StaffanRolfsson's picture
Learning bug fixed
3dcc668 verified
{
"name": "root",
"gauges": {
"SnowballTarget.Policy.Entropy.mean": {
"value": 0.8238164782524109,
"min": 0.8238164782524109,
"max": 2.854252338409424,
"count": 20
},
"SnowballTarget.Policy.Entropy.sum": {
"value": 7829.5517578125,
"min": 7829.5517578125,
"max": 29136.20703125,
"count": 20
},
"SnowballTarget.Step.mean": {
"value": 199984.0,
"min": 9952.0,
"max": 199984.0,
"count": 20
},
"SnowballTarget.Step.sum": {
"value": 199984.0,
"min": 9952.0,
"max": 199984.0,
"count": 20
},
"SnowballTarget.Policy.ExtrinsicValueEstimate.mean": {
"value": 12.826810836791992,
"min": 0.35294824838638306,
"max": 12.826810836791992,
"count": 20
},
"SnowballTarget.Policy.ExtrinsicValueEstimate.sum": {
"value": 2501.22802734375,
"min": 68.47196197509766,
"max": 2612.640380859375,
"count": 20
},
"SnowballTarget.Losses.PolicyLoss.mean": {
"value": 0.06773690539115455,
"min": 0.05984110653619556,
"max": 0.07475832272716883,
"count": 20
},
"SnowballTarget.Losses.PolicyLoss.sum": {
"value": 0.2709476215646182,
"min": 0.23936442614478223,
"max": 0.3737916136358441,
"count": 20
},
"SnowballTarget.Losses.ValueLoss.mean": {
"value": 0.19683007815597103,
"min": 0.13347096463628844,
"max": 0.2806715703039777,
"count": 20
},
"SnowballTarget.Losses.ValueLoss.sum": {
"value": 0.7873203126238841,
"min": 0.5338838585451537,
"max": 1.3936233032567829,
"count": 20
},
"SnowballTarget.Policy.LearningRate.mean": {
"value": 8.082097306000005e-06,
"min": 8.082097306000005e-06,
"max": 0.000291882002706,
"count": 20
},
"SnowballTarget.Policy.LearningRate.sum": {
"value": 3.232838922400002e-05,
"min": 3.232838922400002e-05,
"max": 0.00138516003828,
"count": 20
},
"SnowballTarget.Policy.Epsilon.mean": {
"value": 0.10269400000000001,
"min": 0.10269400000000001,
"max": 0.19729400000000002,
"count": 20
},
"SnowballTarget.Policy.Epsilon.sum": {
"value": 0.41077600000000003,
"min": 0.41077600000000003,
"max": 0.96172,
"count": 20
},
"SnowballTarget.Policy.Beta.mean": {
"value": 0.0001444306000000001,
"min": 0.0001444306000000001,
"max": 0.0048649706,
"count": 20
},
"SnowballTarget.Policy.Beta.sum": {
"value": 0.0005777224000000004,
"min": 0.0005777224000000004,
"max": 0.023089828,
"count": 20
},
"SnowballTarget.Environment.EpisodeLength.mean": {
"value": 199.0,
"min": 199.0,
"max": 199.0,
"count": 20
},
"SnowballTarget.Environment.EpisodeLength.sum": {
"value": 8756.0,
"min": 8756.0,
"max": 10945.0,
"count": 20
},
"SnowballTarget.Environment.CumulativeReward.mean": {
"value": 25.613636363636363,
"min": 3.7045454545454546,
"max": 25.613636363636363,
"count": 20
},
"SnowballTarget.Environment.CumulativeReward.sum": {
"value": 1127.0,
"min": 163.0,
"max": 1375.0,
"count": 20
},
"SnowballTarget.Policy.ExtrinsicReward.mean": {
"value": 25.613636363636363,
"min": 3.7045454545454546,
"max": 25.613636363636363,
"count": 20
},
"SnowballTarget.Policy.ExtrinsicReward.sum": {
"value": 1127.0,
"min": 163.0,
"max": 1375.0,
"count": 20
},
"SnowballTarget.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 20
},
"SnowballTarget.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 20
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1730457667",
"python_version": "3.10.12 (main, Sep 11 2024, 15:47:36) [GCC 11.4.0]",
"command_line_arguments": "/usr/local/bin/mlagents-learn ./config/ppo/SnowballTarget.yaml --env=./training-envs-executables/linux/SnowballTarget/SnowballTarget --run-id=SnowballTarget3 --no-graphics",
"mlagents_version": "1.2.0.dev0",
"mlagents_envs_version": "1.2.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "2.5.0+cu121",
