slava-medvedev's picture
First Push
47d963f
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.42098483443260193,
"min": 0.41547560691833496,
"max": 1.4425572156906128,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 12609.337890625,
"min": 12417.7353515625,
"max": 43761.4140625,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989967.0,
"min": 29952.0,
"max": 989967.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989967.0,
"min": 29952.0,
"max": 989967.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.4700731039047241,
"min": -0.10974668711423874,
"max": 0.5214748382568359,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 126.91973876953125,
"min": -26.448951721191406,
"max": 141.84115600585938,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.05180627852678299,
"min": 0.004674371797591448,
"max": 0.3023749589920044,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 13.987695693969727,
"min": 1.2761034965515137,
"max": 73.17474365234375,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.061823614483555056,
"min": 0.061823614483555056,
"max": 0.07240533088608726,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.8655306027697708,
"min": 0.5005624706906416,
"max": 1.0832479809663105,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.014668682676745535,
"min": 0.00017302395998833678,
"max": 0.01570857674613806,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.2053615574744375,
"min": 0.002249311479848378,
"max": 0.22674141218158184,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.701976004135714e-06,
"min": 7.701976004135714e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.0001078276640579,
"min": 0.0001078276640579,
"max": 0.0034911292362902998,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10256729285714286,
"min": 0.10256729285714286,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4359421,
"min": 1.3886848,
"max": 2.4854386,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.0002664725564285715,
"min": 0.0002664725564285715,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0037306157900000004,
"min": 0.0037306157900000004,
"max": 0.11638459903000002,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.01107530202716589,
"min": 0.01086423546075821,
"max": 0.4281166195869446,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.1550542265176773,
"min": 0.15209929645061493,
"max": 2.996816396713257,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 422.8767123287671,
"min": 348.6309523809524,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 30870.0,
"min": 15984.0,
"max": 32728.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.4626722026409373,
"min": -1.0000000521540642,
"max": 1.5400537096742373,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 105.3123985901475,
"min": -31.99520167708397,
"max": 128.71219782531261,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.4626722026409373,
"min": -1.0000000521540642,
"max": 1.5400537096742373,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 105.3123985901475,
"min": -31.99520167708397,
"max": 128.71219782531261,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.04892451667728892,
"min": 0.0393231837480839,
"max": 8.276825843844563,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 3.5225652007648023,
"min": 3.3031474348390475,
"max": 132.429213501513,
"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": "1698236550",
"python_version": "3.10.12 (main, Jun 11 2023, 05:26:28) [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": "2.1.0+cu118",
"numpy_version": "1.23.5",
"end_time_seconds": "1698238685"
},
"total": 2135.290643153,
"count": 1,
"self": 0.47869165600059205,
"children": {
"run_training.setup": {
"total": 0.046051006999732635,
"count": 1,
"self": 0.046051006999732635
},
"TrainerController.start_learning": {
"total": 2134.7659004899997,
"count": 1,
"self": 1.3700840529772904,
"children": {
"TrainerController._reset_env": {
"total": 3.431643236000127,
"count": 1,
"self": 3.431643236000127
},
"TrainerController.advance": {
"total": 2129.8926128790226,
"count": 63707,
"self": 1.4205622069739547,
"children": {
"env_step": {
"total": 1505.3886351619922,
"count": 63707,
"self": 1371.4256809710196,
"children": {
"SubprocessEnvManager._take_step": {
"total": 133.106052826025,
"count": 63707,
"self": 4.659463870032596,
"children": {
"TorchPolicy.evaluate": {
"total": 128.4465889559924,
"count": 62546,
"self": 128.4465889559924
}
}
},
"workers": {
"total": 0.8569013649475892,
"count": 63707,
"self": 0.0,
"children": {
"worker_root": {
"total": 2129.9131777720095,
"count": 63707,
"is_parallel": true,
"self": 874.3799233020127,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0018315900006200536,
"count": 1,
"is_parallel": true,
"self": 0.000596575002418831,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0012350149982012226,
"count": 8,
"is_parallel": true,
"self": 0.0012350149982012226
}
}
},
"UnityEnvironment.step": {
"total": 0.04980811699988408,
"count": 1,
"is_parallel": true,
"self": 0.0006347080006889882,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.00043348500003048684,
"count": 1,
"is_parallel": true,
"self": 0.00043348500003048684
},
"communicator.exchange": {
"total": 0.04693138899983751,
"count": 1,
"is_parallel": true,
"self": 0.04693138899983751
},
"steps_from_proto": {
"total": 0.0018085349993270938,
"count": 1,
"is_parallel": true,
"self": 0.00037562100169452606,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0014329139976325678,
"count": 8,
"is_parallel": true,
"self": 0.0014329139976325678
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1255.5332544699968,
"count": 63706,
"is_parallel": true,
"self": 34.374128663032025,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 24.843582830080777,
"count": 63706,
"is_parallel": true,
"self": 24.843582830080777
},
"communicator.exchange": {
"total": 1095.7539137979893,
"count": 63706,
"is_parallel": true,
"self": 1095.7539137979893
},
"steps_from_proto": {
"total": 100.56162917889469,
"count": 63706,
"is_parallel": true,
"self": 20.638753640169853,
"children": {
"_process_rank_one_or_two_observation": {
"total": 79.92287553872484,
"count": 509648,
"is_parallel": true,
"self": 79.92287553872484
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 623.0834155100565,
"count": 63707,
"self": 2.587555203120246,
"children": {
"process_trajectory": {
"total": 121.84149535693723,
"count": 63707,
"self": 121.66328953993707,
"children": {
"RLTrainer._checkpoint": {
"total": 0.17820581700016191,
"count": 2,
"self": 0.17820581700016191
}
}
},
"_update_policy": {
"total": 498.65436494999904,
"count": 452,
"self": 294.22260972296317,
"children": {
"TorchPPOOptimizer.update": {
"total": 204.43175522703586,
"count": 22797,
"self": 204.43175522703586
}
}
}
}
}
}
},
"trainer_threads": {
"total": 9.959994713426568e-07,
"count": 1,
"self": 9.959994713426568e-07
},
"TrainerController._save_models": {
"total": 0.07155932600016968,
"count": 1,
"self": 0.001283175000025949,
"children": {
"RLTrainer._checkpoint": {
"total": 0.07027615100014373,
"count": 1,
"self": 0.07027615100014373
}
}
}
}
}
}
}