ppo-PyramidsRND / run_logs /timers.json
shunnaidder's picture
First Push
a97bebd
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.7915821671485901,
"min": 0.7500590085983276,
"max": 1.4331189393997192,
"count": 10
},
"Pyramids.Policy.Entropy.sum": {
"value": 23709.46875,
"min": 22645.78125,
"max": 43475.09765625,
"count": 10
},
"Pyramids.Step.mean": {
"value": 299950.0,
"min": 29952.0,
"max": 299950.0,
"count": 10
},
"Pyramids.Step.sum": {
"value": 299950.0,
"min": 29952.0,
"max": 299950.0,
"count": 10
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.028268778696656227,
"min": -0.15444013476371765,
"max": 0.028268778696656227,
"count": 10
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 6.897582054138184,
"min": -36.60231018066406,
"max": 6.897582054138184,
"count": 10
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.04551688954234123,
"min": 0.04551688954234123,
"max": 0.47999802231788635,
"count": 10
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 11.106121063232422,
"min": 11.106121063232422,
"max": 113.75952911376953,
"count": 10
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06687248608588042,
"min": 0.066063088960938,
"max": 0.0726406413123963,
"count": 10
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9362148052023259,
"min": 0.508484489186774,
"max": 0.9909463344140701,
"count": 10
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.005245659170690315,
"min": 0.00212205375107772,
"max": 0.006739944880324441,
"count": 10
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.07343922838966441,
"min": 0.026387406382250074,
"max": 0.07343922838966441,
"count": 10
},
"Pyramids.Policy.LearningRate.mean": {
"value": 1.46805951065e-05,
"min": 1.46805951065e-05,
"max": 0.0002838354339596191,
"count": 10
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.000205528331491,
"min": 0.000205528331491,
"max": 0.0027272734909089998,
"count": 10
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.1048935,
"min": 0.1048935,
"max": 0.19461180952380958,
"count": 10
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.468509,
"min": 1.362282666666667,
"max": 2.3090910000000004,
"count": 10
},
"Pyramids.Policy.Beta.mean": {
"value": 0.0004988606500000001,
"min": 0.0004988606500000001,
"max": 0.00946171977142857,
"count": 10
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0069840491000000005,
"min": 0.0069840491000000005,
"max": 0.09095819090000001,
"count": 10
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.037923283874988556,
"min": 0.037923283874988556,
"max": 0.49516552686691284,
"count": 10
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.530925989151001,
"min": 0.530925989151001,
"max": 3.466158628463745,
"count": 10
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 898.6666666666666,
"min": 868.6774193548387,
"max": 999.0,
"count": 10
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 32352.0,
"min": 15984.0,
"max": 32920.0,
"count": 10
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": -0.34360004630353713,
"min": -1.0000000521540642,
"max": -0.28847747344163155,
"count": 10
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": -12.369601666927338,
"min": -28.547001741826534,
"max": -8.942801676690578,
"count": 10
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": -0.34360004630353713,
"min": -1.0000000521540642,
"max": -0.28847747344163155,
"count": 10
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": -12.369601666927338,
"min": -28.547001741826534,
"max": -8.942801676690578,
"count": 10
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.35213732569698347,
"min": 0.35213732569698347,
"max": 10.101594628766179,
"count": 10
},
"Pyramids.Policy.RndReward.sum": {
"value": 12.676943725091405,
"min": 12.306526397820562,
"max": 161.62551406025887,
"count": 10
},
"Pyramids.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 10
},
"Pyramids.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 10
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1702474173",
"python_version": "3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0]",
"command_line_arguments": "/usr/local/bin/mlagents-learn ./config/ppo/PyramidsRND.yaml --env=./training-envs-executables/linux/Pyramids/Pyramids --run-id=Pyramids Training --no-graphics",
"mlagents_version": "1.1.0.dev0",
"mlagents_envs_version": "1.1.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "2.1.1+cu121",
"numpy_version": "1.23.5",
"end_time_seconds": "1702474818"
},
"total": 645.47309445,
"count": 1,
"self": 1.0287514819999615,
"children": {
"run_training.setup": {
"total": 0.07843891099992106,
"count": 1,
"self": 0.07843891099992106
},
"TrainerController.start_learning": {
"total": 644.3659040570001,
"count": 1,
"self": 0.4658393350116512,
"children": {
"TrainerController._reset_env": {
"total": 3.475722655000027,
"count": 1,
"self": 3.475722655000027
},
"TrainerController.advance": {
"total": 640.3104600979884,
"count": 18961,
"self": 0.48303391197828205,
"children": {
"env_step": {
"total": 443.1479580879907,
"count": 18961,
"self": 400.28245262095425,
"children": {
"SubprocessEnvManager._take_step": {
"total": 42.60271816601812,
"count": 18961,
"self": 1.518379775009862,
"children": {
"TorchPolicy.evaluate": {
"total": 41.08433839100826,
"count": 18806,
"self": 41.08433839100826
}
}
},
"workers": {
"total": 0.2627873010183066,
"count": 18961,
"self": 0.0,
"children": {
"worker_root": {
"total": 642.922447972984,
"count": 18961,
"is_parallel": true,
"self": 280.2867014579974,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0027527439999630587,
"count": 1,
"is_parallel": true,
"self": 0.0008221379996484757,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.001930606000314583,
"count": 8,
"is_parallel": true,
"self": 0.001930606000314583
}
}
},
"UnityEnvironment.step": {
"total": 0.05183443100008844,
"count": 1,
"is_parallel": true,
"self": 0.0005349120001483243,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0005191669999931037,
"count": 1,
"is_parallel": true,
"self": 0.0005191669999931037
},
"communicator.exchange": {
"total": 0.04913662800004204,
"count": 1,
"is_parallel": true,
"self": 0.04913662800004204
},
"steps_from_proto": {
"total": 0.001643723999904978,
"count": 1,
"is_parallel": true,
"self": 0.00040894099947763607,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.001234783000427342,
"count": 8,
"is_parallel": true,
"self": 0.001234783000427342
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 362.6357465149865,
"count": 18960,
"is_parallel": true,
"self": 10.551039080983628,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 7.892912841002499,
"count": 18960,
"is_parallel": true,
"self": 7.892912841002499
},
"communicator.exchange": {
"total": 312.67527769898925,
"count": 18960,
"is_parallel": true,
"self": 312.67527769898925
},
"steps_from_proto": {
"total": 31.516516894011147,
"count": 18960,
"is_parallel": true,
"self": 6.36816530697638,
"children": {
"_process_rank_one_or_two_observation": {
"total": 25.148351587034767,
"count": 151680,
"is_parallel": true,
"self": 25.148351587034767
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 196.67946809801947,
"count": 18961,
"self": 0.7725336560274627,
"children": {
"process_trajectory": {
"total": 39.28591321399267,
"count": 18961,
"self": 39.28591321399267
},
"_update_policy": {
"total": 156.62102122799934,
"count": 126,
"self": 94.26334410200889,
"children": {
"TorchPPOOptimizer.update": {
"total": 62.35767712599045,
"count": 6843,
"self": 62.35767712599045
}
}
}
}
}
}
},
"trainer_threads": {
"total": 9.210000371240312e-07,
"count": 1,
"self": 9.210000371240312e-07
},
"TrainerController._save_models": {
"total": 0.11388104799993926,
"count": 1,
"self": 0.0017284580001160066,
"children": {
"RLTrainer._checkpoint": {
"total": 0.11215258999982325,
"count": 1,
"self": 0.11215258999982325
}
}
}
}
}
}
}