ppo-Pyramids / run_logs /timers.json
MarcLinder's picture
First Push
8fe00a4 verified
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.3534260094165802,
"min": 0.34234288334846497,
"max": 1.434495210647583,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 10715.876953125,
"min": 10111.439453125,
"max": 43516.84765625,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989936.0,
"min": 29952.0,
"max": 989936.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989936.0,
"min": 29952.0,
"max": 989936.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.5991044640541077,
"min": -0.10557456314563751,
"max": 0.6793408989906311,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 167.74925231933594,
"min": -25.654619216918945,
"max": 194.29148864746094,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.011266621761023998,
"min": -0.007196594029664993,
"max": 0.30974820256233215,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 3.154654026031494,
"min": -1.8423280715942383,
"max": 74.6493148803711,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06622426579200519,
"min": 0.06391665406616605,
"max": 0.07553025259193238,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9271397210880726,
"min": 0.5287117681435267,
"max": 1.0573538166160386,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.01501914461585416,
"min": 0.0016738163737367974,
"max": 0.01659836420149077,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.21026802462195823,
"min": 0.015064347363631176,
"max": 0.24897546302236154,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.249619012064286e-06,
"min": 7.249619012064286e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.0001014946661689,
"min": 0.0001014946661689,
"max": 0.0036363409878864,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10241650714285712,
"min": 0.10241650714285712,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4338310999999997,
"min": 1.3886848,
"max": 2.6121136,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.0002514090635714286,
"min": 0.0002514090635714286,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0035197268900000003,
"min": 0.0035197268900000003,
"max": 0.12123014864000002,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.010126019828021526,
"min": 0.010126019828021526,
"max": 0.49672189354896545,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.14176428318023682,
"min": 0.14176428318023682,
"max": 3.477053165435791,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 319.1818181818182,
"min": 307.970297029703,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 31599.0,
"min": 15984.0,
"max": 32972.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.599712227832298,
"min": -1.0000000521540642,
"max": 1.6685115608729815,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 156.7717983275652,
"min": -30.66200165450573,
"max": 166.89359805732965,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.599712227832298,
"min": -1.0000000521540642,
"max": 1.6685115608729815,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 156.7717983275652,
"min": -30.66200165450573,
"max": 166.89359805732965,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.033309855512925424,
"min": 0.033309855512925424,
"max": 10.439564069733024,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 3.2643658402666915,
"min": 3.1224805404926883,
"max": 167.03302511572838,
"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": "1705325726",
"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": "2.1.2+cu121",
"numpy_version": "1.23.5",
"end_time_seconds": "1705329333"
},
"total": 3607.841228728,
"count": 1,
"self": 0.829808476999915,
"children": {
"run_training.setup": {
"total": 0.06330247400001099,
"count": 1,
"self": 0.06330247400001099
},
"TrainerController.start_learning": {
"total": 3606.948117777,
"count": 1,
"self": 2.724354571927961,
"children": {
"TrainerController._reset_env": {
"total": 3.683230705999904,
"count": 1,
"self": 3.683230705999904
},
"TrainerController.advance": {
"total": 3600.4390712730724,
"count": 64015,
"self": 3.239658485975724,
"children": {
"env_step": {
"total": 2444.4357356711002,
"count": 64015,
"self": 2258.944618052115,
"children": {
"SubprocessEnvManager._take_step": {
"total": 183.73349205010936,
"count": 64015,
"self": 8.02384787193887,
"children": {
"TorchPolicy.evaluate": {
"total": 175.7096441781705,
"count": 62550,
"self": 175.7096441781705
}
}
},
"workers": {
"total": 1.7576255688759375,
"count": 64015,
"self": 0.0,
"children": {
"worker_root": {
"total": 3598.8315337819304,
"count": 64015,
"is_parallel": true,
"self": 1546.187217109898,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0033766870001272764,
"count": 1,
"is_parallel": true,
"self": 0.0010693330004869495,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.002307353999640327,
"count": 8,
"is_parallel": true,
"self": 0.002307353999640327
}
}
},
"UnityEnvironment.step": {
"total": 0.13677786700009165,
"count": 1,
"is_parallel": true,
"self": 0.004070116000093549,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0006074970001463953,
"count": 1,
"is_parallel": true,
"self": 0.0006074970001463953
},
"communicator.exchange": {
"total": 0.12992459099996267,
"count": 1,
"is_parallel": true,
"self": 0.12992459099996267
},
"steps_from_proto": {
"total": 0.0021756629998890276,
"count": 1,
"is_parallel": true,
"self": 0.00046744300016143825,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0017082199997275893,
"count": 8,
"is_parallel": true,
"self": 0.0017082199997275893
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 2052.6443166720323,
"count": 64014,
"is_parallel": true,
"self": 51.93975271503132,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 30.290297101948,
"count": 64014,
"is_parallel": true,
"self": 30.290297101948
},
"communicator.exchange": {
"total": 1831.2920596630715,
"count": 64014,
"is_parallel": true,
"self": 1831.2920596630715
},
"steps_from_proto": {
"total": 139.1222071919815,
"count": 64014,
"is_parallel": true,
"self": 31.258446414703485,
"children": {
"_process_rank_one_or_two_observation": {
"total": 107.863760777278,
"count": 512112,
"is_parallel": true,
"self": 107.863760777278
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 1152.7636771159964,
"count": 64015,
"self": 5.937606095236788,
"children": {
"process_trajectory": {
"total": 185.01342999076223,
"count": 64015,
"self": 184.82084157576173,
"children": {
"RLTrainer._checkpoint": {
"total": 0.1925884150005004,
"count": 2,
"self": 0.1925884150005004
}
}
},
"_update_policy": {
"total": 961.8126410299974,
"count": 456,
"self": 388.2790339838939,
"children": {
"TorchPPOOptimizer.update": {
"total": 573.5336070461035,
"count": 22815,
"self": 573.5336070461035
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.2539994713733904e-06,
"count": 1,
"self": 1.2539994713733904e-06
},
"TrainerController._save_models": {
"total": 0.101459972000157,
"count": 1,
"self": 0.0038363170006050495,
"children": {
"RLTrainer._checkpoint": {
"total": 0.09762365499955195,
"count": 1,
"self": 0.09762365499955195
}
}
}
}
}
}
}