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: VitCon/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:b64e1387900dd8524354562dea8a98e5acf03d247f9deac73bc9687a1642bfb1
+size 1768747

SoccerTwos/SoccerTwos-48499970.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dbcfd011624e8b59381d61efe62ccd2d2b2e69c9a9ae9867f9f31a3dd20e91eb
+size 1768747

SoccerTwos/SoccerTwos-48499970.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bbd938ecab7054383fc2f2ad1f2e3ce2d8999b17d4771b7277e5890311e565cc
+size 28431689

SoccerTwos/SoccerTwos-48999956.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6abf32f698130ac36be908c5bfd371a69b83953cdab9d3e7484df33b2f308fa6
+size 1768747

SoccerTwos/SoccerTwos-48999956.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7cb466ea4a0d8bc11a0859e296fdd647de316c4b5ffe44a0926ca86c236e5436
+size 28431689

SoccerTwos/SoccerTwos-49499990.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aca1bd5c6cc2672d32ef5b2eaa90fe58488c70a285526956fb26cc41e14ca1c6
+size 1768747

SoccerTwos/SoccerTwos-49499990.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8515302c8e92a2c87251252c80967b27879c98afbf361e084ed3ae6fab85fddd
+size 28431689

SoccerTwos/SoccerTwos-49999941.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b64e1387900dd8524354562dea8a98e5acf03d247f9deac73bc9687a1642bfb1
+size 1768747

SoccerTwos/SoccerTwos-49999941.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0863e58d5623beae495ff3d9f5f139dcea0d3a621b3624dc3ac3ade94dc45aee
+size 28431689

SoccerTwos/SoccerTwos-50000024.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b64e1387900dd8524354562dea8a98e5acf03d247f9deac73bc9687a1642bfb1
+size 1768747

SoccerTwos/SoccerTwos-50000024.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b57c9c5ad0075b88e43ca3bd2b43ecdd8450ce0595eb2968a5fa6cfcf551400d
+size 28431689

SoccerTwos/checkpoint.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76c8b59ee77f634c139f4871c799369b4dcd0c670adbb3d2fa12d5de05309395
+size 28430591

SoccerTwos/events.out.tfevents.1689125757.VitCon.20372.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:294a5c5a20a853710927787f0cd625442b4e2c49684dd8345763c763e4b3a961
+size 17862475

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.exe", "env_args": null, "base_port": 5005, "num_envs": 1, "num_areas": 1, "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}, "environment_parameters": null, "checkpoint_settings": {"run_id": "SoccerTwos", "initialize_from": null, "load_model": false, "resume": false, "force": false, "train_model": false, "inference": false, "results_dir": "results"}, "torch_settings": {"device": null}, "debug": false}

