ppo-PyramidsRND / run_logs /timers.json
mttpv's picture
Pyramids Training
b7d4749 verified
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.41503360867500305,
"min": 0.4128900170326233,
"max": 1.411365032196045,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 12570.5380859375,
"min": 12393.306640625,
"max": 42815.16796875,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989904.0,
"min": 29952.0,
"max": 989904.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989904.0,
"min": 29952.0,
"max": 989904.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.49937739968299866,
"min": -0.0919191986322403,
"max": 0.6232285499572754,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 134.33251953125,
"min": -22.33636474609375,
"max": 174.50399780273438,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": -0.06947971880435944,
"min": -0.06947971880435944,
"max": 0.24687962234020233,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": -18.690044403076172,
"min": -18.690044403076172,
"max": 59.2511100769043,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.0656725510129587,
"min": 0.0656725510129587,
"max": 0.07437242245641217,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9194157141814219,
"min": 0.49473383736859566,
"max": 1.0752122778121345,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.013842523494027776,
"min": 0.0012337335404105185,
"max": 0.01519192946683948,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.19379532891638887,
"min": 0.014804802484926223,
"max": 0.21553598299129484,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.772218837864284e-06,
"min": 7.772218837864284e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.00010881106373009999,
"min": 0.00010881106373009999,
"max": 0.0035080655306449,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10259070714285713,
"min": 0.10259070714285713,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4362698999999999,
"min": 1.3886848,
"max": 2.569355100000001,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00026881164357142856,
"min": 0.00026881164357142856,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.00376336301,
"min": 0.00376336301,
"max": 0.11695857449,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.009324366226792336,
"min": 0.009324366226792336,
"max": 0.38314852118492126,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.130541130900383,
"min": 0.130541130900383,
"max": 2.682039737701416,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 370.85526315789474,
"min": 314.8265306122449,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 28185.0,
"min": 15984.0,
"max": 32705.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.5764973443001509,
"min": -1.0000000521540642,
"max": 1.6664725080296234,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 119.81379816681147,
"min": -30.636601738631725,
"max": 161.14639915525913,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.5764973443001509,
"min": -1.0000000521540642,
"max": 1.6664725080296234,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 119.81379816681147,
"min": -30.636601738631725,
"max": 161.14639915525913,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.036074979535968804,
"min": 0.03428913571309143,
"max": 7.151281975209713,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 2.741698444733629,
"min": 2.741698444733629,
"max": 114.42051160335541,
"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": "1772027989",
"python_version": "3.10.9 (main, Jan 11 2023, 15:21:40) [GCC 11.2.0]",
"command_line_arguments": "/kaggle/working/ml-agents/ml-agents/mlagents/trainers/learn.py /kaggle/working/ml-agents/config/ppo/PyramidsRND.yaml --env=./training-envs-executables/linux/Pyramids/Pyramids --run-id=Pyramids Training --no-graphics",
"mlagents_version": "1.2.0.dev0",
"mlagents_envs_version": "1.2.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "2.8.0+cu128",
"numpy_version": "1.23.5",
"end_time_seconds": "1772029736"
},
"total": 1747.0475950730001,
"count": 1,
"self": 0.3728982600000563,
"children": {
"run_training.setup": {
"total": 0.022490938999908394,
"count": 1,
"self": 0.022490938999908394
},
"TrainerController.start_learning": {
"total": 1746.6522058740002,
"count": 1,
"self": 1.1394194010483716,
"children": {
"TrainerController._reset_env": {
"total": 2.260783475999915,
"count": 1,
"self": 2.260783475999915
},
"TrainerController.advance": {
"total": 1743.1634055099516,
"count": 63901,
"self": 1.1594478419053758,
"children": {
"env_step": {
"total": 1146.420251743013,
"count": 63901,
"self": 1008.907274759109,
"children": {
"SubprocessEnvManager._take_step": {
"total": 136.7805468990016,
"count": 63901,
"self": 4.209268716949055,
"children": {
"TorchPolicy.evaluate": {
"total": 132.57127818205254,
"count": 62544,
"self": 132.57127818205254
}
}
},
"workers": {
"total": 0.7324300849024894,
"count": 63901,
"self": 0.0,
"children": {
"worker_root": {
"total": 1743.8016829630287,
"count": 63901,
"is_parallel": true,
"self": 825.2574948080903,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0017856210001809814,
"count": 1,
"is_parallel": true,
"self": 0.0005716009998195659,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0012140200003614154,
"count": 8,
"is_parallel": true,
"self": 0.0012140200003614154
}
}
},
"UnityEnvironment.step": {
"total": 0.03632395499994345,
"count": 1,
"is_parallel": true,
"self": 0.00034760700009428547,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0005186119999507355,
"count": 1,
"is_parallel": true,
"self": 0.0005186119999507355
},
"communicator.exchange": {
"total": 0.03438278299995545,
"count": 1,
"is_parallel": true,
"self": 0.03438278299995545
},
"steps_from_proto": {
"total": 0.0010749529999429797,
"count": 1,
"is_parallel": true,
"self": 0.00024059899988060351,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0008343540000623761,
"count": 8,
"is_parallel": true,
"self": 0.0008343540000623761
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 918.5441881549384,
"count": 63900,
"is_parallel": true,
"self": 20.998192820015447,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 14.438102907973871,
"count": 63900,
"is_parallel": true,
"self": 14.438102907973871
},
"communicator.exchange": {
"total": 820.1258685330031,
"count": 63900,
"is_parallel": true,
"self": 820.1258685330031
},
"steps_from_proto": {
"total": 62.98202389394601,
"count": 63900,
"is_parallel": true,
"self": 12.758106718965337,
"children": {
"_process_rank_one_or_two_observation": {
"total": 50.22391717498067,
"count": 511200,
"is_parallel": true,
"self": 50.22391717498067
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 595.5837059250332,
"count": 63901,
"self": 2.3803986290788544,
"children": {
"process_trajectory": {
"total": 113.77861637695287,
"count": 63901,
"self": 113.5664345779528,
"children": {
"RLTrainer._checkpoint": {
"total": 0.21218179900006362,
"count": 2,
"self": 0.21218179900006362
}
}
},
"_update_policy": {
"total": 479.4246909190015,
"count": 451,
"self": 259.14279181993857,
"children": {
"TorchPPOOptimizer.update": {
"total": 220.2818990990629,
"count": 22767,
"self": 220.2818990990629
}
}
}
}
}
}
},
"trainer_threads": {
"total": 9.380000847158954e-07,
"count": 1,
"self": 9.380000847158954e-07
},
"TrainerController._save_models": {
"total": 0.08859654900015812,
"count": 1,
"self": 0.0010688840002330835,
"children": {
"RLTrainer._checkpoint": {
"total": 0.08752766499992504,
"count": 1,
"self": 0.08752766499992504
}
}
}
}
}
}
}