{ "name": "root", "gauges": { "SoccerTwos.Policy.Entropy.mean": { "value": 1.9916212558746338, "min": 1.9346994161605835, "max": 2.040637254714966, "count": 22 }, "SoccerTwos.Policy.Entropy.sum": { "value": 79856.046875, "min": 3466.9814453125, "max": 83491.359375, "count": 22 }, "SoccerTwos.Environment.EpisodeLength.mean": { "value": 43.10043668122271, "min": 30.636363636363637, "max": 50.04081632653061, "count": 22 }, "SoccerTwos.Environment.EpisodeLength.sum": { "value": 39480.0, "min": 1348.0, "max": 39480.0, "count": 22 }, "SoccerTwos.Self-play.ELO.mean": { "value": 1591.110419882527, "min": 1570.81481481915, "max": 1599.5770602729604, "count": 22 }, "SoccerTwos.Self-play.ELO.sum": { "value": 728728.5723061974, "min": 34808.60446692368, "max": 799140.6304508327, "count": 22 }, "SoccerTwos.Step.mean": { "value": 9999874.0, "min": 9579978.0, "max": 9999874.0, "count": 22 }, "SoccerTwos.Step.sum": { "value": 9999874.0, "min": 9579978.0, "max": 9999874.0, "count": 22 }, "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": { "value": 0.0063628037460148335, "min": -0.03889418765902519, "max": 0.06105382367968559, "count": 22 }, "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": { "value": 2.907801389694214, "min": -15.479887008666992, "max": 23.305809020996094, "count": 22 }, "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": { "value": 0.0062186759896576405, "min": -0.03762693330645561, "max": 0.0645577684044838, "count": 22 }, "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": { "value": 2.8419349193573, "min": -15.878498077392578, "max": 24.109832763671875, "count": 22 }, "SoccerTwos.Environment.CumulativeReward.mean": { "value": 0.0, "min": 0.0, "max": 0.0, "count": 22 }, "SoccerTwos.Environment.CumulativeReward.sum": { "value": 0.0, "min": 0.0, "max": 0.0, "count": 22 }, "SoccerTwos.Policy.ExtrinsicReward.mean": { "value": -0.015454705486673234, "min": -0.1476120163442751, "max": 0.10037647178642164, "count": 22 }, "SoccerTwos.Policy.ExtrinsicReward.sum": { "value": -7.062800407409668, "min": -68.78719961643219, "max": 49.48560059070587, "count": 22 }, "SoccerTwos.Environment.GroupCumulativeReward.mean": { "value": -0.015454705486673234, "min": -0.1476120163442751, "max": 0.10037647178642164, "count": 22 }, "SoccerTwos.Environment.GroupCumulativeReward.sum": { "value": -7.062800407409668, "min": -68.78719961643219, "max": 49.48560059070587, "count": 22 }, "SoccerTwos.IsTraining.mean": { "value": 1.0, "min": 1.0, "max": 1.0, "count": 22 }, "SoccerTwos.IsTraining.sum": { "value": 1.0, "min": 1.0, "max": 1.0, "count": 22 }, "SoccerTwos.Losses.PolicyLoss.mean": { "value": 0.018135219111496553, "min": 0.017422182935843012, "max": 0.018135219111496553, "count": 4 }, "SoccerTwos.Losses.PolicyLoss.sum": { "value": 0.018135219111496553, "min": 0.017422182935843012, "max": 0.018135219111496553, "count": 4 }, "SoccerTwos.Losses.ValueLoss.mean": { "value": 0.08954002186655999, "min": 0.08799008548259735, "max": 0.08954002186655999, "count": 4 }, "SoccerTwos.Losses.ValueLoss.sum": { "value": 0.08954002186655999, "min": 0.08799008548259735, "max": 0.08954002186655999, "count": 4 }, "SoccerTwos.Losses.BaselineLoss.mean": { "value": 0.0894136626124382, "min": 0.08891006422042846, "max": 0.08992617464065551, "count": 4 }, "SoccerTwos.Losses.BaselineLoss.sum": { "value": 0.0894136626124382, "min": 0.08891006422042846, "max": 0.08992617464065551, "count": 4 }, "SoccerTwos.Policy.LearningRate.mean": { "value": 5.372998925600155e-07, "min": 5.372998925600155e-07, "max": 1.590259681949999e-05, "count": 4 }, "SoccerTwos.Policy.LearningRate.sum": { "value": 5.372998925600155e-07, "min": 5.372998925600155e-07, "max": 1.590259681949999e-05, "count": 4 }, "SoccerTwos.Policy.Epsilon.mean": { "value": 0.10010744, "min": 0.10010744, "max": 0.10318050000000002, "count": 4 }, "SoccerTwos.Policy.Epsilon.sum": { "value": 0.10010744, "min": 0.10010744, "max": 0.10318050000000002, "count": 4 }, "SoccerTwos.Policy.Beta.mean": { "value": 1.536125600000015e-05, "min": 1.536125600000015e-05, "max": 0.00016870694999999988, "count": 4 }, "SoccerTwos.Policy.Beta.sum": { "value": 1.536125600000015e-05, "min": 1.536125600000015e-05, "max": 0.00016870694999999988, "count": 4 } }, "metadata": { "timer_format_version": "0.1.0", "start_time_seconds": "1733560359", "python_version": "3.10.12 (main, Jul 5 2023, 15:34:07) [Clang 14.0.6 ]", "command_line_arguments": "/Users/mango/anaconda3/envs/rl/bin/mlagents-learn ./config/poca/SoccerTwos.yaml --env=./training-envs-executables/SoccerTwos/SoccerTwos.app --run-id=SoccerTwos-v2 --no-graphics --resume", "mlagents_version": "1.2.0.dev0", "mlagents_envs_version": "1.2.0.dev0", "communication_protocol_version": "1.5.0", "pytorch_version": "2.2.2", "numpy_version": "1.23.5", "end_time_seconds": "1733563316" }, "total": 2957.137057906002, "count": 1, "self": 1.3709337539912667, "children": { "run_training.setup": { "total": 0.09655873500742018, "count": 1, "self": 0.09655873500742018 }, "TrainerController.start_learning": { "total": 2955.6695654170035, "count": 1, "self": 0.5951568851160118, "children": { "TrainerController._reset_env": { "total": 7.347605434988509, "count": 3, "self": 7.347605434988509 }, "TrainerController.advance": { "total": 2947.538644401895, "count": 30071, "self": 0.5614396156015573, "children": { "env_step": { "total": 2116.8447047862865, "count": 30071, "self": 2045.0511811070028, "children": { "SubprocessEnvManager._take_step": { "total": 71.43473334351438, "count": 30071, "self": 2.8646667561260983, "children": { "TorchPolicy.evaluate": { "total": 68.57006658738828, "count": 52724, "self": 68.57006658738828 } } }, "workers": { "total": 0.3587903357692994, "count": 30071, "self": 0.0, "children": { "worker_root": { "total": 2948.541456867024, "count": 30071, "is_parallel": true, "self": 964.6685758058011, "children": { "steps_from_proto": { "total": 0.011270272021647543, "count": 6, "is_parallel": true, "self": 0.0017144339653896168, "children": { "_process_rank_one_or_two_observation": { "total": 0.009555838056257926, "count": 24, "is_parallel": true, "self": 0.009555838056257926 } } }, "UnityEnvironment.step": { "total": 1983.8616107892012, "count": 30071, "is_parallel": true, "self": 6.1401000669866335, "children": { "UnityEnvironment._generate_step_input": { "total": 39.32464096340118, "count": 30071, "is_parallel": true, "self": 39.32464096340118 }, "communicator.exchange": { "total": 1865.1993332414131, "count": 30071, "is_parallel": true, "self": 1865.1993332414131 }, "steps_from_proto": { "total": 73.1975365174003, "count": 60142, "is_parallel": true, "self": 9.851312138678622, "children": { "_process_rank_one_or_two_observation": { "total": 63.34622437872167, "count": 240568, "is_parallel": true, "self": 63.34622437872167 } } } } } } } } } } }, "trainer_advance": { "total": 830.132500000007, "count": 30071, "self": 3.491149840861908, "children": { "process_trajectory": { "total": 90.44449889313546, "count": 30071, "self": 90.10556022413948, "children": { "RLTrainer._checkpoint": { "total": 0.33893866899597924, "count": 1, "self": 0.33893866899597924 } } }, "_update_policy": { "total": 736.1968512660096, "count": 4, "self": 82.8476835929905, "children": { "TorchPOCAOptimizer.update": { "total": 653.3491676730191, "count": 1000, "self": 653.3491676730191 } } } } } } }, "trainer_threads": { "total": 8.119968697428703e-07, "count": 1, "self": 8.119968697428703e-07 }, "TrainerController._save_models": { "total": 0.18815788300707936, "count": 1, "self": 0.0019104440143564716, "children": { "RLTrainer._checkpoint": { "total": 0.1862474389927229, "count": 1, "self": 0.1862474389927229 } } } } } } }