configuration.yaml

@@ -0,0 +1,82 @@
+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.exe
+  env_args: null
+  base_port: 5005
+  num_envs: 1
+  num_areas: 1
+  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
+environment_parameters: null
+checkpoint_settings:
+  run_id: SoccerTwos
+  initialize_from: null
+  load_model: false
+  resume: false
+  force: false
+  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": 1.8859961032867432,
+            "min": 0.7960292100906372,
+            "max": 3.2957451343536377,
+            "count": 5000
+        },
+        "SoccerTwos.Policy.Entropy.sum": {
+            "value": 37056.05078125,
+            "min": 10420.7392578125,
+            "max": 160892.15625,
+            "count": 5000
+        },
+        "SoccerTwos.Environment.EpisodeLength.mean": {
+            "value": 61.69230769230769,
+            "min": 47.300970873786405,
+            "max": 999.0,
+            "count": 5000
+        },
+        "SoccerTwos.Environment.EpisodeLength.sum": {
+            "value": 19248.0,
+            "min": 12720.0,
+            "max": 30472.0,
+            "count": 5000
+        },
+        "SoccerTwos.Self-play.ELO.mean": {
+            "value": 1596.9313248102283,
+            "min": 1181.879118812825,
+            "max": 1607.88234923467,
+            "count": 1668
+        },
+        "SoccerTwos.Self-play.ELO.sum": {
+            "value": 249121.2866703956,
+            "min": 2363.75823762565,
+            "max": 320736.99087702495,
+            "count": 1668
+        },
+        "SoccerTwos.Step.mean": {
+            "value": 49999941.0,
+            "min": 9252.0,
+            "max": 49999941.0,
+            "count": 5000
+        },
+        "SoccerTwos.Step.sum": {
+            "value": 49999941.0,
+            "min": 9252.0,
+            "max": 49999941.0,
+            "count": 5000
+        },
+        "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": {
+            "value": -0.0425206795334816,
+            "min": -0.12837325036525726,
+            "max": 0.15666325390338898,
+            "count": 5000
+        },
+        "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": {
+            "value": -6.675746917724609,
+            "min": -18.100627899169922,
+            "max": 27.25940704345703,
+            "count": 5000
+        },
+        "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": {
+            "value": -0.0382288359105587,
+            "min": -0.1278502345085144,
+            "max": 0.16222737729549408,
+            "count": 5000
+        },
+        "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": {
+            "value": -6.001927375793457,
+            "min": -18.455331802368164,
+            "max": 28.227563858032227,
+            "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": -0.30523311864038943,
+            "min": -0.6923076923076923,
+            "max": 0.42062222295337254,
+            "count": 5000
+        },
+        "SoccerTwos.Policy.ExtrinsicReward.sum": {
+            "value": -47.92159962654114,
+            "min": -47.92159962654114,
+            "max": 47.269000351428986,
+            "count": 5000
+        },
+        "SoccerTwos.Environment.GroupCumulativeReward.mean": {
+            "value": -0.30523311864038943,
+            "min": -0.6923076923076923,
+            "max": 0.42062222295337254,
+            "count": 5000
+        },
+        "SoccerTwos.Environment.GroupCumulativeReward.sum": {
+            "value": -47.92159962654114,
+            "min": -47.92159962654114,
+            "max": 47.269000351428986,
+            "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": 0.014865773305064067,
+            "min": 0.009734752150446486,
+            "max": 0.02486639428922596,
+            "count": 2314
+        },
+        "SoccerTwos.Losses.PolicyLoss.sum": {
+            "value": 0.014865773305064067,
+            "min": 0.009734752150446486,
+            "max": 0.02486639428922596,
+            "count": 2314
+        },
+        "SoccerTwos.Losses.ValueLoss.mean": {
+            "value": 0.09687695155541103,
+            "min": 5.549574153806044e-12,
+            "max": 0.10573716560999552,
+            "count": 2314
+        },
+        "SoccerTwos.Losses.ValueLoss.sum": {
+            "value": 0.09687695155541103,
+            "min": 5.549574153806044e-12,
+            "max": 0.10573716560999552,
+            "count": 2314
+        },
+        "SoccerTwos.Losses.BaselineLoss.mean": {
+            "value": 0.09823337495326996,
+            "min": 8.280764818060135e-12,
+            "max": 0.10774662320812543,
+            "count": 2314
+        },
+        "SoccerTwos.Losses.BaselineLoss.sum": {
+            "value": 0.09823337495326996,
+            "min": 8.280764818060135e-12,
+            "max": 0.10774662320812543,
+            "count": 2314
+        },
+        "SoccerTwos.Policy.LearningRate.mean": {
+            "value": 0.0003,
+            "min": 0.0003,
+            "max": 0.0003,
+            "count": 2314
+        },
+        "SoccerTwos.Policy.LearningRate.sum": {
+            "value": 0.0003,
+            "min": 0.0003,
+            "max": 0.0003,
+            "count": 2314
+        },
+        "SoccerTwos.Policy.Epsilon.mean": {
+            "value": 0.20000000000000007,
+            "min": 0.20000000000000007,
+            "max": 0.20000000000000007,
+            "count": 2314
+        },
+        "SoccerTwos.Policy.Epsilon.sum": {
+            "value": 0.20000000000000007,
+            "min": 0.20000000000000007,
+            "max": 0.20000000000000007,
+            "count": 2314
+        },
+        "SoccerTwos.Policy.Beta.mean": {
+            "value": 0.005000000000000001,
+            "min": 0.005000000000000001,
+            "max": 0.005000000000000001,
+            "count": 2314
+        },
+        "SoccerTwos.Policy.Beta.sum": {
+            "value": 0.005000000000000001,
+            "min": 0.005000000000000001,
+            "max": 0.005000000000000001,
+            "count": 2314
+        }
+    },
+    "metadata": {
+        "timer_format_version": "0.1.0",
+        "start_time_seconds": "1689125754",
+        "python_version": "3.9.13 (tags/v3.9.13:6de2ca5, May 17 2022, 16:36:42) [MSC v.1929 64 bit (AMD64)]",
+        "command_line_arguments": "C:\\Users\\vitco\\Desktop\\SoccerTwos\\env\\Scripts\\mlagents-learn ./config/poca/SoccerTwos.yaml --env=./training-envs-executables/SoccerTwos.exe --run-id=SoccerTwos --no-graphics",
+        "mlagents_version": "0.31.0.dev0",
+        "mlagents_envs_version": "0.31.0.dev0",
+        "communication_protocol_version": "1.5.0",
+        "pytorch_version": "1.13.1+cu116",
+        "numpy_version": "1.21.2",
+        "end_time_seconds": "1689205231"
+    },
+    "total": 79476.61104039999,
+    "count": 1,
+    "self": 0.7239865999872563,
+    "children": {
+        "run_training.setup": {
+            "total": 0.2028278000000001,
+            "count": 1,
+            "self": 0.2028278000000001
+        },
+        "TrainerController.start_learning": {
+            "total": 79475.684226,
+            "count": 1,
+            "self": 53.68059590390476,
+            "children": {
+                "TrainerController._reset_env": {
+                    "total": 17.12137330000531,
+                    "count": 249,
+                    "self": 17.12137330000531
+                },
+                "TrainerController.advance": {
+                    "total": 79404.7220559961,
+                    "count": 3278979,
+                    "self": 58.7541511883901,
+                    "children": {
+                        "env_step": {
+                            "total": 56214.57777440655,
+                            "count": 3278979,
+                            "self": 34536.39524380873,
