{ "name": "root", "gauges": { "SoccerTwos.Policy.Entropy.mean": { "value": 1.3606890439987183, "min": 1.3207616806030273, "max": 3.2958154678344727, "count": 5000 }, "SoccerTwos.Policy.Entropy.sum": { "value": 27997.537109375, "min": 7348.11328125, "max": 167852.40625, "count": 5000 }, "SoccerTwos.Environment.EpisodeLength.mean": { "value": 64.92207792207792, "min": 41.31858407079646, "max": 999.0, "count": 5000 }, "SoccerTwos.Environment.EpisodeLength.sum": { "value": 19996.0, "min": 15984.0, "max": 26616.0, "count": 5000 }, "SoccerTwos.Self-play.ELO.mean": { "value": 1944.2369976366379, "min": 1186.2772326853458, "max": 1956.3347176419152, "count": 4794 }, "SoccerTwos.Self-play.ELO.sum": { "value": 299412.49763604224, "min": 2372.5625412596405, "max": 425002.35477958986, "count": 4794 }, "SoccerTwos.Step.mean": { "value": 49999862.0, "min": 9326.0, "max": 49999862.0, "count": 5000 }, "SoccerTwos.Step.sum": { "value": 49999862.0, "min": 9326.0, "max": 49999862.0, "count": 5000 }, "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": { "value": -0.003132409183308482, "min": -0.10575120896100998, "max": 0.3213760256767273, "count": 5000 }, "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": { "value": -0.47925859689712524, "min": -20.159502029418945, "max": 44.83538055419922, "count": 5000 }, "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": { "value": -0.0028535863384604454, "min": -0.10418880730867386, "max": 0.32389914989471436, "count": 5000 }, "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": { "value": -0.4365987181663513, "min": -21.083391189575195, "max": 44.63798141479492, "count": 5000 }, "SoccerTwos.Environment.CumulativeReward.mean": { "value": 0.0, "min": 0.0, "max": 0.0, "count": 5000 }, "SoccerTwos.Environment.CumulativeReward.sum": { "value": 0.0, "min": 0.0, "max": 0.0, "count": 5000 }, "SoccerTwos.Policy.ExtrinsicReward.mean": { "value": 0.10041568676630656, "min": -0.5714285714285714, "max": 0.8144077510796777, "count": 5000 }, "SoccerTwos.Policy.ExtrinsicReward.sum": { "value": 15.363600075244904, "min": -62.99719989299774, "max": 105.05859988927841, "count": 5000 }, "SoccerTwos.Environment.GroupCumulativeReward.mean": { "value": 0.10041568676630656, "min": -0.5714285714285714, "max": 0.8144077510796777, "count": 5000 }, "SoccerTwos.Environment.GroupCumulativeReward.sum": { "value": 15.363600075244904, "min": -62.99719989299774, "max": 105.05859988927841, "count": 5000 }, "SoccerTwos.IsTraining.mean": { "value": 1.0, "min": 1.0, "max": 1.0, "count": 5000 }, "SoccerTwos.IsTraining.sum": { "value": 1.0, "min": 1.0, "max": 1.0, "count": 5000 }, "SoccerTwos.Losses.PolicyLoss.mean": { "value": 0.013992966905546685, "min": 0.009055629027837615, "max": 0.026194270861257488, "count": 2408 }, "SoccerTwos.Losses.PolicyLoss.sum": { "value": 0.013992966905546685, "min": 0.009055629027837615, "max": 0.026194270861257488, "count": 2408 }, "SoccerTwos.Losses.ValueLoss.mean": { "value": 0.08764854321877162, "min": 1.2683622570008878e-07, "max": 0.10234996452927589, "count": 2408 }, "SoccerTwos.Losses.ValueLoss.sum": { "value": 0.08764854321877162, "min": 1.2683622570008878e-07, "max": 0.10234996452927589, "count": 2408 }, "SoccerTwos.Losses.BaselineLoss.mean": { "value": 0.08808425143361091, "min": 1.2518245879888922e-07, "max": 0.10266090234120687, "count": 2408 }, "SoccerTwos.Losses.BaselineLoss.sum": { "value": 0.08808425143361091, "min": 1.2518245879888922e-07, "max": 0.10266090234120687, "count": 2408 }, "SoccerTwos.Policy.LearningRate.mean": { "value": 4.656996902000402e-08, "min": 4.656996902000402e-08, "max": 0.000149936022042652, "count": 2408 }, "SoccerTwos.Policy.LearningRate.sum": { "value": 4.656996902000402e-08, "min": 4.656996902000402e-08, "max": 0.000149936022042652, "count": 2408 }, "SoccerTwos.Policy.Epsilon.mean": { "value": 