user05181824's picture
Train a Snowball agent with PPO
5373c8f verified
{
"name": "root",
"gauges": {
"SnowballTarget.Policy.Entropy.mean": {
"value": 0.9829105138778687,
"min": 0.9829105138778687,
"max": 2.86360502243042,
"count": 20
},
"SnowballTarget.Policy.Entropy.sum": {
"value": 9352.3935546875,
"min": 9352.3935546875,
"max": 29263.1796875,
"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": 12.728230476379395,
"min": 0.23321369290351868,
"max": 12.728230476379395,
"count": 20
},
"SnowballTarget.Policy.ExtrinsicValueEstimate.sum": {
"value": 2482.0048828125,
"min": 45.24345779418945,
"max": 2554.23291015625,
"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.Losses.PolicyLoss.mean": {
"value": 0.06822366482390184,
"min": 0.06287294063019777,
"max": 0.07442477357165213,
"count": 20
},
"SnowballTarget.Losses.PolicyLoss.sum": {
"value": 0.27289465929560736,
"min": 0.2514917625207911,
"max": 0.36435870416736343,
"count": 20
},
"SnowballTarget.Losses.ValueLoss.mean": {
"value": 0.23009178970082134,
"min": 0.10962210796252989,
"max": 0.2729754504035501,
"count": 20
},
"SnowballTarget.Losses.ValueLoss.sum": {
"value": 0.9203671588032853,
"min": 0.43848843185011954,
"max": 1.3648772520177506,
"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.CumulativeReward.mean": {
"value": 25.363636363636363,
"min": 3.159090909090909,
"max": 25.363636363636363,
"count": 20
},
"SnowballTarget.Environment.CumulativeReward.sum": {
"value": 1116.0,
"min": 139.0,
"max": 1374.0,
"count": 20
},
"SnowballTarget.Policy.ExtrinsicReward.mean": {
"value": 25.363636363636363,
"min": 3.159090909090909,
"max": 25.363636363636363,
"count": 20
},
"SnowballTarget.Policy.ExtrinsicReward.sum": {
"value": 1116.0,
"min": 139.0,
"max": 1374.0,
"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": "1758906028",
"python_version": "3.11.13 (main, Jul 11 2025, 22:43:55) [Clang 20.1.4 ]",
"command_line_arguments": "/home/klinghu/code/github/other/ml-agents/ml-agents/mlagents/trainers/learn.py --train ./SnowballTarget.yaml --env=./training-envs-executables/linux/SnowballTarget/SnowballTarget --run-id=SnowballTarget1 --no-graphics --torch-device cuda",
"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": "1758906533"
},
"total": 494.99436059899745,
"count": 1,
"self": 0.16675337999186013,
"children": {
"run_training.setup": {
"total": 0.013972544002172071,
"count": 1,
"self": 0.013972544002172071
},
"TrainerController.start_learning": {
"total": 494.8136346750034,
"count": 1,
"self": 0.2763546610731282,
"children": {
"TrainerController._reset_env": {
"total": 1.444417800004885,
"count": 1,
"self": 1.444417800004885
},
"TrainerController.advance": {
"total": 493.0330200559256,
"count": 18196,
"self": 0.1201501714749611,
"children": {
"env_step": {
"total": 492.91286988445063,
"count": 18196,
"self": 385.6620164292399,
"children": {
"SubprocessEnvManager._take_step": {
"total": 107.13010976585792,
"count": 18196,
"self": 0.7269739701223443,
"children": {
"TorchPolicy.evaluate": {
"total": 106.40313579573558,
"count": 18196,
"self": 106.40313579573558
}
}
},
"workers": {
"total": 0.12074368935282109,
"count": 18196,
"self": 0.0,
"children": {
"worker_root": {
"total": 494.14473895351693,
"count": 18196,
"is_parallel": true,
"self": 131.17301208998833,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0011569010021048598,
"count": 1,
"is_parallel": true,
"self": 0.0004002900095656514,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0007566109925392084,
"count": 10,
"is_parallel": true,
"self": 0.0007566109925392084
}
}
},
"UnityEnvironment.step": {
"total": 0.025385741006175522,
"count": 1,
"is_parallel": true,
"self": 7.079900387907401e-05,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0002776300025288947,
"count": 1,
"is_parallel": true,
"self": 0.0002776300025288947
},
"communicator.exchange": {
"total": 0.024533028998121154,
"count": 1,
"is_parallel": true,
"self": 0.024533028998121154
},
"steps_from_proto": {
"total": 0.0005042830016463995,
"count": 1,
"is_parallel": true,
"self": 0.00011495100625324994,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.00038933199539314955,
"count": 10,
"is_parallel": true,
"self": 0.00038933199539314955
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 362.9717268635286,
"count": 18195,
"is_parallel": true,
"self": 1.12122208213259,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 3.589410522232356,
"count": 18195,
"is_parallel": true,
"self": 3.589410522232356
},
"communicator.exchange": {
"total": 350.6904604559095,
"count": 18195,
"is_parallel": true,
"self": 350.6904604559095
},
"steps_from_proto": {
"total": 7.570633803254168,
"count": 18195,
"is_parallel": true,
"self": 1.6110896334212157,
"children": {
"_process_rank_one_or_two_observation": {
"total": 5.959544169832952,
"count": 181950,
"is_parallel": true,
"self": 5.959544169832952
}
}
}
}
}
}
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.707699993858114e-05,
"count": 1,
"self": 1.707699993858114e-05,
"children": {
"thread_root": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"trainer_advance": {
"total": 487.485358452177,
"count": 1349044,
"is_parallel": true,
"self": 13.953153022906918,
"children": {
"process_trajectory": {
"total": 341.6691653822927,
"count": 1349044,
"is_parallel": true,
"self": 341.39889234829025,
"children": {
"RLTrainer._checkpoint": {
"total": 0.2702730340024573,
"count": 4,
"is_parallel": true,
"self": 0.2702730340024573
}
}
},
"_update_policy": {
"total": 131.8630400469774,
"count": 90,
"is_parallel": true,
"self": 13.161599861756258,
"children": {
"TorchPPOOptimizer.update": {
"total": 118.70144018522114,
"count": 4587,
"is_parallel": true,
"self": 118.70144018522114
}
}
}
}
}
}
}
}
},
"TrainerController._save_models": {
"total": 0.059825080999871716,
"count": 1,
"self": 0.0008039809981710277,
"children": {
"RLTrainer._checkpoint": {
"total": 0.05902110000170069,
"count": 1,
"self": 0.05902110000170069
}
}
}
}
}
}
}