ppo-PyramidsRND / run_logs /timers.json
rematchka's picture
First Push
cc6e46c
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.3056558668613434,
"min": 0.3056558668613434,
"max": 1.4162405729293823,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 9106.099609375,
"min": 9106.099609375,
"max": 42963.07421875,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989902.0,
"min": 29952.0,
"max": 989902.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989902.0,
"min": 29952.0,
"max": 989902.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.6013326644897461,
"min": -0.10716479271650314,
"max": 0.6211044192314148,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 174.386474609375,
"min": -25.71955108642578,
"max": 175.77255249023438,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.009950211271643639,
"min": -0.027393080294132233,
"max": 0.608289897441864,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 2.885561227798462,
"min": -7.176987171173096,
"max": 144.16470336914062,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.07023719387403357,
"min": 0.06468444040371397,
"max": 0.07359107546391812,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.98332071423647,
"min": 0.5063820474405758,
"max": 1.054098236898426,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.016747928635524264,
"min": 0.000863756886924007,
"max": 0.016747928635524264,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.23447100089733972,
"min": 0.00863756886924007,
"max": 0.23447100089733972,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.604497465200001e-06,
"min": 7.604497465200001e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.00010646296451280002,
"min": 0.00010646296451280002,
"max": 0.0033820526726492,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10253480000000001,
"min": 0.10253480000000001,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4354872,
"min": 1.3886848,
"max": 2.5273508000000007,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00026322652000000007,
"min": 0.00026322652000000007,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.003685171280000001,
"min": 0.003685171280000001,
"max": 0.11276234492,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.01340904738754034,
"min": 0.01340904738754034,
"max": 0.5663581490516663,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.18772666156291962,
"min": 0.18772666156291962,
"max": 3.9645071029663086,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 302.009900990099,
"min": 302.009900990099,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 30503.0,
"min": 15984.0,
"max": 32975.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.6583722602612903,
"min": -1.0000000521540642,
"max": 1.6583722602612903,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 167.4955982863903,
"min": -30.71360170096159,
"max": 167.4955982863903,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.6583722602612903,
"min": -1.0000000521540642,
"max": 1.6583722602612903,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 167.4955982863903,
"min": -30.71360170096159,
"max": 167.4955982863903,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.042047608472352976,
"min": 0.042047608472352976,
"max": 11.50680017657578,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 4.246808455707651,
"min": 4.013747861783486,
"max": 184.10880282521248,
"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": "1676975583",
"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=Pyramids Training --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": "1676978035"
},
"total": 2452.387272752,
"count": 1,
"self": 0.4385405540006104,
"children": {
"run_training.setup": {
"total": 0.1186270880000393,
"count": 1,
"self": 0.1186270880000393
},
"TrainerController.start_learning": {
"total": 2451.8301051099997,
"count": 1,
"self": 1.5178045990060127,
"children": {
"TrainerController._reset_env": {
"total": 6.197488776,
"count": 1,
"self": 6.197488776
},
"TrainerController.advance": {
"total": 2444.0276519849936,
"count": 63898,
"self": 1.5075098049296685,
"children": {
"env_step": {
"total": 1676.9140122699755,
"count": 63898,
"self": 1555.3637564139385,
"children": {
"SubprocessEnvManager._take_step": {
"total": 120.6151101590151,
"count": 63898,
"self": 4.90597812705073,
"children": {
"TorchPolicy.evaluate": {
"total": 115.70913203196437,
"count": 62560,
"self": 38.597625940971284,
"children": {
"TorchPolicy.sample_actions": {
"total": 77.11150609099309,
"count": 62560,
"self": 77.11150609099309
}
}
}
}
},
"workers": {
"total": 0.9351456970218806,
"count": 63898,
"self": 0.0,
"children": {
"worker_root": {
"total": 2446.412113820978,
"count": 63898,
"is_parallel": true,
"self": 1011.4332039470635,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0017574149999290967,
"count": 1,
"is_parallel": true,
"self": 0.0006699090001802688,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0010875059997488279,
"count": 8,
"is_parallel": true,
"self": 0.0010875059997488279
}
}
},
"UnityEnvironment.step": {
"total": 0.044408441000086896,
"count": 1,
"is_parallel": true,
"self": 0.0005222600002525724,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0005068329999176058,
"count": 1,
"is_parallel": true,
"self": 0.0005068329999176058
},
"communicator.exchange": {
"total": 0.04180296799995631,
"count": 1,
"is_parallel": true,
"self": 0.04180296799995631
},
"steps_from_proto": {
"total": 0.001576379999960409,
"count": 1,
"is_parallel": true,
"self": 0.0004149030003191001,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0011614769996413088,
"count": 8,
"is_parallel": true,
"self": 0.0011614769996413088
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1434.9789098739145,
"count": 63897,
"is_parallel": true,
"self": 32.67435278894686,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 23.35456198398083,
"count": 63897,
"is_parallel": true,
"self": 23.35456198398083
},
"communicator.exchange": {
"total": 1281.702453377013,
"count": 63897,
"is_parallel": true,
"self": 1281.702453377013
},
"steps_from_proto": {
"total": 97.24754172397388,
"count": 63897,
"is_parallel": true,
"self": 23.171267542010128,
"children": {
"_process_rank_one_or_two_observation": {
"total": 74.07627418196375,
"count": 511176,
"is_parallel": true,
"self": 74.07627418196375
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 765.6061299100884,
"count": 63898,
"self": 2.755064031153779,
"children": {
"process_trajectory": {
"total": 165.37350648593974,
"count": 63898,
"self": 165.1924692139405,
"children": {
"RLTrainer._checkpoint": {
"total": 0.18103727199923014,
"count": 2,
"self": 0.18103727199923014
}
}
},
"_update_policy": {
"total": 597.4775593929949,
"count": 444,
"self": 234.89255442395302,
"children": {
"TorchPPOOptimizer.update": {
"total": 362.58500496904185,
"count": 22833,
"self": 362.58500496904185
}
}
}
}
}
}
},
"trainer_threads": {
"total": 8.950000847107731e-07,
"count": 1,
"self": 8.950000847107731e-07
},
"TrainerController._save_models": {
"total": 0.08715885499987053,
"count": 1,
"self": 0.0014308420004454092,
"children": {
"RLTrainer._checkpoint": {
"total": 0.08572801299942512,
"count": 1,
"self": 0.08572801299942512
}
}
}
}
}
}
}