knoffel's picture
default-ish parameters
2cb37ac verified
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.12661775946617126,
"min": 0.12661775946617126,
"max": 1.3403754234313965,
"count": 139
},
"Pyramids.Policy.Entropy.sum": {
"value": 3745.85986328125,
"min": 3745.85986328125,
"max": 40661.62890625,
"count": 139
},
"Pyramids.Step.mean": {
"value": 4169953.0,
"min": 29952.0,
"max": 4169953.0,
"count": 139
},
"Pyramids.Step.sum": {
"value": 4169953.0,
"min": 29952.0,
"max": 4169953.0,
"count": 139
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.8268681764602661,
"min": -0.11072057485580444,
"max": 0.9321785569190979,
"count": 139
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 247.23358154296875,
"min": -26.683658599853516,
"max": 293.18743896484375,
"count": 139
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.009303764440119267,
"min": -0.042200781404972076,
"max": 0.6160295009613037,
"count": 139
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 2.781825542449951,
"min": -11.09880542755127,
"max": 145.99899291992188,
"count": 139
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06799422354476216,
"min": 0.06394597697840225,
"max": 0.07374791761004909,
"count": 139
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 1.0199133531714324,
"min": 0.5003612831822194,
"max": 1.0699331861687824,
"count": 139
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.014920670442127934,
"min": 0.0001861820094411787,
"max": 0.016767832649567136,
"count": 139
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.223810056631919,
"min": 0.002420366122735323,
"max": 0.2347496570939399,
"count": 139
},
"Pyramids.Policy.LearningRate.mean": {
"value": 2.8299225506830216e-06,
"min": 2.8299225506830216e-06,
"max": 0.00029884382013985764,
"count": 139
},
"Pyramids.Policy.LearningRate.sum": {
"value": 4.2448838260245327e-05,
"min": 4.2448838260245327e-05,
"max": 0.004001000713742209,
"count": 139
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10094327449798585,
"min": 0.10094327449798585,
"max": 0.1996146065848214,
"count": 139
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.5141491174697876,
"min": 1.3973022460937499,
"max": 2.8336668491363524,
"count": 139
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00010423312234878539,
"min": 0.00010423312234878539,
"max": 0.00996149919782366,
"count": 139
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0015634968352317808,
"min": 0.0015634968352317808,
"max": 0.13338331822872163,
"count": 139
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.005316837690770626,
"min": 0.00511825829744339,
"max": 0.5967947840690613,
"count": 139
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.07975256443023682,
"min": 0.07195756584405899,
"max": 4.177563667297363,
"count": 139
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 230.81203007518798,
"min": 189.74074074074073,
"max": 999.0,
"count": 139
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 30698.0,
"min": 15984.0,
"max": 33497.0,
"count": 139
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.7090150287052743,
"min": -1.0000000521540642,
"max": 1.8060340087632745,
"count": 139
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 227.29899881780148,
"min": -30.17300172895193,
"max": 293.1627984791994,
"count": 139
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.7090150287052743,
"min": -1.0000000521540642,
"max": 1.8060340087632745,
"count": 139
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 227.29899881780148,
"min": -30.17300172895193,
"max": 293.1627984791994,
"count": 139
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.012755914703565142,
"min": 0.01059819941069328,
"max": 12.678219143301249,
"count": 139
},
"Pyramids.Policy.RndReward.sum": {
"value": 1.6965366555741639,
"min": 1.5618635597929824,
"max": 202.85150629281998,
"count": 139
},
"Pyramids.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 139
},
"Pyramids.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 139
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1738594273",