"numpy_version": "1.23.5",
"end_time_seconds": "1730458152"
},
"total": 485.2031802299998,
"count": 1,
"self": 0.4909352700005911,
"children": {
"run_training.setup": {
"total": 0.060118817999864405,
"count": 1,
"self": 0.060118817999864405
},
"TrainerController.start_learning": {
"total": 484.65212614199936,
"count": 1,
"self": 0.4465823720638582,
"children": {
"TrainerController._reset_env": {
"total": 2.097126030999789,
"count": 1,
"self": 2.097126030999789
},
"TrainerController.advance": {
"total": 482.0110896639353,
"count": 18192,
"self": 0.45769853190085996,
"children": {
"env_step": {
"total": 339.99699726599556,
"count": 18192,
"self": 260.36097829788923,
"children": {
"SubprocessEnvManager._take_step": {
"total": 79.34327630703046,
"count": 18192,
"self": 1.4361591449942352,
"children": {
"TorchPolicy.evaluate": {
"total": 77.90711716203623,
"count": 18192,
"self": 77.90711716203623
}
}
},
"workers": {
"total": 0.292742661075863,
"count": 18192,
"self": 0.0,
"children": {
"worker_root": {
"total": 482.89055251393984,
"count": 18192,
"is_parallel": true,
"self": 255.2187081369675,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.002146708000509534,
"count": 1,
"is_parallel": true,
"self": 0.0006805399998484063,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0014661680006611277,
"count": 10,
"is_parallel": true,
"self": 0.0014661680006611277
}
}
},
"UnityEnvironment.step": {
"total": 0.033992140000009385,
"count": 1,
"is_parallel": true,
"self": 0.0006545869991896325,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.00041273799979535397,
"count": 1,
"is_parallel": true,
"self": 0.00041273799979535397
},
"communicator.exchange": {
"total": 0.031093510000573588,
"count": 1,
"is_parallel": true,
"self": 0.031093510000573588
},
"steps_from_proto": {
"total": 0.0018313050004508113,
"count": 1,
"is_parallel": true,
"self": 0.0003532840009938809,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0014780209994569304,
"count": 10,
"is_parallel": true,
"self": 0.0014780209994569304
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 227.67184437697233,
"count": 18191,
"is_parallel": true,
"self": 11.375857074863234,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 5.977144897168728,
"count": 18191,
"is_parallel": true,
"self": 5.977144897168728
},
"communicator.exchange": {
"total": 175.57844794897755,
"count": 18191,
"is_parallel": true,
"self": 175.57844794897755
},
"steps_from_proto": {
"total": 34.740394455962814,
"count": 18191,
"is_parallel": true,
"self": 6.394949745939812,
"children": {
"_process_rank_one_or_two_observation": {
"total": 28.345444710023003,
"count": 181910,
"is_parallel": true,
"self": 28.345444710023003
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 141.5563938660389,
"count": 18192,
"self": 0.5527828149870402,
"children": {
"process_trajectory": {
"total": 30.309547771056714,
"count": 18192,
"self": 29.887203831056468,
"children": {
"RLTrainer._checkpoint": {
"total": 0.4223439400002462,
"count": 4,
"self": 0.4223439400002462
}
}
},
"_update_policy": {
"total": 110.69406327999513,
"count": 90,
"self": 46.07532737297879,
"children": {
"TorchPPOOptimizer.update": {
"total": 64.61873590701634,
"count": 4587,
"self": 64.61873590701634
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.5120003808988258e-06,
"count": 1,
"self": 1.5120003808988258e-06
},
"TrainerController._save_models": {
"total": 0.09732656300002418,
"count": 1,
"self": 0.00097503799952392,
"children": {
"RLTrainer._checkpoint": {
"total": 0.09635152500050026,
"count": 1,
"self": 0.09635152500050026
}
}
}
}
}
}
}