{ "name": "root", "gauges": { "SoccerTwos.Policy.Entropy.mean": { "value": 1.5979747772216797, "min": 1.511555552482605, "max": 3.295804500579834, "count": 2500 }, "SoccerTwos.Policy.Entropy.sum": { "value": 31141.33203125, "min": 3777.708740234375, "max": 123682.1796875, "count": 2500 }, "SoccerTwos.Environment.EpisodeLength.mean": { "value": 71.68115942028986, "min": 42.14782608695652, "max": 999.0, "count": 2500 }, "SoccerTwos.Environment.EpisodeLength.sum": { "value": 19784.0, "min": 14492.0, "max": 25680.0, "count": 2500 }, "SoccerTwos.Self-play.ELO.mean": { "value": 1826.957066560597, "min": 1197.322794256716, "max": 1843.2668183370704, "count": 2467 }, "SoccerTwos.Self-play.ELO.sum": { "value": 252120.0751853624, "min": 2394.645588513432, "max": 404495.6598818271, "count": 2467 }, "SoccerTwos.Step.mean": { "value": 24999998.0, "min": 9856.0, "max": 24999998.0, "count": 2500 }, "SoccerTwos.Step.sum": { "value": 24999998.0, "min": 9856.0, "max": 24999998.0, "count": 2500 }, "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": { "value": -0.011240914463996887, "min": -0.14725421369075775, "max": 0.33110320568084717, "count": 2500 }, "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": { "value": -1.5400053262710571, "min": -23.897327423095703, "max": 49.003273010253906, "count": 2500 }, "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": { "value": -0.011732472106814384, "min": -0.14846444129943848, "max": 0.33885422348976135, "count": 2500 }, "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": { "value": -1.6073486804962158, "min": -23.647857666015625, "max": 50.15042495727539, "count": 2500 }, "SoccerTwos.Environment.CumulativeReward.mean": { "value": 0.0, "min": 0.0, "max": 0.0, "count": 2500 }, "SoccerTwos.Environment.CumulativeReward.sum": { "value": 0.0, "min": 0.0, "max": 0.0, "count": 2500 }, "SoccerTwos.Policy.ExtrinsicReward.mean": { "value": -0.12945985533025142, "min": -0.49676666657129925, "max": 0.7677444407233486, "count": 2500 }, "SoccerTwos.Policy.ExtrinsicReward.sum": { "value": -17.736000180244446, "min": -62.46600019931793, "max": 99.18400037288666, "count": 2500 }, "SoccerTwos.Environment.GroupCumulativeReward.mean": { "value": -0.12945985533025142, "min": -0.49676666657129925, "max": 0.7677444407233486, "count": 2500 }, "SoccerTwos.Environment.GroupCumulativeReward.sum": { "value": -17.736000180244446, "min": -62.46600019931793, "max": 99.18400037288666, "count": 2500 }, "SoccerTwos.IsTraining.mean": { "value": 1.0, "min": 1.0, "max": 1.0, "count": 2500 }, "SoccerTwos.IsTraining.sum": { "value": 1.0, "min": 1.0, "max": 1.0, "count": 2500 }, "SoccerTwos.Losses.PolicyLoss.mean": { "value": 0.01654374592277842, "min": 0.008984572056215256, "max": 0.02581824016136428, "count": 1210 }, "SoccerTwos.Losses.PolicyLoss.sum": { "value": 0.01654374592277842, "min": 0.008984572056215256, "max": 0.02581824016136428, "count": 1210 }, "SoccerTwos.Losses.ValueLoss.mean": { "value": 0.08911005382736524, "min": 3.785484807394823e-07, "max": 0.14597028146187466, "count": 1210 }, "SoccerTwos.Losses.ValueLoss.sum": { "value": 0.08911005382736524, "min": 3.785484807394823e-07, "max": 0.14597028146187466, "count": 1210 }, "SoccerTwos.Losses.BaselineLoss.mean": { "value": 0.09072775915265083, "min": 3.7238030851464526e-07, "max": 0.14897385289271672, "count": 1210 }, "SoccerTwos.Losses.BaselineLoss.sum": { "value": 0.09072775915265083, "min": 3.7238030851464526e-07, "max": 0.14897385289271672, "count": 1210 }, "SoccerTwos.Policy.LearningRate.mean": { "value": 0.00025000000000000006, "min": 0.00025000000000000006, "max": 0.00025000000000000006, "count": 1210 }, "SoccerTwos.Policy.LearningRate.sum": { "value": 0.00025000000000000006, "min": 0.00025000000000000006, "max": 0.00025000000000000006, "count": 1210 }, "SoccerTwos.Policy.Epsilon.mean": { "value": 0.20000000000000007, "min": 0.20000000000000004, "max": 