ppo-PyramidsRND / run_logs /timers.json
y-korobko's picture
First Push
214b9de verified
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.5885481834411621,
"min": 0.5885481834411621,
"max": 1.509569764137268,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 17581.111328125,
"min": 17581.111328125,
"max": 45794.30859375,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989899.0,
"min": 29952.0,
"max": 989899.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989899.0,
"min": 29952.0,
"max": 989899.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.37280794978141785,
"min": -0.09860341995954514,
"max": 0.37280794978141785,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 99.16691589355469,
"min": -23.763423919677734,
"max": 99.16691589355469,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": -0.05186496302485466,
"min": -0.05186496302485466,
"max": 0.24286338686943054,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": -13.796080589294434,
"min": -13.796080589294434,
"max": 58.28721237182617,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06631203180089977,
"min": 0.06475652475170897,
"max": 0.07360094230600453,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9946804770134965,
"min": 0.5102597278514047,
"max": 1.0343581915618487,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.012653406974011149,
"min": 0.00036206810141345777,
"max": 0.013798886060526242,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.18980110461016722,
"min": 0.003982749115548035,
"max": 0.19318440484736737,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.536297487933329e-06,
"min": 7.536297487933329e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.00011304446231899993,
"min": 0.00011304446231899993,
"max": 0.003225435124854999,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10251206666666667,
"min": 0.10251206666666667,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.537681,
"min": 1.3691136000000002,
"max": 2.3597177,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00026095545999999984,
"min": 0.00026095545999999984,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.003914331899999997,
"min": 0.003914331899999997,
"max": 0.10752698550000002,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.012467634864151478,
"min": 0.012315102852880955,
"max": 0.38476961851119995,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.18701452016830444,
"min": 0.17241144180297852,
"max": 2.693387269973755,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 453.50704225352115,
"min": 453.50704225352115,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 32199.0,
"min": 15984.0,
"max": 32787.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.3434685453772546,
"min": -1.0000000521540642,
"max": 1.3434685453772546,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 94.04279817640781,
"min": -32.000001668930054,
"max": 94.04279817640781,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.3434685453772546,
"min": -1.0000000521540642,
"max": 1.3434685453772546,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 94.04279817640781,
"min": -32.000001668930054,
"max": 94.04279817640781,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.05911630411012863,
"min": 0.05911630411012863,
"max": 8.4098148541525,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 4.138141287709004,
"min": 3.543811898736749,
"max": 134.55703766644,
"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": "1772193744",
"python_version": "3.10.18 (main, Sep 18 2025, 19:50:32) [Clang 20.1.4 ]",
"command_line_arguments": "/home/user/projects/hf-deep-reinforcement-learning-course/unit5/.venv/bin/mlagents-learn ./ml-agents/config/ppo/PyramidsRND.yaml --env=./training-envs-executables/linux/Pyramids/Pyramids --run-id=PyramidsRND-v1 --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": "1772195351"
},
"total": 1606.986277116,
"count": 1,
"self": 0.21880261000001155,
"children": {
"run_training.setup": {
"total": 0.012995429000000058,
"count": 1,
"self": 0.012995429000000058
},
"TrainerController.start_learning": {
"total": 1606.754479077,
"count": 1,
"self": 0.871689447017161,
"children": {
"TrainerController._reset_env": {
"total": 1.7584810139999973,
"count": 1,
"self": 1.7584810139999973
},
"TrainerController.advance": {
"total": 1604.0689350099829,
"count": 63368,
"self": 0.790682645015977,
"children": {
"env_step": {
"total": 895.2196596279778,
"count": 63368,
"self": 660.7589839269783,
"children": {
"SubprocessEnvManager._take_step": {
"total": 233.918878046987,
"count": 63368,
"self": 3.121145841965756,
"children": {
"TorchPolicy.evaluate": {
"total": 230.79773220502125,
"count": 62569,
"self": 230.79773220502125
}
}
},
"workers": {
"total": 0.5417976540125125,
"count": 63368,
"self": 0.0,
"children": {
"worker_root": {
"total": 1604.7467451210086,
"count": 63368,
"is_parallel": true,
"self": 998.905150767014,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0013365329999999176,
"count": 1,
"is_parallel": true,
"self": 0.0006909410000091043,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0006455919999908133,
"count": 8,
"is_parallel": true,
"self": 0.0006455919999908133
}
}
},
"UnityEnvironment.step": {
"total": 0.02013354300000003,
"count": 1,
"is_parallel": true,
"self": 0.00013214500000202634,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.00012504099999688378,
"count": 1,
"is_parallel": true,
"self": 0.00012504099999688378
},
"communicator.exchange": {
"total": 0.019497637000000623,
"count": 1,
"is_parallel": true,
"self": 0.019497637000000623
},
"steps_from_proto": {
"total": 0.00037872000000049866,
"count": 1,
"is_parallel": true,
"self": 9.248099999581427e-05,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0002862390000046844,
"count": 8,
"is_parallel": true,
"self": 0.0002862390000046844
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 605.8415943539947,
"count": 63367,
"is_parallel": true,
"self": 8.882935559958696,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 6.488353090028156,
"count": 63367,
"is_parallel": true,
"self": 6.488353090028156
},
"communicator.exchange": {
"total": 566.7105776620081,
"count": 63367,
"is_parallel": true,
"self": 566.7105776620081
},
"steps_from_proto": {
"total": 23.75972804199983,
"count": 63367,
"is_parallel": true,
"self": 5.650316733085518,
"children": {
"_process_rank_one_or_two_observation": {
"total": 18.10941130891431,
"count": 506936,
"is_parallel": true,
"self": 18.10941130891431
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 708.0585927369891,
"count": 63368,
"self": 1.4995599229677055,
"children": {
"process_trajectory": {
"total": 116.16978465202035,
"count": 63368,
"self": 116.03932034502023,
"children": {
"RLTrainer._checkpoint": {
"total": 0.1304643070001248,
"count": 2,
"self": 0.1304643070001248
}
}
},
"_update_policy": {
"total": 590.3892481620011,
"count": 437,
"self": 250.88786889498851,
"children": {
"TorchPPOOptimizer.update": {
"total": 339.50137926701257,
"count": 22767,
"self": 339.50137926701257
}
}
}
}
}
}
},
"trainer_threads": {
"total": 4.2400006350362673e-07,
"count": 1,
"self": 4.2400006350362673e-07
},
"TrainerController._save_models": {
"total": 0.05537318200003938,
"count": 1,
"self": 0.0006585680000625871,
"children": {
"RLTrainer._checkpoint": {
"total": 0.05471461399997679,
"count": 1,
"self": 0.05471461399997679
}
}
}
}
}
}
}