First Push`

README.md

@@ -0,0 +1,35 @@
+---
+library_name: ml-agents
+tags:
+- SoccerTwos
+- deep-reinforcement-learning
+- reinforcement-learning
+- ML-Agents-SoccerTwos
+---
+
+  # **poca** Agent playing **SoccerTwos**
+  This is a trained model of a **poca** agent playing **SoccerTwos**
+  using the [Unity ML-Agents Library](https://github.com/Unity-Technologies/ml-agents).
+
+  ## Usage (with ML-Agents)
+  The Documentation: https://unity-technologies.github.io/ml-agents/ML-Agents-Toolkit-Documentation/
+
+  We wrote a complete tutorial to learn to train your first agent using ML-Agents and publish it to the Hub:
+  - A *short tutorial* where you teach Huggy the Dog 🐶 to fetch the stick and then play with him directly in your
+  browser: https://huggingface.co/learn/deep-rl-course/unitbonus1/introduction
+  - A *longer tutorial* to understand how works ML-Agents:
+  https://huggingface.co/learn/deep-rl-course/unit5/introduction
+
+  ### Resume the training
+  ```bash
+  mlagents-learn <your_configuration_file_path.yaml> --run-id=<run_id> --resume
+  ```
+
+  ### Watch your Agent play
+  You can watch your agent **playing directly in your browser**
+
+  1. If the environment is part of ML-Agents official environments, go to https://huggingface.co/unity
+  2. Step 1: Find your model_id: c-wang/poca-SoccerTwos
+  3. Step 2: Select your *.nn /*.onnx file
+  4. Click on Watch the agent play 👀
+  

SoccerTwos.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f321450ad9481edaf7f87f4716c8ccada6e005ab01fc9fc97ab6fd138b6640fc
+size 1768747

SoccerTwos/SoccerTwos-48499860.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b04f058401f8501090dc3b95eb31ab59a1645396ca09f49511fb43e749df278b
+size 1768747

SoccerTwos/SoccerTwos-48499860.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:413a0ff8e42509d47f2badebeec0a4517f9a3ea5c8e85d67ff1f5653568bc8c2
+size 28430482

SoccerTwos/SoccerTwos-48999946.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1aaf2d9235e221b84863575e87ec14a022f9e1bb504217581aaa1372630a3efc
+size 1768747

SoccerTwos/SoccerTwos-48999946.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2cee7956ce580832f6ba70c4f6e81af49829dd291c8a3843dc25e970c084cf81
+size 28430482

SoccerTwos/SoccerTwos-49499990.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3182ef97e5b67a2227bb0681e09aabf86f8313c94558ba0661cafa64e69a2951
+size 1768747

SoccerTwos/SoccerTwos-49499990.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f06c2eab20d0bf857fba94021311988ee152187f4c0424f43b3a2f2bee636536
+size 28430482

SoccerTwos/SoccerTwos-49999865.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f321450ad9481edaf7f87f4716c8ccada6e005ab01fc9fc97ab6fd138b6640fc
+size 1768747

SoccerTwos/SoccerTwos-49999865.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:560755c0c8edcbba87363929d19efe0d2bc703caec8d4163341caaa548c3aaf5
+size 28430482

SoccerTwos/SoccerTwos-50000036.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f321450ad9481edaf7f87f4716c8ccada6e005ab01fc9fc97ab6fd138b6640fc
+size 1768747

SoccerTwos/SoccerTwos-50000036.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a29b40232a13bd6ee815f27bc9c36949ea019ebfb779bb8fb6492b87b5a049fe
+size 28430482

SoccerTwos/checkpoint.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0548f096aa212ec3fd3a159569beb8a6d425e10e70e642218a415161da955e69
+size 28429366

SoccerTwos/events.out.tfevents.1703464607.DESKTOP-8RQVDLC.17120.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb40c5d2c351e8c256dbb6f47f022aa3ab2faff369f809eeb3c725c521069a64
+size 27741948

