Unity 机器学习代理工具包 (ML-Agents) 是一个开源项目,它使游戏和模拟能够作为训练智能代理的环境。
您最多选择25个主题 主题必须以中文或者字母或数字开头,可以包含连字符 (-),并且长度不得超过35个字符
 
 
 
 
 

298 行
16 KiB

{
"name": "root",
"gauges": {
"WormDynamic.Policy.Entropy.mean": {
"value": 1.3626095056533813,
"min": 1.3626091480255127,
"max": 1.3773400783538818,
"count": 48
},
"WormDynamic.Environment.EpisodeLength.mean": {
"value": 999.0,
"min": 999.0,
"max": 999.0,
"count": 48
},
"WormDynamic.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.8739488124847412,
"min": 0.30814048647880554,
"max": 0.8840816020965576,
"count": 48
},
"WormDynamic.Environment.CumulativeReward.mean": {
"value": 4.433333333333334,
"min": 2.0,
"max": 4.733333333333333,
"count": 48
},
"WormDynamic.Policy.ExtrinsicReward.mean": {
"value": 4.433333333333334,
"min": 2.0,
"max": 4.733333333333333,
"count": 48
},
"WormDynamic.IsTraining.mean": {
"value": 0.0,
"min": 0.0,
"max": 1.0,
"count": 48
},
"WormDynamic.Losses.ValueLoss.mean": {
"value": 0.018619487062096596,
"min": 0.007740612607449293,
"max": 0.018619487062096596,
"count": 47
},
"WormDynamic.Losses.PolicyLoss.mean": {
"value": 0.019054416567087173,
"min": 0.01428273506462574,
"max": 0.02198072336614132,
"count": 47
},
"WormDynamic.Policy.LearningRate.mean": {
"value": 1.7143888726423029e-06,
"min": 1.7143888726423029e-06,
"max": 0.00012000006245216355,
"count": 47
},
"WormDynamic.Policy.Epsilon.mean": {
"value": 0.10057143867015839,
"min": 0.10057143867015839,
"max": 0.13999998569488525,
"count": 47
},
"WormDynamic.Policy.Beta.mean": {
"value": 3.851434667012654e-05,
"min": 3.851434667012654e-05,
"max": 0.002005999907851219,
"count": 47
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1593749147",
"python_version": "3.6.4 (default, Mar 1 2018, 18:36:42) \n[GCC 4.2.1 Compatible Apple LLVM 9.0.0 (clang-900.0.39.2)]",
"command_line_arguments": "/Users/brandonh/unity_projects/ml-agents-master/ml-agents/venv/bin/mlagents-learn config/ppo/WormDynamic.yaml --run-id=WormDynamic2 --resume",
"mlagents_version": "0.18.0.dev0",
"mlagents_envs_version": "0.18.0.dev0",
"communication_protocol_version": "1.0.0",
"tensorflow_version": "2.2.0",
"end_time_seconds": "1593751348"
},
"total": 2200.9439116270514,
"count": 1,
"self": 0.0248973990092054,
"children": {
"run_training.setup": {
"total": 0.01232337299734354,
"count": 1,
"self": 0.01232337299734354
},
"TrainerController.start_learning": {
"total": 2200.906690855045,
"count": 1,
"self": 80.82224849390332,
"children": {
"TrainerController._reset_env": {
"total": 13.719661066075787,
"count": 1,
"self": 13.719661066075787
},
"TrainerController.advance": {
"total": 2106.153355687973,
"count": 144001,
"self": 2.0326165510341525,
"children": {
"env_step": {
"total": 2104.120739136939,
"count": 144001,
"self": 1871.5352950636297,
"children": {
"SubprocessEnvManager._take_step": {
"total": 230.03781107312534,
"count": 144001,
"self": 6.187276195501909,
"children": {
"NNPolicy.evaluate": {
"total": 223.85053487762343,
"count": 144001,
"self": 223.85053487762343
}
}
},
"workers": {
"total": 2.5476330001838505,
"count": 144001,
"self": 0.0,
"children": {
"worker_root": {
"total": 2194.3333130943356,
"count": 144001,
"is_parallel": true,
"self": 615.1715003795689,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.00180063396692276,
"count": 1,
"is_parallel": true,
"self": 0.00022585701663047075,
"children": {
"_process_vector_observation": {
"total": 0.0015747769502922893,
"count": 2,
"is_parallel": true,
"self": 0.0015747769502922893
}
}
},
"UnityEnvironment.step": {
"total": 0.023117784061469138,
"count": 1,
"is_parallel": true,
"self": 0.00016248517204076052,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.000196726992726326,
"count": 1,
"is_parallel": true,
"self": 0.000196726992726326
},
"communicator.exchange": {
"total": 0.02229075797367841,
"count": 1,
"is_parallel": true,
"self": 0.02229075797367841
},
"steps_from_proto": {
"total": 0.0004678139230236411,
"count": 1,
"is_parallel": true,
"self": 0.00011108280159533024,
"children": {
"_process_vector_observation": {
"total": 0.00035673112142831087,
"count": 2,
"is_parallel": true,
"self": 0.00035673112142831087
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1579.1618127147667,
"count": 144000,
"is_parallel": true,
"self": 22.52040944120381,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 23.696352752973326,
"count": 144000,
"is_parallel": true,
"self": 23.696352752973326
},
"communicator.exchange": {
"total": 1475.0952947261976,
"count": 144000,
"is_parallel": true,
"self": 1475.0952947261976
},
"steps_from_proto": {
"total": 57.84975579439197,
"count": 144000,
"is_parallel": true,
"self": 15.751152550568804,
"children": {
"_process_vector_observation": {
"total": 42.09860324382316,
"count": 288000,
"is_parallel": true,
"self": 42.09860324382316
}
}
}
}
}
}
}
}
}
}
}
}
},
"trainer_threads": {
"total": 3.58381075784564e-05,
"count": 1,
"self": 3.58381075784564e-05,
"children": {
"thread_root": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"trainer_advance": {
"total": 2122.309396725381,
"count": 7082971,
"is_parallel": true,
"self": 121.44299733371008,
"children": {
"process_trajectory": {
"total": 1559.3746680220356,
"count": 7082971,
"is_parallel": true,
"self": 1558.636247553979,
"children": {
"Trainer.save_model": {
"total": 0.7384204680565745,
"count": 3,
"is_parallel": true,
"self": 0.7384204680565745
}
}
},
"_update_policy": {
"total": 441.4917313696351,
"count": 47,
"is_parallel": true,
"self": 160.25679189502262,
"children": {
"PPOOptimizer.update": {
"total": 281.2349394746125,
"count": 1974,
"is_parallel": true,
"self": 281.2349394746125
}
}
}
}
}
}
}
}
},
"TrainerController._save_model": {
"total": 0.2113897689851001,
"count": 1,
"self": 0.0001797359436750412,
"children": {
"Trainer.save_model": {
"total": 0.21121003304142505,
"count": 1,
"self": 0.21121003304142505
}
}
}
}
}
}
}