+                            "children": {
+                                "SubprocessEnvManager._take_step": {
+                                    "total": 21642.040069300623,
+                                    "count": 3278979,
+                                    "self": 359.92690518921154,
+                                    "children": {
+                                        "TorchPolicy.evaluate": {
+                                            "total": 21282.11316411141,
+                                            "count": 6431088,
+                                            "self": 21282.11316411141
+                                        }
+                                    }
+                                },
+                                "workers": {
+                                    "total": 36.14246129719899,
+                                    "count": 3278979,
+                                    "self": 0.0,
+                                    "children": {
+                                        "worker_root": {
+                                            "total": 79379.58762889986,
+                                            "count": 3278979,
+                                            "is_parallel": true,
+                                            "self": 51420.96723030831,
+                                            "children": {
+                                                "steps_from_proto": {
+                                                    "total": 0.5842672999857501,
+                                                    "count": 498,
+                                                    "is_parallel": true,
+                                                    "self": 0.11798829990786297,
+                                                    "children": {
+                                                        "_process_rank_one_or_two_observation": {
+                                                            "total": 0.46627900007788714,
+                                                            "count": 1992,
+                                                            "is_parallel": true,
+                                                            "self": 0.46627900007788714
+                                                        }
+                                                    }
+                                                },
+                                                "UnityEnvironment.step": {
+                                                    "total": 27958.036131291556,
+                                                    "count": 3278979,
+                                                    "is_parallel": true,
+                                                    "self": 1510.600279377246,
+                                                    "children": {
+                                                        "UnityEnvironment._generate_step_input": {
+                                                            "total": 1552.2331909067102,
+                                                            "count": 3278979,
+                                                            "is_parallel": true,
+                                                            "self": 1552.2331909067102
+                                                        },
+                                                        "communicator.exchange": {
+                                                            "total": 20337.390693502173,
+                                                            "count": 3278979,
+                                                            "is_parallel": true,
+                                                            "self": 20337.390693502173
+                                                        },
+                                                        "steps_from_proto": {
+                                                            "total": 4557.811967505427,
+                                                            "count": 6557958,
+                                                            "is_parallel": true,
+                                                            "self": 949.4219691056983,
+                                                            "children": {
+                                                                "_process_rank_one_or_two_observation": {
+                                                                    "total": 3608.3899983997285,
+                                                                    "count": 26231832,
+                                                                    "is_parallel": true,
+                                                                    "self": 3608.3899983997285
+                                                                }
+                                                            }
+                                                        }
+                                                    }
+                                                }
+                                            }
+                                        }
+                                    }
+                                }
+                            }
+                        },
+                        "trainer_advance": {
+                            "total": 23131.390130401163,
+                            "count": 3278979,
+                            "self": 429.52601460448204,
+                            "children": {
+                                "process_trajectory": {
+                                    "total": 5416.983307796469,
+                                    "count": 3278979,
+                                    "self": 5399.051477096455,
+                                    "children": {
+                                        "RLTrainer._checkpoint": {
+                                            "total": 17.931830700013734,
+                                            "count": 100,
+                                            "self": 17.931830700013734
+                                        }
+                                    }
+                                },
+                                "_update_policy": {
+                                    "total": 17284.880808000213,
+                                    "count": 2314,
+                                    "self": 7338.387878698481,
+                                    "children": {
+                                        "TorchPOCAOptimizer.update": {
+                                            "total": 9946.492929301732,
+                                            "count": 69420,
+                                            "self": 9946.492929301732
+                                        }
+                                    }
+                                }
+                            }
+                        }
+                    }
+                },
+                "trainer_threads": {
+                    "total": 4.999892553314567e-07,
+                    "count": 1,
+                    "self": 4.999892553314567e-07
+                },
+                "TrainerController._save_models": {
+                    "total": 0.16020030000072438,
+                    "count": 1,
+                    "self": 0.011720599999534898,
+                    "children": {
+                        "RLTrainer._checkpoint": {
+                            "total": 0.14847970000118949,
+                            "count": 1,
+                            "self": 0.14847970000118949
+                        }
+                    }
+                }
+            }
+        }
+    }
+}