config.json

@@ -0,0 +1 @@
+{"default_settings": null, "behaviors": {"SoccerTwos": {"trainer_type": "poca", "hyperparameters": {"batch_size": 2048, "buffer_size": 20480, "learning_rate": 0.0003, "beta": 0.005, "epsilon": 0.2, "lambd": 0.95, "num_epoch": 3, "learning_rate_schedule": "constant", "beta_schedule": "constant", "epsilon_schedule": "constant"}, "checkpoint_interval": 500000, "network_settings": {"normalize": false, "hidden_units": 512, "num_layers": 2, "vis_encode_type": "simple", "memory": null, "goal_conditioning_type": "hyper", "deterministic": false}, "reward_signals": {"extrinsic": {"gamma": 0.99, "strength": 1.0, "network_settings": {"normalize": false, "hidden_units": 128, "num_layers": 2, "vis_encode_type": "simple", "memory": null, "goal_conditioning_type": "hyper", "deterministic": false}}}, "init_path": null, "keep_checkpoints": 5, "even_checkpoints": false, "max_steps": 50000000, "time_horizon": 1000, "summary_freq": 10000, "threaded": false, "self_play": {"save_steps": 50000, "team_change": 200000, "swap_steps": 2000, "window": 10, "play_against_latest_model_ratio": 0.5, "initial_elo": 1200.0}, "behavioral_cloning": null}}, "env_settings": {"env_path": "./training-envs-executables/SoccerTwos/SoccerTwos.exe", "env_args": null, "base_port": 5005, "num_envs": 1, "num_areas": 1, "timeout_wait": 60, "seed": -1, "max_lifetime_restarts": 10, "restarts_rate_limit_n": 1, "restarts_rate_limit_period_s": 60}, "engine_settings": {"width": 84, "height": 84, "quality_level": 5, "time_scale": 20, "target_frame_rate": -1, "capture_frame_rate": 60, "no_graphics": true, "no_graphics_monitor": false}, "environment_parameters": null, "checkpoint_settings": {"run_id": "SoccerTwos", "initialize_from": null, "load_model": false, "resume": false, "force": true, "train_model": false, "inference": false, "results_dir": "results"}, "torch_settings": {"device": null}, "debug": false}

configuration.yaml

@@ -0,0 +1,84 @@
+default_settings: null
+behaviors:
+  SoccerTwos:
+    trainer_type: poca
+    hyperparameters:
+      batch_size: 2048
+      buffer_size: 20480
+      learning_rate: 0.0003
+      beta: 0.005
+      epsilon: 0.2
+      lambd: 0.95
+      num_epoch: 3
+      learning_rate_schedule: constant
+      beta_schedule: constant
+      epsilon_schedule: constant
+    checkpoint_interval: 500000
+    network_settings:
+      normalize: false
+      hidden_units: 512
+      num_layers: 2
+      vis_encode_type: simple
+      memory: null
+      goal_conditioning_type: hyper
+      deterministic: false
+    reward_signals:
+      extrinsic:
+        gamma: 0.99
+        strength: 1.0
+        network_settings:
+          normalize: false
+          hidden_units: 128
+          num_layers: 2
+          vis_encode_type: simple
+          memory: null
+          goal_conditioning_type: hyper
+          deterministic: false
+    init_path: null
+    keep_checkpoints: 5
+    even_checkpoints: false
+    max_steps: 50000000
+    time_horizon: 1000
+    summary_freq: 10000
+    threaded: false
+    self_play:
+      save_steps: 50000
+      team_change: 200000
+      swap_steps: 2000
+      window: 10
+      play_against_latest_model_ratio: 0.5
+      initial_elo: 1200.0
+    behavioral_cloning: null
+env_settings:
+  env_path: ./training-envs-executables/SoccerTwos/SoccerTwos.exe
+  env_args: null
+  base_port: 5005
+  num_envs: 1
+  num_areas: 1
+  timeout_wait: 60
+  seed: -1
+  max_lifetime_restarts: 10
+  restarts_rate_limit_n: 1
+  restarts_rate_limit_period_s: 60
+engine_settings:
+  width: 84
+  height: 84
+  quality_level: 5
+  time_scale: 20
+  target_frame_rate: -1
+  capture_frame_rate: 60
+  no_graphics: true
+  no_graphics_monitor: false
+environment_parameters: null
+checkpoint_settings:
+  run_id: SoccerTwos
+  initialize_from: null
+  load_model: false
+  resume: false
+  force: true
+  train_model: false
+  inference: false
+  results_dir: results
+torch_settings:
+  device: null
+debug: false

