{ "name": "root", "gauges": { "SoccerTwos.Policy.Entropy.mean": { "value": 1.2096617221832275, "min": 0.9391449689865112, "max": 2.010424852371216, "count": 20000 }, "SoccerTwos.Policy.Entropy.sum": { "value": 21406.173828125, "min": 14272.4765625, "max": 68390.7265625, "count": 20000 }, "SoccerTwos.Environment.EpisodeLength.mean": { "value": 66.15068493150685, "min": 36.744360902255636, "max": 115.44186046511628, "count": 20000 }, "SoccerTwos.Environment.EpisodeLength.sum": { "value": 19316.0, "min": 17160.0, "max": 22160.0, "count": 20000 }, "SoccerTwos.Self-play.ELO.mean": { "value": 1216.0945673615486, "min": 978.189206095434, "max": 1262.9238630749649, "count": 20000 }, "SoccerTwos.Self-play.ELO.sum": { "value": 177549.80683478608, "min": 92538.0044447693, "max": 294780.2539102876, "count": 20000 }, "SoccerTwos.Step.mean": { "value": 199999989.0, "min": 9982.0, "max": 199999989.0, "count": 20000 }, "SoccerTwos.Step.sum": { "value": 199999989.0, "min": 9982.0, "max": 199999989.0, "count": 20000 }, "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": { "value": -0.04544258490204811, "min": -0.22158707678318024, "max": 0.2172376811504364, "count": 20000 }, "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": { "value": -6.680059909820557, "min": -47.788673400878906, "max": 31.285858154296875, "count": 20000 }, "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": { "value": -0.04707314446568489, "min": -0.2094786912202835, "max": 0.20654548704624176, "count": 20000 }, "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": { "value": -6.91975212097168, "min": -45.967620849609375, "max": 32.832611083984375, "count": 20000 }, "SoccerTwos.Environment.CumulativeReward.mean": { "value": 0.0, "min": 0.0, "max": 0.0, "count": 20000 }, "SoccerTwos.Environment.CumulativeReward.sum": { "value": 0.0, "min": 0.0, "max": 0.0, "count": 20000 }, "SoccerTwos.Policy.ExtrinsicReward.mean": { "value": -0.10335782352758914, "min": -0.5316902246690334, "max": 0.4482817514969485, "count": 20000 }, "SoccerTwos.Policy.ExtrinsicReward.sum": { "value": -15.193600058555603, "min": -70.71479988098145, "max": 61.41459995508194, "count": 20000 }, "SoccerTwos.Environment.GroupCumulativeReward.mean": { "value": -0.10335782352758914, "min": -0.5316902246690334, "max": 0.4482817514969485, "count": 20000 }, "SoccerTwos.Environment.GroupCumulativeReward.sum": { "value": -15.193600058555603, "min": -70.71479988098145, "max": 61.41459995508194, "count": 20000 }, "SoccerTwos.IsTraining.mean": { "value": 1.0, "min": 1.0, "max": 1.0, "count": 20000 }, "SoccerTwos.IsTraining.sum": { "value": 1.0, "min": 1.0, "max": 1.0, "count": 20000 }, "SoccerTwos.Losses.PolicyLoss.mean": { "value": 0.04526555237049858, "min": 0.025337440768877666, "max": 0.06725332758699855, "count": 5393 }, "SoccerTwos.Losses.PolicyLoss.sum": { "value": 0.04526555237049858, "min": 0.025337440768877666, "max": 0.06725332758699855, "count": 5393 }, "SoccerTwos.Losses.ValueLoss.mean": { "value": 0.05701572448015213, "min": 0.05284883367518584, "max": 0.09262013187011083, "count": 5393 }, "SoccerTwos.Losses.ValueLoss.sum": { "value": 0.05701572448015213, "min": 0.05284883367518584, "max": 0.09262013187011083, "count": 5393 }, "SoccerTwos.Losses.BaselineLoss.mean": { "value": 0.060283756752808886, "min": 0.05522047355771065, "max": 0.10195552408695222, "count": 5393 }, "SoccerTwos.Losses.BaselineLoss.sum": { "value": 0.060283756752808886, "min": 0.05522047355771065, "max": 0.10195552408695222, "count": 5393 }, "SoccerTwos.Policy.LearningRate.mean": { "value": 1.773099804096243e-08, "min": 1.773099804096243e-08, "max": 0.0008998363800181802, "count": 5393 }, "SoccerTwos.Policy.LearningRate.sum": { "value": 1.773099804096243e-08, "min": 1.773099804096243e-08, "max": 0.0008998363800181802, "count": 5393 }, "SoccerTwos.Policy.Epsilon.mean": { "value": 0.10000048975000005, "min": 0.10000048975000005, "max": 0.12499545500000003, "count": 5393 }, "SoccerTwos.Policy.Epsilon.sum": { "value": 0.10000048975000005, "min": 0.10000048975000005, "max": 0.12499545500000003, "count": 5393 }, "SoccerTwos.Policy.Beta.mean": { "value": 1.0097754099999798e-05, "min": 1.0097754099999798e-05, "max": 0.004999092818000001, "count": 5393 }, "SoccerTwos.Policy.Beta.sum": { "value": 1.0097754099999798e-05, "min": 1.0097754099999798e-05, "max": 0.004999092818000001, "count": 5393 } }, "metadata": { "timer_format_version": "0.1.0", "start_time_seconds": "1678044037", "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-mem_2x_09_l3_64_64_16-linear-200-ft.yaml --env=./training-envs-executables/SoccerTwos/SoccerTwos.x86_64 --run-id=SoccerTwos --no-graphics --results-dir results-mem_2x_09_l3_64_64_16-linear-200-ft --base-port 5500 --num-envs=6 --initialize-from SoccerTwos --force", "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": "1678233893" }, "total": 189855.29048984684, "count": 1, "self": 0.3207913697697222, "children": { "run_training.setup": { "total": 0.02204182604327798, "count": 1, "self": 0.02204182604327798 }, "TrainerController.start_learning": { "total": 189854.94765665103, "count": 1, "self": 181.66945680743083, "children": { "TrainerController._reset_env": { "total": 98.39931596489623, "count": 1000, "self": 98.39931596489623 }, "TrainerController.advance": { "total": 189574.73793974053, "count": 7344410, "self": 165.12496986985207, "children": { "env_step": { "total": 96603.2392786229, "count": 7344410, "self": 34313.6553129293, "children": { "SubprocessEnvManager._take_step": { "total": 62147.62801608583, "count": 14027689, "self": 2248.422056724783, "children": { "TorchPolicy.evaluate": { "total": 59899.20595936105, "count": 25727990, "self": 59899.20595936105 } } }, "workers": { "total": 141.9559496077709, "count": 7344410, "self": 0.0, "children": { "worker_root": { "total": 1138019.8113890202, "count": 14024247, "is_parallel": true, "self": 783723.7413149513, "children": { "run_training.setup": { "total": 0.0, "count": 0, "is_parallel": true, "self": 0.0, "children": { "steps_from_proto": { "total": 0.022877162788063288, "count": 12, "is_parallel": true, "self": 0.005851720925420523, "children": { "_process_rank_one_or_two_observation": { "total": 0.017025441862642765, "count": 48, "is_parallel": true, "self": 0.017025441862642765 } } }, "UnityEnvironment.step": { "total": 0.2780195097438991, "count": 6, "is_parallel": true, "self": 0.013382475357502699, "children": { "UnityEnvironment._generate_step_input": { "total": 0.006043791305273771, "count": 6, "is_parallel": true, "self": 0.006043791305273771 }, "communicator.exchange": { "total": 0.23156442120671272, "count": 6, "is_parallel": true, "self": 0.23156442120671272 }, "steps_from_proto": { "total": 0.027028821874409914, "count": 12, "is_parallel": true, "self": 0.0037005161866545677, "children": { "_process_rank_one_or_two_observation": { "total": 0.023328305687755346, "count": 48, "is_parallel": true, "self": 0.023328305687755346 } } } } } } }, "UnityEnvironment.step": { "total": 354273.18451454956, "count": 14024241, "is_parallel": true, "self": 27041.204451641068, "children": { "UnityEnvironment._generate_step_input": { "total": 16506.659348304383, "count": 14024241, "is_parallel": true, "self": 16506.659348304383 }, "communicator.exchange": { "total": 243595.4176135431, "count": 14024241, "is_parallel": true, "self": 243595.4176135431 }, "steps_from_proto": { "total": 67129.903101061, "count": 28048482, "is_parallel": true, "self": 10519.045234736055, "children": { "_process_rank_one_or_two_observation": { "total": 56610.85786632495, "count": 112193928, "is_parallel": true, "self": 56610.85786632495 } } } } }, "steps_from_proto": { "total": 22.8855595192872, "count": 11988, "is_parallel": true, "self": 4.090687413234264, "children": { "_process_rank_one_or_two_observation": { "total": 18.794872106052935, "count": 47952, "is_parallel": true, "self": 18.794872106052935 } } } } } } } } }, "trainer_advance": { "total": 92806.37369124778, "count": 7344410, "self": 2120.3723945165984, "children": { "process_trajectory": { "total": 72153.54126273142, "count": 7344410, "self": 72093.49027456343, "children": { "RLTrainer._checkpoint": { "total": 60.05098816799, "count": 400, "self": 60.05098816799 } } }, "_update_policy": { "total": 18532.460033999756, "count": 5393, "self": 5504.250082283281, "children": { "TorchPOCAOptimizer.update": { "total": 13028.209951716475, "count": 161790, "self": 13028.209951716475 } } } } } } }, "trainer_threads": { "total": 7.208436727523804e-07, "count": 1, "self": 7.208436727523804e-07 }, "TrainerController._save_models": { "total": 0.14094341732561588, "count": 1, "self": 0.00044933659955859184, "children": { "RLTrainer._checkpoint": { "total": 0.1404940807260573, "count": 1, "self": 0.1404940807260573 } } } } } } }