0.10005266599999998, "min": 0.10005266599999998, "max": 0.26992749159999996, "count": 2408 }, "SoccerTwos.Policy.Epsilon.sum": { "value": 0.10005266599999998, "min": 0.10005266599999998, "max": 0.26992749159999996, "count": 2408 }, "SoccerTwos.Policy.Beta.mean": { "value": 1.3094902000000267e-05, "min": 1.3094902000000267e-05, "max": 0.0099957390652, "count": 2408 }, "SoccerTwos.Policy.Beta.sum": { "value": 1.3094902000000267e-05, "min": 1.3094902000000267e-05, "max": 0.0099957390652, "count": 2408 } }, "metadata": { "timer_format_version": "0.1.0", "start_time_seconds": "1717231552", "python_version": "3.10.12 (main, Jul 5 2023, 15:02:25) [Clang 14.0.6 ]", "command_line_arguments": "/Users/prithvi/anaconda3/envs/rl/bin/mlagents-learn ./config/poca/SoccerTwos.yaml --env=./training-envs-executables/SoccerTwos/SoccerTwos.app --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", "numpy_version": "1.23.5", "end_time_seconds": "1717459891" }, "total": 228338.0476951669, "count": 1, "self": 0.5725247919326648, "children": { "run_training.setup": { "total": 0.024256707984022796, "count": 1, "self": 0.024256707984022796 }, "TrainerController.start_learning": { "total": 228337.45091366698, "count": 1, "self": 42.539694306091405, "children": { "TrainerController._reset_env": { "total": 19.4115347933257, "count": 250, "self": 19.4115347933257 }, "TrainerController.advance": { "total": 228275.35854177666, "count": 3423208, "self": 39.36356943077408, "children": { "env_step": { "total": 180749.49001260777, "count": 3423208, "self": 174422.59828599403, "children": { "SubprocessEnvManager._take_step": { "total": 6301.572822949383, "count": 3423208, "self": 194.92717085045297, "children": { "TorchPolicy.evaluate": { "total": 6106.64565209893, "count": 6310568, "self": 6106.64565209893 } } }, "workers": { "total": 25.318903664359823, "count": 3423208, "self": 0.0, "children": { "worker_root": { "total": 228269.80472033995, "count": 3423208, "is_parallel": true, "self": 58624.286830272176, "children": { "steps_from_proto": { "total": 0.5045759890927002, "count": 500, "is_parallel": true, "self": 0.05740836798213422, "children": { "_process_rank_one_or_two_observation": { "total": 0.44716762111056596, "count": 2000, "is_parallel": true, "self": 0.44716762111056596 } } }, "UnityEnvironment.step": { "total": 169645.01331407868, "count": 3423208, "is_parallel": true, "self": 456.1211591894971, "children": { "UnityEnvironment._generate_step_input": { "total": 3106.9896800647257, "count": 3423208, "is_parallel": true, "self": 3106.9896800647257 }, "communicator.exchange": { "total": 159872.00061112107, "count": 3423208, "is_parallel": true, "self": 159872.00061112107 }, "steps_from_proto": { "total": 6209.901863703388, "count": 6846416, "is_parallel": true, "self": 679.2257801635424, "children": { "_process_rank_one_or_two_observation": { "total": 5530.676083539845, "count": 27385664, "is_parallel": true, "self": 5530.676083539845 } } } } } } } } } } }, "trainer_advance": { "total": 47486.504959738115, "count": 3423208, "self": 353.96039831102826, "children": { "process_trajectory": { "total": 7820.259917136049, "count": 3423208, "self": 7791.687315305346, "children": { "RLTrainer._checkpoint": { "total": 28.572601830703206, "count": 100, "self": 28.572601830703206 } } }, "_update_policy": { "total": 39312.28464429104, "count": 2408, "self": 3858.2221490669763, "children": { "TorchPOCAOptimizer.update": { "total": 35454.06249522406, "count": 72240, "self": 35454.06249522406 } } } } } } }, "trainer_threads": { "total": 4.5797787606716156e-07, "count": 1, "self": 4.5797787606716156e-07 }, "TrainerController._save_models": { "total": 0.14114233292639256, "count": 1, "self": 0.0030910419300198555, "children": { "RLTrainer._checkpoint": { "total": 0.1380512909963727, "count": 1, "self": 0.1380512909963727 } } } } } } }