run_logs/timers.json

@@ -0,0 +1,368 @@
+{
+    "name": "root",
+    "gauges": {
+        "SoccerTwos.Policy.Entropy.mean": {
+            "value": 1.775998592376709,
+            "min": 1.6040146350860596,
+            "max": 3.2957441806793213,
+            "count": 5000
+        },
+        "SoccerTwos.Policy.Entropy.sum": {
+            "value": 35747.30078125,
+            "min": 5111.42138671875,
+            "max": 138394.046875,
+            "count": 5000
+        },
+        "SoccerTwos.Environment.EpisodeLength.mean": {
+            "value": 82.56666666666666,
+            "min": 55.28888888888889,
+            "max": 999.0,
+            "count": 5000
+        },
+        "SoccerTwos.Environment.EpisodeLength.sum": {
+            "value": 19816.0,
+            "min": 15564.0,
+            "max": 24952.0,
+            "count": 5000
+        },
+        "SoccerTwos.Self-play.ELO.mean": {
+            "value": 1573.510181424287,
+            "min": 1176.3367770620202,
+            "max": 1643.2444482422695,
+            "count": 2891
+        },
+        "SoccerTwos.Self-play.ELO.sum": {
+            "value": 188821.22177091445,
+            "min": 2354.063791122488,
+            "max": 291336.48131911457,
+            "count": 2891
+        },
+        "SoccerTwos.Step.mean": {
+            "value": 49999865.0,
+            "min": 9498.0,
+            "max": 49999865.0,
+            "count": 5000
+        },
+        "SoccerTwos.Step.sum": {
+            "value": 49999865.0,
+            "min": 9498.0,
+            "max": 49999865.0,
+            "count": 5000
+        },
+        "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": {
+            "value": -0.06021175906062126,
+            "min": -0.29010486602783203,
+            "max": 0.12045252323150635,
+            "count": 5000
+        },
+        "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": {
+            "value": -7.2254109382629395,
+            "min": -18.919986724853516,
+            "max": 9.369237899780273,
+            "count": 5000
+        },
+        "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": {
+            "value": -0.05843336135149002,
+            "min": -0.29126614332199097,
+            "max": 0.12659993767738342,
+            "count": 5000
+        },
+        "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": {
+            "value": -7.012003421783447,
+            "min": -19.16312026977539,
+            "max": 9.595020294189453,
+            "count": 5000
+        },
+        "SoccerTwos.Environment.CumulativeReward.mean": {
+            "value": 0.0,
+            "min": 0.0,
+            "max": 0.0,
+            "count": 5000
+        },
+        "SoccerTwos.Environment.CumulativeReward.sum": {
+            "value": 0.0,
+            "min": 0.0,
+            "max": 0.0,
+            "count": 5000
+        },
+        "SoccerTwos.Policy.ExtrinsicReward.mean": {
+            "value": -0.06249999751647314,
+            "min": -0.8743764705517713,
+            "max": 0.6009942889213562,
+            "count": 5000
+        },
+        "SoccerTwos.Policy.ExtrinsicReward.sum": {
+            "value": -7.499999701976776,
+            "min": -55.73639988899231,
+            "max": 43.6576002240181,
+            "count": 5000
+        },
+        "SoccerTwos.Environment.GroupCumulativeReward.mean": {
+            "value": -0.06249999751647314,
+            "min": -0.8743764705517713,
+            "max": 0.6009942889213562,
+            "count": 5000
+        },
+        "SoccerTwos.Environment.GroupCumulativeReward.sum": {
+            "value": -7.499999701976776,
+            "min": -55.73639988899231,
+            "max": 43.6576002240181,
+            "count": 5000
+        },
+        "SoccerTwos.IsTraining.mean": {
+            "value": 1.0,
+            "min": 1.0,
+            "max": 1.0,
+            "count": 5000
+        },
+        "SoccerTwos.IsTraining.sum": {
+            "value": 1.0,
+            "min": 1.0,
+            "max": 1.0,
+            "count": 5000
+        },
+        "SoccerTwos.Losses.PolicyLoss.mean": {
+            "value": 0.016774558438919486,
+            "min": 0.010072278234777817,
+            "max": 0.025630577544992168,
+            "count": 2341
+        },
+        "SoccerTwos.Losses.PolicyLoss.sum": {
+            "value": 0.016774558438919486,
+            "min": 0.010072278234777817,
+            "max": 0.025630577544992168,
+            "count": 2341
+        },
+        "SoccerTwos.Losses.ValueLoss.mean": {
+            "value": 0.09155314341187477,
+            "min": 3.240071249493992e-11,
+            "max": 0.09859520792961121,
+            "count": 2341
+        },
+        "SoccerTwos.Losses.ValueLoss.sum": {
+            "value": 0.09155314341187477,
+            "min": 3.240071249493992e-11,
+            "max": 0.09859520792961121,
+            "count": 2341
+        },
+        "SoccerTwos.Losses.BaselineLoss.mean": {
+            "value": 0.09291486019889514,
+            "min": 4.150054478309813e-11,
+            "max": 0.10025302891929945,
+            "count": 2341
+        },
+        "SoccerTwos.Losses.BaselineLoss.sum": {
+            "value": 0.09291486019889514,
+            "min": 4.150054478309813e-11,
+            "max": 0.10025302891929945,
+            "count": 2341
+        },
+        "SoccerTwos.Policy.LearningRate.mean": {
+            "value": 0.0003,
+            "min": 0.0003,
+            "max": 0.0003,
+            "count": 2341
+        },
+        "SoccerTwos.Policy.LearningRate.sum": {
+            "value": 0.0003,
+            "min": 0.0003,
+            "max": 0.0003,
+            "count": 2341
+        },
+        "SoccerTwos.Policy.Epsilon.mean": {
+            "value": 0.20000000000000007,
+            "min": 0.20000000000000007,
+            "max": 0.20000000000000007,
+            "count": 2341
+        },
+        "SoccerTwos.Policy.Epsilon.sum": {
+            "value": 0.20000000000000007,
+            "min": 0.20000000000000007,
+            "max": 0.20000000000000007,
+            "count": 2341
+        },
+        "SoccerTwos.Policy.Beta.mean": {
+            "value": 0.005000000000000001,
+            "min": 0.005000000000000001,
+            "max": 0.005000000000000001,
+            "count": 2341
+        },
+        "SoccerTwos.Policy.Beta.sum": {
+            "value": 0.005000000000000001,
+            "min": 0.005000000000000001,
+            "max": 0.005000000000000001,
+            "count": 2341
+        }
+    },
+    "metadata": {
+        "timer_format_version": "0.1.0",
+        "start_time_seconds": "1703464603",
+        "python_version": "3.10.12 | packaged by Anaconda, Inc. | (main, Jul  5 2023, 19:01:18) [MSC v.1916 64 bit (AMD64)]",
+        "command_line_arguments": "\\\\?\\C:\\Users\\chang\\anaconda3\\envs\\rl\\Scripts\\mlagents-learn ./config/poca/SoccerTwos.yaml --env=./training-envs-executables/SoccerTwos/SoccerTwos.exe --run-id=SoccerTwos --no-graphics --force",
+        "mlagents_version": "1.1.0.dev0",
+        "mlagents_envs_version": "1.1.0.dev0",
+        "communication_protocol_version": "1.5.0",
+        "pytorch_version": "2.1.2+cpu",
+        "numpy_version": "1.23.5",
+        "end_time_seconds": "1703627288"
+    },
+    "total": 162713.8910461,
+    "count": 1,
+    "self": 2.877047700021649,
+    "children": {
+        "run_training.setup": {
+            "total": 0.10487579999971786,
+            "count": 1,
+            "self": 0.10487579999971786
+        },
+        "TrainerController.start_learning": {
+            "total": 162710.9091226,
+            "count": 1,
+            "self": 95.03098711458733,
+            "children": {
+                "TrainerController._reset_env": {
+                    "total": 22.870975599962094,
+                    "count": 250,
+                    "self": 22.870975599962094
+                },
+                "TrainerController.advance": {
+                    "total": 162592.78000478543,
+                    "count": 3285140,
