{ "name": "root", "gauges": { "SoccerTwos.Policy.Entropy.mean": { "value": 1.11602783203125, "min": 1.1030526161193848, "max": 1.4285537004470825, "count": 500 }, "SoccerTwos.Policy.Entropy.sum": { "value": 22320.556640625, "min": 21029.970703125, "max": 28662.501953125, "count": 500 }, "SoccerTwos.Environment.EpisodeLength.mean": { "value": 45.905660377358494, "min": 35.28, "max": 55.38202247191011, "count": 500 }, "SoccerTwos.Environment.EpisodeLength.sum": { "value": 19464.0, "min": 14112.0, "max": 21312.0, "count": 500 }, "SoccerTwos.Self-play.ELO.mean": { "value": 1672.5986965358836, "min": 1594.9761687604023, "max": 1684.8696129621046, "count": 500 }, "SoccerTwos.Self-play.ELO.sum": { "value": 354590.92366560735, "min": 291606.7260394614, "max": 450906.7595913422, "count": 500 }, "SoccerTwos.Step.mean": { "value": 14999982.0, "min": 10009972.0, "max": 14999982.0, "count": 500 }, "SoccerTwos.Step.sum": { "value": 14999982.0, "min": 10009972.0, "max": 14999982.0, "count": 500 }, "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": { "value": -0.033846165984869, "min": -0.10714015364646912, "max": 0.09757562726736069, "count": 500 }, "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": { "value": -13.876928329467773, "min": -45.2131462097168, "max": 41.95751953125, "count": 500 }, "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": { "value": -0.03533671796321869, "min": -0.11142526566982269, "max": 0.10013997554779053, "count": 500 }, "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": { "value": -14.488054275512695, "min": -47.021461486816406, "max": 43.06018829345703, "count": 500 }, "SoccerTwos.Environment.CumulativeReward.mean": { "value": 0.0, "min": 0.0, "max": 0.0, "count": 500 }, "SoccerTwos.Environment.CumulativeReward.sum": { "value": 0.0, "min": 0.0, "max": 0.0, "count": 500 }, "SoccerTwos.Policy.ExtrinsicReward.mean": { "value": -0.07970754661650029, "min": -0.26102524301380786, "max": 0.3510517215934293, "count": 500 }, "SoccerTwos.Policy.ExtrinsicReward.sum": { "value": -16.89799988269806, "min": -55.66319966316223, "max": 81.4439994096756, "count": 500 }, "SoccerTwos.Environment.GroupCumulativeReward.mean": { "value": -0.07970754661650029, "min": -0.26102524301380786, "max": 0.3510517215934293, "count": 500 }, "SoccerTwos.Environment.GroupCumulativeReward.sum": { "value": -16.89799988269806, "min": -55.66319966316223, "max": 81.4439994096756, "count": 500 }, "SoccerTwos.IsTraining.mean": { "value": 1.0, "min": 1.0, "max": 1.0, "count": 500 }, "SoccerTwos.IsTraining.sum": { "value": 1.0, "min": 1.0, "max": 1.0, "count": 500 }, "SoccerTwos.Losses.PolicyLoss.mean": { "value": 0.012091718217061499, "min": 0.009822013491066173, "max": 0.015032402960059699, "count": 76 }, "SoccerTwos.Losses.PolicyLoss.sum": { "value": 0.012091718217061499, "min": 0.009822013491066173, "max": 0.015032402960059699, "count": 76 }, "SoccerTwos.Losses.ValueLoss.mean": { "value": 0.05806648256257176, "min": 0.05226881531998515, "max": 0.059615218080580235, "count": 76 }, "SoccerTwos.Losses.ValueLoss.sum": { "value": 0.05806648256257176, "min": 0.05226881531998515, "max": 0.059615218080580235, "count": 76 }, "SoccerTwos.Losses.BaselineLoss.mean": { "value": 0.05836009904742241, "min": 0.0542876117862761, "max": 0.06110596880316734, "count": 76 }, "SoccerTwos.Losses.BaselineLoss.sum": { "value": 0.05836009904742241, "min": 0.0542876117862761, "max": 0.06110596880316734, "count": 76 }, "SoccerTwos.Policy.LearningRate.mean": { "value": 4.0553810392494676e-07, "min": 4.0553810392494676e-07, "max": 0.0001428860023203168, "count": 76 }, "SoccerTwos.Policy.LearningRate.sum": { "value": 4.0553810392494676e-07, "min": 4.0553810392494676e-07, "max": 0.0001428860023203168, "count": 76 }, "SoccerTwos.Policy.Epsilon.mean": { "value": 0.10005440800688417, "min": 0.10005440800688417, "max": 0.11917465842958205, "count": 76 }, "SoccerTwos.Policy.Epsilon.sum": { "value": 0.10005440800688417, "min": 0.10005440800688417, "max": 0.11917465842958205, "count": 76 }, "SoccerTwos.Policy.Beta.mean": { "value": 1.0169934717551432e-05, "min": 1.0169934717551432e-05, "max": 6.988898235716646e-05, "count": 76 }, "SoccerTwos.Policy.Beta.sum": { "value": 1.0169934717551432e-05, "min": 1.0169934717551432e-05, "max": 6.988898235716646e-05, "count": 76 } }, "metadata": { "timer_format_version": "0.1.0", "start_time_seconds": "1710358573", "python_version": "3.10.12 (main, Jul 