"python_version": "3.10.16 (main, Dec 4 2024, 08:53:38) [GCC 13.2.0]",
"command_line_arguments": "/home/hrichter/projects/hf_rl_course/.venv/bin/mlagents-learn ml-agents/config/ppo/PyramidsRND.yaml --env=./training-envs-executables/linux/Pyramids/Pyramids --run-id=Pyramids Training RND --no-graphics --force",
"mlagents_version": "1.2.0.dev0",
"mlagents_envs_version": "1.2.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "2.5.1+cu124",
"numpy_version": "1.23.5",
"end_time_seconds": "1738598179"
},
"total": 3905.9609358640005,
"count": 1,
"self": 0.2180168859999867,
"children": {
"run_training.setup": {
"total": 0.015057243000001108,
"count": 1,
"self": 0.015057243000001108
},
"TrainerController.start_learning": {
"total": 3905.7278617350003,
"count": 1,
"self": 2.4446929558953343,
"children": {
"TrainerController._reset_env": {
"total": 1.4401749290000225,
"count": 1,
"self": 1.4401749290000225
},
"TrainerController.advance": {
"total": 3901.7912311101054,
"count": 273835,
"self": 2.5200714938191595,
"children": {
"env_step": {
"total": 2490.929578938104,
"count": 273835,
"self": 2185.725352476874,
"children": {
"SubprocessEnvManager._take_step": {
"total": 303.5715818341579,
"count": 273835,
"self": 8.230848409151918,
"children": {
"TorchPolicy.evaluate": {
"total": 295.34073342500596,
"count": 262221,
"self": 295.34073342500596
}
}
},
"workers": {
"total": 1.6326446270722954,
"count": 273835,
"self": 0.0,
"children": {
"worker_root": {
"total": 3901.762734884753,
"count": 273835,
"is_parallel": true,
"self": 1897.81562272177,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0009466559999964375,
"count": 1,
"is_parallel": true,
"self": 0.00022919100001672632,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0007174649999797111,
"count": 8,
"is_parallel": true,
"self": 0.0007174649999797111
}
}
},
"UnityEnvironment.step": {
"total": 0.02041816700000254,
"count": 1,
"is_parallel": true,
"self": 0.00020874099993761774,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.00019684100004724314,
"count": 1,
"is_parallel": true,
"self": 0.00019684100004724314
},
"communicator.exchange": {
"total": 0.01937417099998129,
"count": 1,
"is_parallel": true,
"self": 0.01937417099998129
},
"steps_from_proto": {
"total": 0.0006384140000363914,
"count": 1,
"is_parallel": true,
"self": 0.00014979100001255574,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0004886230000238356,
"count": 8,
"is_parallel": true,
"self": 0.0004886230000238356
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 2003.947112162983,
"count": 273834,
"is_parallel": true,
"self": 47.876944490791175,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 33.30915514407741,
"count": 273834,
"is_parallel": true,
"self": 33.30915514407741
},
"communicator.exchange": {
"total": 1783.656841315873,
"count": 273834,
"is_parallel": true,
"self": 1783.656841315873
},
"steps_from_proto": {
"total": 139.10417121224145,
"count": 273834,
"is_parallel": true,
"self": 30.037486382786255,
"children": {
"_process_rank_one_or_two_observation": {
"total": 109.0666848294552,
"count": 2190672,
"is_parallel": true,
"self": 109.0666848294552
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 1408.3415806781823,
"count": 273835,
"self": 5.083995277255326,
"children": {
"process_trajectory": {
"total": 274.06235688592847,
"count": 273835,
"self": 273.650273610928,
"children": {
"RLTrainer._checkpoint": {
"total": 0.41208327500044106,
"count": 8,
"self": 0.41208327500044106
}
}
},
"_update_policy": {
"total": 1129.1952285149985,
"count": 1962,
"self": 609.6197948401142,
"children": {
"TorchPPOOptimizer.update": {
"total": 519.5754336748843,
"count": 95631,
"self": 519.5754336748843
}
}
}
}
}
}
},
"trainer_threads": {
"total": 5.699994289898314e-07,
"count": 1,
"self": 5.699994289898314e-07
},
"TrainerController._save_models": {
"total": 0.05176217000007455,
"count": 1,
"self": 0.0017888270003822981,
"children": {
"RLTrainer._checkpoint": {
"total": 0.04997334299969225,
"count": 1,
"self": 0.04997334299969225
}
}
}
}
}
}
}