+                    "self": 99.11501299220254,
+                    "children": {
+                        "env_step": {
+                            "total": 66458.05725349297,
+                            "count": 3285140,
+                            "self": 49901.25500668623,
+                            "children": {
+                                "SubprocessEnvManager._take_step": {
+                                    "total": 16496.2410986158,
+                                    "count": 3285140,
+                                    "self": 579.729718210041,
+                                    "children": {
+                                        "TorchPolicy.evaluate": {
+                                            "total": 15916.511380405758,
+                                            "count": 6413848,
+                                            "self": 15916.511380405758
+                                        }
+                                    }
+                                },
+                                "workers": {
+                                    "total": 60.56114819093136,
+                                    "count": 3285140,
+                                    "self": 0.0,
+                                    "children": {
+                                        "worker_root": {
+                                            "total": 162593.12074150195,
+                                            "count": 3285140,
+                                            "is_parallel": true,
+                                            "self": 124623.27507769279,
+                                            "children": {
+                                                "steps_from_proto": {
+                                                    "total": 0.5632270000055541,
+                                                    "count": 500,
+                                                    "is_parallel": true,
+                                                    "self": 0.10653560005812324,
+                                                    "children": {
+                                                        "_process_rank_one_or_two_observation": {
+                                                            "total": 0.4566913999474309,
+                                                            "count": 2000,
+                                                            "is_parallel": true,
+                                                            "self": 0.4566913999474309
+                                                        }
+                                                    }
+                                                },
+                                                "UnityEnvironment.step": {
+                                                    "total": 37969.28243680915,
+                                                    "count": 3285140,
+                                                    "is_parallel": true,
+                                                    "self": 2018.8212593716598,
+                                                    "children": {
+                                                        "UnityEnvironment._generate_step_input": {
+                                                            "total": 2098.5394663058187,
+                                                            "count": 3285140,
+                                                            "is_parallel": true,
+                                                            "self": 2098.5394663058187
+                                                        },
+                                                        "communicator.exchange": {
+                                                            "total": 26754.034175819048,
+                                                            "count": 3285140,
+                                                            "is_parallel": true,
+                                                            "self": 26754.034175819048
+                                                        },
+                                                        "steps_from_proto": {
+                                                            "total": 7097.887535312628,
+                                                            "count": 6570280,
+                                                            "is_parallel": true,
+                                                            "self": 1349.507690489363,
+                                                            "children": {
+                                                                "_process_rank_one_or_two_observation": {
+                                                                    "total": 5748.379844823265,
+                                                                    "count": 26281120,
+                                                                    "is_parallel": true,
+                                                                    "self": 5748.379844823265
+                                                                }
+                                                            }
+                                                        }
+                                                    }
+                                                }
+                                            }
+                                        }
+                                    }
+                                }
+                            }
+                        },
+                        "trainer_advance": {
+                            "total": 96035.60773830024,
+                            "count": 3285140,
+                            "self": 705.0338465059904,
+                            "children": {
+                                "process_trajectory": {
+                                    "total": 13425.446054794105,
+                                    "count": 3285140,
+                                    "self": 13407.045061593923,
+                                    "children": {
+                                        "RLTrainer._checkpoint": {
+                                            "total": 18.40099320018271,
+                                            "count": 100,
+                                            "self": 18.40099320018271
+                                        }
+                                    }
+                                },
+                                "_update_policy": {
+                                    "total": 81905.12783700015,
+                                    "count": 2341,
+                                    "self": 8293.239140300793,
+                                    "children": {
+                                        "TorchPOCAOptimizer.update": {
+                                            "total": 73611.88869669936,
+                                            "count": 70230,
+                                            "self": 73611.88869669936
+                                        }
+                                    }
+                                }
+                            }
+                        }
+                    }
+                },
+                "trainer_threads": {
+                    "total": 9.00006853044033e-07,
+                    "count": 1,
+                    "self": 9.00006853044033e-07
+                },
+                "TrainerController._save_models": {
+                    "total": 0.227154199994402,
+                    "count": 1,
+                    "self": 0.05800160000217147,
+                    "children": {
+                        "RLTrainer._checkpoint": {
+                            "total": 0.16915259999223053,
+                            "count": 1,
+                            "self": 0.16915259999223053
+                        }
+                    }
+                }
+            }
+        }
+    }
+}

