{ "name": "root", "gauges": { "SoccerTwos.Policy.Entropy.mean": { "value": 1.1532409191131592, "min": 1.1196668148040771, "max": 3.2957353591918945, "count": 1000 }, "SoccerTwos.Policy.Entropy.sum": { "value": 23987.41015625, "min": 7871.72021484375, "max": 112492.0390625, "count": 1000 }, "SoccerTwos.Environment.EpisodeLength.mean": { "value": 62.79746835443038, "min": 45.00917431192661, "max": 999.0, "count": 1000 }, "SoccerTwos.Environment.EpisodeLength.sum": { "value": 19844.0, "min": 7992.0, "max": 30788.0, "count": 1000 }, "SoccerTwos.Self-play.ELO.mean": { "value": 1709.0109510622178, "min": 1190.391640367494, "max": 1720.9898581771342, "count": 991 }, "SoccerTwos.Self-play.ELO.sum": { "value": 270023.7302678304, "min": 2381.3820137520165, "max": 367918.0238814893, "count": 991 }, "SoccerTwos.Step.mean": { "value": 9999962.0, "min": 9412.0, "max": 9999962.0, "count": 1000 }, "SoccerTwos.Step.sum": { "value": 9999962.0, "min": 9412.0, "max": 9999962.0, "count": 1000 }, "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": { "value": 0.01338563859462738, "min": -0.11655034869909286, "max": 0.16845153272151947, "count": 1000 }, "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": { "value": 2.1149308681488037, "min": -22.494216918945312, "max": 23.403825759887695, "count": 1000 }, "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": { "value": 0.014068040996789932, "min": -0.11802836507558823, "max": 0.17341575026512146, "count": 1000 }, "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": { "value": 2.222750425338745, "min": -22.422487258911133, "max": 22.71746253967285, "count": 1000 }, "SoccerTwos.Environment.CumulativeReward.mean": { "value": 0.0, "min": 0.0, "max": 0.0, "count": 1000 }, "SoccerTwos.Environment.CumulativeReward.sum": { "value": 0.0, "min": 0.0, "max": 0.0, "count": 1000 }, "SoccerTwos.Policy.ExtrinsicReward.mean": { "value": -0.0220329112644437, "min": -0.5286444425582886, "max": 0.40512195447596105, "count": 1000 }, "SoccerTwos.Policy.ExtrinsicReward.sum": { "value": -3.4811999797821045, "min": -57.29039967060089, "max": 47.71400010585785, "count": 1000 }, "SoccerTwos.Environment.GroupCumulativeReward.mean": { "value": -0.0220329112644437, "min": -0.5286444425582886, "max": 0.40512195447596105, "count": 1000 }, "SoccerTwos.Environment.GroupCumulativeReward.sum": { "value": -3.4811999797821045, "min": -57.29039967060089, "max": 47.71400010585785, "count": 1000 }, "SoccerTwos.IsTraining.mean": { "value": 1.0, "min": 1.0, "max": 1.0, "count": 1000 }, "SoccerTwos.IsTraining.sum": { "value": 1.0, "min": 1.0, "max": 1.0, "count": 1000 }, "SoccerTwos.Losses.PolicyLoss.mean": { "value": 0.018708063105683927, "min": 0.01079763230518438, "max": 0.0237867681561814, "count": 482 }, "SoccerTwos.Losses.PolicyLoss.sum": { "value": 0.018708063105683927, "min": 0.01079763230518438, "max": 0.0237867681561814, "count": 482 }, "SoccerTwos.Losses.ValueLoss.mean": { "value": 0.10804408763845762, "min": 0.0008606591868253115, "max": 0.1229787160952886, "count": 482 }, "SoccerTwos.Losses.ValueLoss.sum": { "value": 0.10804408763845762, "min": 0.0008606591868253115, "max": 0.1229787160952886, "count": 482 }, "SoccerTwos.Losses.BaselineLoss.mean": { "value": 0.10806638523936271, "min": 0.0008650216366125581, "max": 0.12314595381418864, "count": 482 }, "SoccerTwos.Losses.BaselineLoss.sum": { "value": 0.10806638523936271, "min": 0.0008650216366125581, "max": 0.12314595381418864, "count": 482 }, "SoccerTwos.Policy.LearningRate.mean": { "value": 7.12599857500005e-07, "min": 7.12599857500005e-07, "max": 0.0004986521002695799, "count": 482 }, "SoccerTwos.Policy.LearningRate.sum": { "value": 7.12599857500005e-07, "min": 7.12599857500005e-07, "max": 0.0004986521002695799, "count": 482 }, "SoccerTwos.Policy.Epsilon.mean": { "value": 0.10014250000000001, "min": 0.10014250000000001, "max": 0.19973042, "count": 482 }, "SoccerTwos.Policy.Epsilon.sum": { "value": 0.10014250000000001, "min": 0.10014250000000001, "max": 0.19973042, "count": 482 }, "SoccerTwos.Policy.Beta.mean": { "value": 1.7110750000000053e-05, "min": 1.7110750000000053e-05, "max": 0.004986547957999999, "count": 482 }, "SoccerTwos.Policy.Beta.sum": { "value": 1.7110750000000053e-05, "min": 1.7110750000000053e-05, "max": 0.004986547957999999, "count": 482 } }, "metadata": { "timer_format_version": "0.1.0", "start_time_seconds": "1703855988", "python_version": "3.10.12 (main, Jul 5 2023, 15:02:25) [Clang 14.0.6 ]", "command_line_arguments": "/opt/homebrew/Caskroom/miniconda/base/envs/rl/bin/mlagents-learn ./config/poca/SoccerTwos.yaml --env=./training-envs-executables/SoccerTwos.app --run-id=SoccerTwos --no-graphics --force", "mlagents_version": "1.1.0.dev0", "mlagents_envs_version": "1.1.0.dev0", "communication_protocol_version": "1.5.0", "pytorch_version": "2.1.2", "numpy_version": "1.23.5", "end_time_seconds": "1703866208" }, "total": 10219.73770537498, "count": 1, "self": 0.19493024999974295, "children": { "run_training.setup": { "total": 0.013058999989880249, "count": 1, "self": 0.013058999989880249 }, "TrainerController.start_learning": { "total": 10219.529716124991, "count": 1, "self": 4.421711311646504, "children": { "TrainerController._reset_env": { "total": 1.822008292976534, "count": 40, "self": 1.822008292976534 }, "TrainerController.advance": { "total": 10213.13215393637, "count": 681034, "self": 3.9921176846837625, "children": { "env_step": { "total": 4570.680127619009, "count": 681034, "self": 3851.7968664654472, "children": { "SubprocessEnvManager._take_step": { "total": 715.9787979857938, "count": 681034, "self": 21.6306824052881, "children": { "TorchPolicy.evaluate": { "total": 694.3481155805057, "count": 1257830, "self": 694.3481155805057 } } }, "workers": { "total": 2.9044631677679718, "count": 681034, "self": 0.0, "children": { "worker_root": { "total": 10211.220314579754, "count": 681034, "is_parallel": true, "self": 6963.624994570651, "children": { "steps_from_proto": { "total": 0.025804257107665762, "count": 80, "is_parallel": true, "self": 0.00424826008384116, "children": { "_process_rank_one_or_two_observation": { "total": 0.021555997023824602, "count": 320, "is_parallel": true, "self": 0.021555997023824602 } } }, "UnityEnvironment.step": { "total": 3247.569515751995, "count": 681034, "is_parallel": true, "self": 157.49633839310263, "children": { "UnityEnvironment._generate_step_input": { "total": 75.07220318031614, "count": 681034, "is_parallel": true, "self": 75.07220318031614 }, "communicator.exchange": { "total": 2617.271251162514, "count": 681034, "is_parallel": true, "self": 2617.271251162514 }, "steps_from_proto": { "total": 397.72972301606205, "count": 1362068, "is_parallel": true, "self": 61.088766790606314, "children": { "_process_rank_one_or_two_observation": { "total": 336.64095622545574, "count": 5448272, "is_parallel": true, "self": 336.64095622545574 } } } } } } } } } } }, "trainer_advance": { "total": 5638.459908632678, "count": 681034, "self": 41.62850201164838, "children": { "process_trajectory": { "total": 1114.3295763299102, "count": 681034, "self": 1111.2247817909229, "children": { "RLTrainer._checkpoint": { "total": 3.104794538987335, "count": 20, "self": 3.104794538987335 } } }, "_update_policy": { "total": 4482.501830291119, "count": 482, "self": 476.7318739280745, "children": { "TorchPOCAOptimizer.update": { "total": 4005.769956363045, "count": 14469, "self": 4005.769956363045 } } } } } } }, "trainer_threads": { "total": 2.500019036233425e-07, "count": 1, "self": 2.500019036233425e-07 }, "TrainerController._save_models": { "total": 0.15384233399527147, "count": 1, "self": 0.0008283340139314532, "children": { "RLTrainer._checkpoint": { "total": 0.15301399998134002, "count": 1, "self": 0.15301399998134002 } } } } } } }