "children": { "_process_rank_one_or_two_observation": { "total": 0.002205973956733942, "count": 8, "is_parallel": true, "self": 0.002205973956733942 } } } } } } }, "steps_from_proto": { "total": 5.926313872681931, "count": 2800, "is_parallel": true, "self": 0.8897827882319689, "children": { "_process_rank_one_or_two_observation": { "total": 5.036531084449962, "count": 11200, "is_parallel": true, "self": 5.036531084449962 } } }, "UnityEnvironment.step": { "total": 63293.01067247684, "count": 2814279, "is_parallel": true, "self": 5000.369204876246, "children": { "UnityEnvironment._generate_step_input": { "total": 2897.470194905065, "count": 2814279, "is_parallel": true, "self": 2897.470194905065 }, "communicator.exchange": { "total": 42666.08549953322, "count": 2814279, "is_parallel": true, "self": 42666.08549953322 }, "steps_from_proto": { "total": 12729.085773162311, "count": 5628558, "is_parallel": true, "self": 1811.1029579455499, "children": { "_process_rank_one_or_two_observation": { "total": 10917.982815216761, "count": 22514232, "is_parallel": true, "self": 10917.982815216761 } } } } } } } } } } }, "trainer_advance": { "total": 18973.574030024698, "count": 930879, "self": 424.5758848800324, "children": { "process_trajectory": { "total": 7482.0648319923785, "count": 930879, "self": 7462.430964581668, "children": { "RLTrainer._checkpoint": { "total": 19.63386741071008, "count": 80, "self": 19.63386741071008 } } }, "_update_policy": { "total": 11066.933313152287, "count": 973, "self": 7924.29891131958, "children": { "TorchPOCAOptimizer.update": { "total": 3142.634401832707, "count": 29190, "self": 3142.634401832707 } } } } } } }, "trainer_threads": { "total": 1.0898802429437637e-06, "count": 1, "self": 1.0898802429437637e-06 }, "TrainerController._save_models": { "total": 0.2578146900050342, "count": 1, "self": 0.0019481771159917116, "children": { "RLTrainer._checkpoint": { "total": 0.2558665128890425, "count": 1, "self": 0.2558665128890425 } } } } } } }