run_logs/training_status.json

@@ -0,0 +1,66 @@
+{
+    "SoccerTwos": {
+        "checkpoints": [
+            {
+                "steps": 48499860,
+                "file_path": "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-48499860.onnx",
+                "reward": 0.0,
+                "creation_time": 1703621868.135585,
+                "auxillary_file_paths": [
+                    "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-48499860.pt"
+                ]
+            },
+            {
+                "steps": 48999946,
+                "file_path": "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-48999946.onnx",
+                "reward": 0.0,
+                "creation_time": 1703623713.5747406,
+                "auxillary_file_paths": [
+                    "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-48999946.pt"
+                ]
+            },
+            {
+                "steps": 49499990,
+                "file_path": "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-49499990.onnx",
+                "reward": 0.0,
+                "creation_time": 1703625514.037782,
+                "auxillary_file_paths": [
+                    "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-49499990.pt"
+                ]
+            },
+            {
+                "steps": 49999865,
+                "file_path": "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-49999865.onnx",
+                "reward": 0.0,
+                "creation_time": 1703627285.4809556,
+                "auxillary_file_paths": [
+                    "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-49999865.pt"
+                ]
+            },
+            {
+                "steps": 50000036,
+                "file_path": "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-50000036.onnx",
+                "reward": 0.0,
+                "creation_time": 1703627285.704455,
+                "auxillary_file_paths": [
+                    "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-50000036.pt"
+                ]
+            }
+        ],
+        "elo": 1571.5942980926663,
+        "final_checkpoint": {
+            "steps": 50000036,
+            "file_path": "results\\SoccerTwos\\SoccerTwos.onnx",
+            "reward": 0.0,
+            "creation_time": 1703627285.704455,
+            "auxillary_file_paths": [
+                "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-50000036.pt"
+            ]
+        }
+    },
+    "metadata": {
+        "stats_format_version": "0.3.0",
+        "mlagents_version": "1.1.0.dev0",
+        "torch_version": "2.1.2+cpu"
+    }
+}