ppo-Pyramids / run_logs /timers.json
TikhonRadkevich's picture
First Push
025e1f6 verified
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.15648406744003296,
"min": 0.15526911616325378,
"max": 1.4576489925384521,
"count": 100
},
"Pyramids.Policy.Entropy.sum": {
"value": 4682.00341796875,
"min": 4618.32470703125,
"max": 44219.23828125,
"count": 100
},
"Pyramids.Step.mean": {
"value": 2999950.0,
"min": 29952.0,
"max": 2999950.0,
"count": 100
},
"Pyramids.Step.sum": {
"value": 2999950.0,
"min": 29952.0,
"max": 2999950.0,
"count": 100
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.8419708013534546,
"min": -0.09999159723520279,
"max": 0.8908884525299072,
"count": 100
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 254.27517700195312,
"min": -23.997983932495117,
"max": 269.0483093261719,
"count": 100
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": -0.0005854564369656146,
"min": -0.025869233533740044,
"max": 0.3011578619480133,
"count": 100
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": -0.17680785059928894,
"min": -6.726000785827637,
"max": 72.88020324707031,
"count": 100
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06635388502500904,
"min": 0.06375473797067702,
"max": 0.07373669026297643,
"count": 100
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9289543903501266,
"min": 0.4816754833897496,
"max": 1.079091215909655,
"count": 100
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.01574389996727425,
"min": 0.00020649086813176947,
"max": 0.017026865324142444,
"count": 100
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.2204145995418395,
"min": 0.0028908721538447725,
"max": 0.2486327111905934,
"count": 100
},
"Pyramids.Policy.LearningRate.mean": {
"value": 1.560278051369048e-06,
"min": 1.560278051369048e-06,
"max": 0.00029838354339596195,
"count": 100
},
"Pyramids.Policy.LearningRate.sum": {
"value": 2.1843892719166674e-05,
"min": 2.1843892719166674e-05,
"max": 0.003969440876853066,
"count": 100
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10052005952380953,
"min": 0.10052005952380953,
"max": 0.19946118095238097,
"count": 100
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4072808333333335,
"min": 1.3962282666666668,
"max": 2.7674191333333344,
"count": 100
},
"Pyramids.Policy.Beta.mean": {
"value": 6.195394642857145e-05,
"min": 6.195394642857145e-05,
"max": 0.009946171977142856,
"count": 100
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0008673552500000003,
"min": 0.0008673552500000003,
"max": 0.13232237864,
"count": 100
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.005206058733165264,
"min": 0.0050305589102208614,
"max": 0.529541015625,
"count": 100
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.07288482040166855,
"min": 0.07042782753705978,
"max": 3.706787109375,
"count": 100
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 206.6978417266187,
"min": 204.82119205298014,
"max": 999.0,
"count": 100
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 28731.0,
"min": 15984.0,
"max": 33496.0,
"count": 100
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.7805956389377082,
"min": -1.0000000521540642,
"max": 1.7843846004742843,
"count": 100
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 245.72219817340374,
"min": -29.802001617848873,
"max": 267.411798119545,
"count": 100
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.7805956389377082,
"min": -1.0000000521540642,
"max": 1.7843846004742843,
"count": 100
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 245.72219817340374,
"min": -29.802001617848873,
"max": 267.411798119545,
"count": 100
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.011204389614051805,
"min": 0.011204389614051805,
"max": 10.77631295658648,
"count": 100
},
"Pyramids.Policy.RndReward.sum": {
"value": 1.5462057667391491,
"min": 1.5462057667391491,
"max": 172.42100730538368,
"count": 100
},
"Pyramids.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 100
},
"Pyramids.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 100
