ppo-Pyramids / run_logs /timers.json
asuzuki's picture
1M training steps
736cbf3
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.3894749581813812,
"min": 0.3894749581813812,
"max": 1.37165105342865,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 11553.384765625,
"min": 11553.384765625,
"max": 41610.40625,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989894.0,
"min": 29875.0,
"max": 989894.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989894.0,
"min": 29875.0,
"max": 989894.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.608856201171875,
"min": -0.08415204286575317,
"max": 0.6389569044113159,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 168.65316772460938,
"min": -20.36479377746582,
"max": 182.10272216796875,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": -0.04042782634496689,
"min": -0.04042782634496689,
"max": 0.27796033024787903,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": -11.198508262634277,
"min": -11.198508262634277,
"max": 67.26640319824219,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.0648559197642924,
"min": 0.0648559197642924,
"max": 0.07340954700836509,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9079828767000936,
"min": 0.5872763760669207,
"max": 1.056355008837201,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.016720666352935665,
"min": 0.001097141176042019,
"max": 0.016948496229555788,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.2340893289410993,
"min": 0.014262835288546246,
"max": 0.23727894721378104,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.72051171224286e-06,
"min": 7.72051171224286e-06,
"max": 0.0002948486642171125,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.00010808716397140003,
"min": 0.00010808716397140003,
"max": 0.0036337429887524,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10257347142857146,
"min": 0.10257347142857146,
"max": 0.1982828875,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4360286000000004,
"min": 1.4360286000000004,
"max": 2.6112476,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.0002670897957142858,
"min": 0.0002670897957142858,
"max": 0.009828460461249999,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.003739257140000001,
"min": 0.003739257140000001,
"max": 0.12114363524000002,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.01342105958610773,
"min": 0.013147766701877117,
"max": 0.4541656970977783,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.18789483606815338,
"min": 0.1840687394142151,
"max": 3.6333255767822266,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 321.468085106383,
"min": 281.9622641509434,
"max": 991.5625,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 30218.0,
"min": 16418.0,
"max": 33210.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.657244667014543,
"min": -0.9213313006330281,
"max": 1.6972616670705447,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 155.78099869936705,
"min": -29.4826016202569,
"max": 181.6069983765483,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.657244667014543,
"min": -0.9213313006330281,
"max": 1.6972616670705447,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 155.78099869936705,
"min": -29.4826016202569,
"max": 181.6069983765483,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.044402356304109415,
"min": 0.03842810736033439,
"max": 8.547281347653445,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 4.173821492586285,
"min": 4.015445108001586,
"max": 145.30378291010857,
"count": 33
},
"Pyramids.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 33
},
"Pyramids.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 33
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1676067862",
"python_version": "3.8.10 (default, Nov 14 2022, 12:59:47) \n[GCC 9.4.0]",
"command_line_arguments": "/usr/local/bin/mlagents-learn ./config/ppo/PyramidsRND.yaml --env=./training-envs-executables/linux/Pyramids/Pyramids --run-id=PyramidsTraining --no-graphics",
"mlagents_version": "0.29.0.dev0",
"mlagents_envs_version": "0.29.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "1.8.1+cu102",
"numpy_version": "1.21.6",
"end_time_seconds": "1676070138"
},
"total": 2275.3318676300005,
"count": 1,
"self": 0.4240907820008033,
"children": {
"run_training.setup": {
"total": 0.18653028799963067,
"count": 1,
"self": 0.18653028799963067
},
"TrainerController.start_learning": {
"total": 2274.72124656,
"count": 1,
"self": 1.2637923200622936,
"children": {
"TrainerController._reset_env": {
"total": 7.196981993999998,
"count": 1,
"self": 7.196981993999998
},
"TrainerController.advance": {
"total": 2266.1763559119368,
"count": 63966,
"self": 1.3838251719598702,
"children": {
"env_step": {
"total": 1534.723390248947,
"count": 63966,
"self": 1424.269694713982,
"children": {
"SubprocessEnvManager._take_step": {
"total": 109.65634118300386,
"count": 63966,
"self": 4.442792347150316,
"children": {
"TorchPolicy.evaluate": {
"total": 105.21354883585354,
"count": 62551,
"self": 35.271450269884554,
"children": {
"TorchPolicy.sample_actions": {
"total": 69.94209856596899,
"count": 62551,
"self": 69.94209856596899
}
}
}
}
},
"workers": {
"total": 0.7973543519610757,
"count": 63966,
"self": 0.0,
"children": {
"worker_root": {
"total": 2269.7959138470055,
"count": 63966,
"is_parallel": true,
"self": 955.8003631160987,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0024254790000668436,
"count": 1,
"is_parallel": true,
"self": 0.0007860470004743547,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0016394319995924889,
"count": 8,
"is_parallel": true,
"self": 0.0016394319995924889
}
}
},
"UnityEnvironment.step": {
"total": 0.04411653900024248,
"count": 1,
"is_parallel": true,
"self": 0.0005285590004859841,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0004589199998008553,
"count": 1,
"is_parallel": true,
"self": 0.0004589199998008553
},
"communicator.exchange": {
"total": 0.041528408999965905,
"count": 1,
"is_parallel": true,
"self": 0.041528408999965905
},
"steps_from_proto": {
"total": 0.001600650999989739,
"count": 1,
"is_parallel": true,
"self": 0.00042424499997650855,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0011764060000132304,
"count": 8,
"is_parallel": true,
"self": 0.0011764060000132304
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1313.9955507309069,
"count": 63965,
"is_parallel": true,
"self": 30.756449033914123,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 22.189855891916068,
"count": 63965,
"is_parallel": true,
"self": 22.189855891916068
},
"communicator.exchange": {
"total": 1161.2007584510093,
"count": 63965,
"is_parallel": true,
"self": 1161.2007584510093
},
"steps_from_proto": {
"total": 99.8484873540674,
"count": 63965,
"is_parallel": true,
"self": 21.836080953852615,
"children": {
"_process_rank_one_or_two_observation": {
"total": 78.01240640021479,
"count": 511720,
"is_parallel": true,
"self": 78.01240640021479
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 730.0691404910299,
"count": 63966,
"self": 2.4411927351670784,
"children": {
"process_trajectory": {
"total": 159.45392961886773,
"count": 63966,
"self": 159.27467489286846,
"children": {
"RLTrainer._checkpoint": {
"total": 0.1792547259992716,
"count": 2,
"self": 0.1792547259992716
}
}
},
"_update_policy": {
"total": 568.1740181369951,
"count": 456,
"self": 218.23581475598485,
"children": {
"TorchPPOOptimizer.update": {
"total": 349.9382033810102,
"count": 22785,
"self": 349.9382033810102
}
}
}
}
}
}
},
"trainer_threads": {
"total": 9.690002116258256e-07,
"count": 1,
"self": 9.690002116258256e-07
},
"TrainerController._save_models": {
"total": 0.08411536500079819,
"count": 1,
"self": 0.0014051470016056555,
"children": {
"RLTrainer._checkpoint": {
"total": 0.08271021799919254,
"count": 1,
"self": 0.08271021799919254
}
}
}
}
}
}
}