5 2023, 18:54:27) [GCC 11.2.0]", "command_line_arguments": "/home/stephan/.conda/envs/rl/bin/mlagents-learn ./ml-agents/config/poca/SoccerTwos.yaml --env=./ml-agents/training-envs-executables/SoccerTwos.exe --run-id=SoccerTwos-read-deal --width=1280 --height=720 --resume", "mlagents_version": "1.1.0.dev0", "mlagents_envs_version": "1.1.0.dev0", "communication_protocol_version": "1.5.0", "pytorch_version": "2.2.1+cu121", "numpy_version": "1.23.5", "end_time_seconds": "1710372965" }, "total": 14392.268066103992, "count": 1, "self": 0.3203680879960302, "children": { "run_training.setup": { "total": 0.017996288996073417, "count": 1, "self": 0.017996288996073417 }, "TrainerController.start_learning": { "total": 14391.929701727, "count": 1, "self": 7.222826109427842, "children": { "TrainerController._reset_env": { "total": 4.3356985290010925, "count": 10, "self": 4.3356985290010925 }, "TrainerController.advance": { "total": 14380.252779454575, "count": 357131, "self": 6.432589866788476, "children": { "env_step": { "total": 6466.631959918537, "count": 357131, "self": 5498.500712261317, "children": { "SubprocessEnvManager._take_step": { "total": 963.7606686544896, "count": 357131, "self": 29.2971219017345, "children": { "TorchPolicy.evaluate": { "total": 934.4635467527551, "count": 624880, "self": 934.4635467527551 } } }, "workers": { "total": 4.370579002730665, "count": 357131, "self": 0.0, "children": { "worker_root": { "total": 14380.838316511697, "count": 357131, "is_parallel": true, "self": 9673.599672775075, "children": { "run_training.setup": { "total": 0.0, "count": 0, "is_parallel": true, "self": 0.0, "children": { "steps_from_proto": { "total": 0.06619343400234357, "count": 2, "is_parallel": true, "self": 0.0004880309716099873, "children": { "_process_rank_one_or_two_observation": { "total": 0.06570540303073358, "count": 8, "is_parallel": true, "self": 0.06570540303073358 } } }, "UnityEnvironment.step": { "total": 0.027265490003628656, "count": 1, "is_parallel": true, "self": 0.0006784550059819594, "children": { "UnityEnvironment._generate_step_input": { "total": 0.0004930220020469278, "count": 1, "is_parallel": true, "self": 0.0004930220020469278 }, "communicator.exchange": { "total": 0.024494213997968473, "count": 1, "is_parallel": true, "self": 0.024494213997968473 }, "steps_from_proto": { "total": 0.0015997989976312965, "count": 2, "is_parallel": true, "self": 0.00032821799686644226, "children": { "_process_rank_one_or_two_observation": { "total": 0.0012715810007648543, "count": 8, "is_parallel": true, "self": 0.0012715810007648543 } } } } } } }, "steps_from_proto": { "total": 0.018047201985609718, "count": 18, "is_parallel": true, "self": 0.0030916809337213635, "children": { "_process_rank_one_or_two_observation": { "total": 0.014955521051888354, "count": 72, "is_parallel": true, "self": 0.014955521051888354 } } }, "UnityEnvironment.step": { "total": 4707.220596534637, "count": 357130, "is_parallel": true, "self": 257.41373912687413, "children": { "UnityEnvironment._generate_step_input": { "total": 162.74674167780904, "count": 357130, "is_parallel": true, "self": 162.74674167780904 }, "communicator.exchange": { "total": 3645.2510425291694, "count": 357130, "is_parallel": true, "self": 3645.2510425291694 }, "steps_from_proto": { "total": 641.809073200784, "count": 714260, "is_parallel": true, "self": 114.93468356617086, "children": { "_process_rank_one_or_two_observation": { "total": 526.8743896346132, "count": 2857040, "is_parallel": true, "self": 526.8743896346132 } } } } } } } } } } }, "trainer_advance": { "total": 7907.188229669249, "count": 357131, "self": 41.56422363645106, "children": { "process_trajectory": { "total": 1667.4114032577781, "count": 357131, "self": 1666.2047581057705, "children": { "RLTrainer._checkpoint": { "total": 1.2066451520076953, "count": 10, "self": 1.2066451520076953 } } }, "_update_policy": { "total": 6198.21260277502, "count": 76, "self": 967.3907746167824, "children": { "TorchPOCAOptimizer.update": { "total": 5230.821828158238, "count": 6080, "self": 5230.821828158238 } } } } } } }, "trainer_threads": { "total": 7.199996616691351e-07, "count": 1, "self": 7.199996616691351e-07 }, "TrainerController._save_models": { "total": 0.11839691399654839, "count": 1, "self": 0.0024417589884251356, "children": { "RLTrainer._checkpoint": { "total": 0.11595515500812326, "count": 1, "self": 0.11595515500812326 } } } } } } }