run_logs/training_status.json

@@ -0,0 +1,66 @@
+{
+    "SoccerTwos": {
+        "checkpoints": [
+            {
+                "steps": 48499970,
+                "file_path": "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-48499970.onnx",
+                "reward": 0.0,
+                "creation_time": 1689203016.1253088,
+                "auxillary_file_paths": [
+                    "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-48499970.pt"
+                ]
+            },
+            {
+                "steps": 48999956,
+                "file_path": "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-48999956.onnx",
+                "reward": 0.0,
+                "creation_time": 1689203754.341808,
+                "auxillary_file_paths": [
+                    "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-48999956.pt"
+                ]
+            },
+            {
+                "steps": 49499990,
+                "file_path": "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-49499990.onnx",
+                "reward": 0.0,
+                "creation_time": 1689204489.358575,
+                "auxillary_file_paths": [
+                    "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-49499990.pt"
+                ]
+            },
+            {
+                "steps": 49999941,
+                "file_path": "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-49999941.onnx",
+                "reward": 0.0,
+                "creation_time": 1689205230.671029,
+                "auxillary_file_paths": [
+                    "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-49999941.pt"
+                ]
+            },
+            {
+                "steps": 50000024,
+                "file_path": "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-50000024.onnx",
+                "reward": 0.0,
+                "creation_time": 1689205230.8380823,
+                "auxillary_file_paths": [
+                    "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-50000024.pt"
+                ]
+            }
+        ],
+        "elo": 1591.5127129992936,
+        "final_checkpoint": {
+            "steps": 50000024,
+            "file_path": "results\\SoccerTwos\\SoccerTwos.onnx",
+            "reward": 0.0,
+            "creation_time": 1689205230.8380823,
+            "auxillary_file_paths": [
+                "results\\SoccerTwos\\SoccerTwos\\SoccerTwos-50000024.pt"
+            ]
+        }
+    },
+    "metadata": {
+        "stats_format_version": "0.3.0",
+        "mlagents_version": "0.31.0.dev0",
+        "torch_version": "1.13.1+cu116"
+    }
+}