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1710437298",
"python_version": "3.10.12 (main, Jul 5 2023, 18:54:27) [GCC 11.2.0]",
"command_line_arguments": "/home/tikhon/anaconda3/envs/mlagents/bin/mlagents-learn ../config/ppo/PyramidsRND.yaml --env=./training-envs-executables/linux/Pyramids/Pyramids --run-id=Pyramids Training --no-graphics",
"mlagents_version": "1.0.0",
"mlagents_envs_version": "1.0.0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "2.2.1",
"numpy_version": "1.21.2",
"end_time_seconds": "1710441946"
},
"total": 4647.532118725001,
"count": 1,
"self": 0.31966334099888627,
"children": {
"run_training.setup": {
"total": 0.018745831001069746,
"count": 1,
"self": 0.018745831001069746
},
"TrainerController.start_learning": {
"total": 4647.193709553001,
"count": 1,
"self": 3.5560735466660844,
"children": {
"TrainerController._reset_env": {
"total": 1.5141093009988253,
"count": 1,
"self": 1.5141093009988253
},
"TrainerController.advance": {
"total": 4642.067872695332,
"count": 195069,
"self": 3.4164983091650356,
"children": {
"env_step": {
"total": 3056.4457490606,
"count": 195069,
"self": 2829.550677944624,
"children": {
"SubprocessEnvManager._take_step": {
"total": 224.5273600923192,
"count": 195069,
"self": 10.340624971253419,
"children": {
"TorchPolicy.evaluate": {
"total": 214.1867351210658,
"count": 187547,
"self": 214.1867351210658
}
}
},
"workers": {
"total": 2.3677110236567387,
"count": 195069,
"self": 0.0,
"children": {
"worker_root": {
"total": 4642.366809390001,
"count": 195069,
"is_parallel": true,
"self": 2068.1822051252075,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.001196270999571425,
"count": 1,
"is_parallel": true,
"self": 0.00033221499870705884,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0008640560008643661,
"count": 8,
"is_parallel": true,
"self": 0.0008640560008643661
}
}
},
"UnityEnvironment.step": {
"total": 0.0372296980003739,
"count": 1,
"is_parallel": true,
"self": 0.0002497889981896151,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0003989140004705405,
"count": 1,
"is_parallel": true,
"self": 0.0003989140004705405
},
"communicator.exchange": {
"total": 0.035797853000985924,
"count": 1,
"is_parallel": true,
"self": 0.035797853000985924
},
"steps_from_proto": {
"total": 0.0007831420007278211,
"count": 1,
"is_parallel": true,
"self": 0.00017009699877235107,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0006130450019554701,
"count": 8,
"is_parallel": true,
"self": 0.0006130450019554701
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 2574.184604264794,
"count": 195068,
"is_parallel": true,
"self": 71.99591521421826,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 47.65340357073546,
"count": 195068,
"is_parallel": true,
"self": 47.65340357073546
},
"communicator.exchange": {
"total": 2247.579691982224,
"count": 195068,
"is_parallel": true,
"self": 2247.579691982224
},
"steps_from_proto": {
"total": 206.95559349761606,
"count": 195068,
"is_parallel": true,
"self": 41.96395975608539,
"children": {
"_process_rank_one_or_two_observation": {
"total": 164.99163374153068,
"count": 1560544,
"is_parallel": true,
"self": 164.99163374153068
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 1582.2056253255669,
"count": 195069,
"self": 7.049938452297283,
"children": {
"process_trajectory": {
"total": 250.09003616930022,
"count": 195069,
"self": 249.75688899029774,
"children": {
"RLTrainer._checkpoint": {
"total": 0.3331471790024807,
"count": 6,
"self": 0.3331471790024807
}
}
},
"_update_policy": {
"total": 1325.0656507039694,
"count": 1397,
"self": 719.2999167647304,
"children": {
"TorchPPOOptimizer.update": {
"total": 605.765733939239,
"count": 68331,
"self": 605.765733939239
}
}
}
}
}
}
},
"trainer_threads": {
"total": 7.570015441160649e-07,
"count": 1,
"self": 7.570015441160649e-07
},
"TrainerController._save_models": {
"total": 0.05565325300267432,
"count": 1,
"self": 0.0016590530030953232,
"children": {
"RLTrainer._checkpoint": {
"total": 0.053994199999578996,
"count": 1,
"self": 0.053994199999578996
}
}
}
}
}
}
}