0.20000000000000007, "count": 1210 }, "SoccerTwos.Policy.Epsilon.sum": { "value": 0.20000000000000007, "min": 0.20000000000000004, "max": 0.20000000000000007, "count": 1210 }, "SoccerTwos.Policy.Beta.mean": { "value": 0.005000000000000001, "min": 0.005000000000000001, "max": 0.005000000000000001, "count": 1210 }, "SoccerTwos.Policy.Beta.sum": { "value": 0.005000000000000001, "min": 0.005000000000000001, "max": 0.005000000000000001, "count": 1210 } }, "metadata": { "timer_format_version": "0.1.0", "start_time_seconds": "1714197161", "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\\Rahil\\miniconda3\\envs\\huggingface_rl\\Scripts\\mlagents-learn ./config/poca/SoccerTwos.yaml --env=./training-envs-executables/SoccerTwos.exe --run-id=SoccerTwos --no-graphics", "mlagents_version": "1.1.0.dev0", "mlagents_envs_version": "1.1.0.dev0", "communication_protocol_version": "1.5.0", "pytorch_version": "2.3.0+cu121", "numpy_version": "1.23.5", "end_time_seconds": "1714242608" }, "total": 45445.911744100005, "count": 1, "self": 0.8529118000078597, "children": { "run_training.setup": { "total": 0.09120809999876656, "count": 1, "self": 0.09120809999876656 }, "TrainerController.start_learning": { "total": 45444.9676242, "count": 1, "self": 32.451057508427766, "children": { "TrainerController._reset_env": { "total": 9.359218000085093, "count": 125, "self": 9.359218000085093 }, "TrainerController.advance": { "total": 45402.9393858915, "count": 1712458, "self": 30.118018284418213, "children": { "env_step": { "total": 32565.312563302898, "count": 1712458, "self": 19509.227607912057, "children": { "SubprocessEnvManager._take_step": { "total": 13034.285687099866, "count": 1712458, "self": 201.94590439824242, "children": { "TorchPolicy.evaluate": { "total": 12832.339782701623, "count": 3143974, "self": 12832.339782701623 } } }, "workers": { "total": 21.799268290975306, "count": 1712458, "self": 0.0, "children": { "worker_root": { "total": 45398.33970200704, "count": 1712458, "is_parallel": true, "self": 29911.087757495094, "children": { "steps_from_proto": { "total": 0.2044884001079481, "count": 250, "is_parallel": true, "self": 0.03695939997851383, "children": { "_process_rank_one_or_two_observation": { "total": 0.16752900012943428, "count": 1000, "is_parallel": true, "self": 0.16752900012943428 } } }, "UnityEnvironment.step": { "total": 15487.047456111839, "count": 1712458, "is_parallel": true, "self": 792.354284125744, "children": { "UnityEnvironment._generate_step_input": { "total": 694.8910819937155, "count": 1712458, "is_parallel": true, "self": 694.8910819937155 }, "communicator.exchange": { "total": 11344.200517301026, "count": 1712458, "is_parallel": true, "self": 11344.200517301026 }, "steps_from_proto": { "total": 2655.601572691354, "count": 3424916, "is_parallel": true, "self": 487.3085465980548, "children": { "_process_rank_one_or_two_observation": { "total": 2168.293026093299, "count": 13699664, "is_parallel": true, "self": 2168.293026093299 } } } } } } } } } } }, "trainer_advance": { "total": 12807.508804304183, "count": 1712458, "self": 246.52064090550266, "children": { "process_trajectory": { "total": 6219.173412498698, "count": 1712458, "self": 6209.509433798732, "children": { "RLTrainer._checkpoint": { "total": 9.66397869996581, "count": 50, "self": 9.66397869996581 } } }, "_update_policy": { "total": 6341.814750899983, "count": 1210, "self": 4088.667170400142, "children": { "TorchPOCAOptimizer.update": { "total": 2253.147580499841, "count": 36309, "self": 2253.147580499841 } } } } } } }, "trainer_threads": { "total": 8.999923011288047e-07, "count": 1, "self": 8.999923011288047e-07 }, "TrainerController._save_models": { "total": 0.21796189999440685, "count": 1, "self": 0.04979129999992438, "children": { "RLTrainer._checkpoint": { "total": 0.16817059999448247, "count": 1, "self": 0.16817059999448247 } } } } } } }