{ "name": "root", "gauges": { "SoccerTwos.Policy.Entropy.mean": { "value": 1.6892871856689453, "min": 1.6637274026870728, "max": 3.2958061695098877, "count": 2504 }, "SoccerTwos.Policy.Entropy.sum": { "value": 18055.1015625, "min": 7680.09765625, "max": 105465.796875, "count": 2504 }, "SoccerTwos.Environment.EpisodeLength.mean": { "value": 73.37142857142857, "min": 41.73684210526316, "max": 999.0, "count": 2504 }, "SoccerTwos.Environment.EpisodeLength.sum": { "value": 10272.0, "min": 6556.0, "max": 24920.0, "count": 2504 }, "SoccerTwos.Self-play.ELO.mean": { "value": 1563.2672605397256, "min": 1197.6947597061876, "max": 1584.7162481858609, "count": 2485 }, "SoccerTwos.Self-play.ELO.sum": { "value": 109428.7082377808, "min": 2396.068531600837, "max": 180657.65229318815, "count": 2485 }, "SoccerTwos.Step.mean": { "value": 12524938.0, "min": 4476.0, "max": 12524938.0, "count": 2505 }, "SoccerTwos.Step.sum": { "value": 12524938.0, "min": 4476.0, "max": 12524938.0, "count": 2505 }, "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": { "value": -0.012251640670001507, "min": -0.1846155971288681, "max": 0.152823343873024, "count": 2505 }, "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": { "value": -0.8576148748397827, "min": -13.990399360656738, "max": 12.732059478759766, "count": 2505 }, "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": { "value": -0.008902576752007008, "min": -0.18532951176166534, "max": 0.1499340832233429, "count": 2505 }, "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": { "value": -0.6231803894042969, "min": -14.163141250610352, "max": 12.865997314453125, "count": 2505 }, "SoccerTwos.Environment.CumulativeReward.mean": { "value": 0.0, "min": 0.0, "max": 0.0, "count": 2505 }, "SoccerTwos.Environment.CumulativeReward.sum": { "value": 0.0, "min": 0.0, "max": 0.0, "count": 2505 }, "SoccerTwos.Policy.ExtrinsicReward.mean": { "value": -0.06346285683768137, "min": -0.8181818181818182, "max": 0.6240166574716568, "count": 2505 }, "SoccerTwos.Policy.ExtrinsicReward.sum": { "value": -4.442399978637695, "min": -42.39199995994568, "max": 37.171600103378296, "count": 2505 }, "SoccerTwos.Environment.GroupCumulativeReward.mean": { "value": -0.06346285683768137, "min": -0.8181818181818182, "max": 0.6240166574716568, "count": 2505 }, "SoccerTwos.Environment.GroupCumulativeReward.sum": { "value": -4.442399978637695, "min": -42.39199995994568, "max": 37.171600103378296, "count": 2505 }, "SoccerTwos.IsTraining.mean": { "value": 1.0, "min": 1.0, "max": 1.0, "count": 2505 }, "SoccerTwos.IsTraining.sum": { "value": 1.0, "min": 1.0, "max": 1.0, "count": 2505 }, "SoccerTwos.Losses.PolicyLoss.mean": { "value": 0.027965155991397397, "min": 0.022734053095931208, "max": 0.03164981808044508, "count": 249 }, "SoccerTwos.Losses.PolicyLoss.sum": { "value": 0.027965155991397397, "min": 0.022734053095931208, "max": 0.03164981808044508, "count": 249 }, "SoccerTwos.Losses.ValueLoss.mean": { "value": 0.052076314860540956, "min": 0.0007150877383636583, "max": 0.061866030692019396, "count": 249 }, "SoccerTwos.Losses.ValueLoss.sum": { "value": 0.052076314860540956, "min": 0.0007150877383636583, "max": 0.061866030692019396, "count": 249 }, "SoccerTwos.Losses.BaselineLoss.mean": { "value": 0.054013946714500584, "min": 0.0007275409075779188, "max": 0.06516992378359039, "count": 249 }, "SoccerTwos.Losses.BaselineLoss.sum": { "value": 0.054013946714500584, "min": 0.0007275409075779188, "max": 0.06516992378359039, "count": 249 }, "SoccerTwos.Policy.LearningRate.mean": { "value": 0.000225084528971832, "min": 0.000225084528971832, "max": 0.000299688888103704, "count": 249 }, "SoccerTwos.Policy.LearningRate.sum": { "value": 0.000225084528971832, "min": 0.000225084528971832, "max": 0.000299688888103704, "count": 249 }, "SoccerTwos.Policy.Epsilon.mean": { "value": 0.25005633600000005, "min": 0.25005633600000005, "max": 0.29979259199999997, "count": 249 }, "SoccerTwos.Policy.Epsilon.sum": { "value": 0.25005633600000005, "min": 0.25005633600000005, "max": 0.29979259199999997, "count": 249 }, "SoccerTwos.Policy.Beta.mean": { "value": 0.007505313983200001, "min": 0.007505313983200001, "max": 0.009989639970400001, "count": 249 }, "SoccerTwos.Policy.Beta.sum": { "value": 0.007505313983200001, "min": 0.007505313983200001, "max": 0.009989639970400001, "count": 249 } }, "metadata": { "timer_format_version": "0.1.0", "start_time_seconds": "1711360012", "python_version": "3.10.12 | packaged by Anaconda, Inc. | (main, Jul 5 2023, 19:01:18) [MSC v.1916 64 bit (AMD64)]", "command_line_arguments": "\\\\?\\C:\\Users\\gael1\\anaconda3\\envs\\rl\\Scripts\\mlagents-learn ./config/poca/SoccerTwos.yaml --env=./training-envs-executables/SoccerTwos.exe --run-id=SoccerTwos3 --no-graphics", "mlagents_version": "1.1.0.dev0", "mlagents_envs_version": "1.1.0.dev0", "communication_protocol_version": "1.5.0", "pytorch_version": "2.2.1+cpu", "numpy_version": "1.23.5", "end_time_seconds": "1711383372" }, "total": 23360.173033299972, "count": 1, "self": 0.22835280001163483, "children": { "run_training.setup": { "total": 0.08911920001264662, "count": 1, "self": 0.08911920001264662 }, "TrainerController.start_learning": { "total": 23359.855561299948, "count": 1, "self": 19.48167141363956, "children": { "TrainerController._reset_env": { "total": 7.970532500185072, "count": 126, "self": 7.970532500185072 }, "TrainerController.advance": { "total": 23332.289733886137, "count": 860657, "self": 17.791757446713746, "children": { "env_step": { "total": 12407.574141020654, "count": 860657, "self": 9478.773316386738, "children": { "SubprocessEnvManager._take_step": { "total": 2916.604442927055, "count": 860657, "self": 107.05926301924046, "children": { "TorchPolicy.evaluate": { "total": 2809.5451799078146, "count": 1583938, "self": 2809.5451799078146 } } }, "workers": { "total": 12.196381706860848, "count": 860656, "self": 0.0, "children": { "worker_root": { "total": 23327.420373377157, "count": 860656, "is_parallel": true, "self": 16007.075391619117, "children": { "steps_from_proto": { "total": 0.18477399949915707, "count": 252, "is_parallel": true, "self": 0.037741698790341616, "children": { "_process_rank_one_or_two_observation": { "total": 0.14703230070881546, "count": 1008, "is_parallel": true, "self": 0.14703230070881546 } } }, "UnityEnvironment.step": { "total": 7320.160207758541, "count": 860656, "is_parallel": true, "self": 373.96662447741255, "children": { "UnityEnvironment._generate_step_input": { "total": 313.44459212664515, "count": 860656, "is_parallel": true, "self": 313.44459212664515 }, "communicator.exchange": { "total": 5443.519387620385, "count": 860656, "is_parallel": true, "self": 5443.519387620385 }, "steps_from_proto": { "total": 1189.2296035340987, "count": 1721312, "is_parallel": true, "self": 241.04875135957263, "children": { "_process_rank_one_or_two_observation": { "total": 948.1808521745261, "count": 6885248, "is_parallel": true, "self": 948.1808521745261 } } } } } } } } } } }, "trainer_advance": { "total": 10906.92383541877, "count": 860656, "self": 167.86869756120723, "children": { "process_trajectory": { "total": 2661.9871018583653, "count": 860656, "self": 2659.399978758651, "children": { "RLTrainer._checkpoint": { "total": 2.5871230997145176, "count": 25, "self": 2.5871230997145176 } } }, "_update_policy": { "total": 8077.068035999197, "count": 249, "self": 1635.0027775001945, "children": { "TorchPOCAOptimizer.update": { "total": 6442.065258499002, "count": 36048, "self": 6442.065258499002 } } } } } } }, "trainer_threads": { "total": 1.00000761449337e-06, "count": 1, "self": 1.00000761449337e-06 }, "TrainerController._save_models": { "total": 0.11362249997910112, "count": 1, "self": 0.008432799950242043, "children": { "RLTrainer._checkpoint": { "total": 0.10518970002885908, "count": 1, "self": 0.10518970002885908 } } } } } } }