Commit ·
9b910bc
1
Parent(s): 2d57dab
First Push`
Browse files- README.md +31 -0
- SoccerTwos.onnx +3 -0
- SoccerTwos/SoccerTwos-10000050.onnx +3 -0
- SoccerTwos/SoccerTwos-10000050.pt +3 -0
- SoccerTwos/SoccerTwos-17598.onnx +3 -0
- SoccerTwos/SoccerTwos-17598.pt +3 -0
- SoccerTwos/SoccerTwos-51174.onnx +3 -0
- SoccerTwos/SoccerTwos-51174.pt +3 -0
- SoccerTwos/SoccerTwos-8499972.onnx +3 -0
- SoccerTwos/SoccerTwos-8499972.pt +3 -0
- SoccerTwos/SoccerTwos-8999992.onnx +3 -0
- SoccerTwos/SoccerTwos-8999992.pt +3 -0
- SoccerTwos/SoccerTwos-9499927.onnx +3 -0
- SoccerTwos/SoccerTwos-9499927.pt +3 -0
- SoccerTwos/SoccerTwos-9999966.onnx +3 -0
- SoccerTwos/SoccerTwos-9999966.pt +3 -0
- SoccerTwos/checkpoint.pt +3 -0
- SoccerTwos/events.out.tfevents.1675468472.DESKTOP-CG26U2K.16928.0 +3 -0
- config.json +1 -0
- configuration.yaml +82 -0
- run_logs/Player-0.log +0 -0
- run_logs/timers.json +368 -0
- run_logs/training_status.json +66 -0
README.md
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
---
|
| 3 |
+
tags:
|
| 4 |
+
- unity-ml-agents
|
| 5 |
+
- ml-agents
|
| 6 |
+
- deep-reinforcement-learning
|
| 7 |
+
- reinforcement-learning
|
| 8 |
+
- ML-Agents-SoccerTwos
|
| 9 |
+
library_name: ml-agents
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# **poca** Agent playing **SoccerTwos**
|
| 13 |
+
This is a trained model of a **poca** agent playing **SoccerTwos** using the [Unity ML-Agents Library](https://github.com/Unity-Technologies/ml-agents).
|
| 14 |
+
|
| 15 |
+
## Usage (with ML-Agents)
|
| 16 |
+
The Documentation: https://github.com/huggingface/ml-agents#get-started
|
| 17 |
+
We wrote a complete tutorial to learn to train your first agent using ML-Agents and publish it to the Hub:
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
### Resume the training
|
| 21 |
+
```
|
| 22 |
+
mlagents-learn <your_configuration_file_path.yaml> --run-id=<run_id> --resume
|
| 23 |
+
```
|
| 24 |
+
### Watch your Agent play
|
| 25 |
+
You can watch your agent **playing directly in your browser:**.
|
| 26 |
+
|
| 27 |
+
1. Go to https://huggingface.co/spaces/unity/ML-Agents-SoccerTwos
|
| 28 |
+
2. Step 1: Write your model_id: RamonAnkersmit/poca-SoccerTwos-Gpu
|
| 29 |
+
3. Step 2: Select your *.nn /*.onnx file
|
| 30 |
+
4. Click on Watch the agent play 👀
|
| 31 |
+
|
SoccerTwos.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1aa5143ecdf1ed803ab060164382857c958396d6e44251a14e3d2d44b2e43ad7
|
| 3 |
+
size 1768747
|
SoccerTwos/SoccerTwos-10000050.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1aa5143ecdf1ed803ab060164382857c958396d6e44251a14e3d2d44b2e43ad7
|
| 3 |
+
size 1768747
|
SoccerTwos/SoccerTwos-10000050.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e3f5163629783fc7ff21a2c12679a2c2bc6b5b52c603de3ec08daecfc7c5a6f5
|
| 3 |
+
size 28431689
|
SoccerTwos/SoccerTwos-17598.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8fecf24e61447c757910c6d668e7681c9dfa42c33fd1b482b51b2f200770bc8d
|
| 3 |
+
size 1766874
|
SoccerTwos/SoccerTwos-17598.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d314457b1b81e2a0beaa4a25c84e1860125eb3b536128ffb5fe788ef3006ce48
|
| 3 |
+
size 9479503
|
SoccerTwos/SoccerTwos-51174.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:23a7cc8bf370c8c4ea1ad7732640392439602b4aefff11832768ca11091c6343
|
| 3 |
+
size 1768747
|
SoccerTwos/SoccerTwos-51174.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ba54e7a6ffacdf0e99abe8f7ad1d664d3a70432a75927ea18c8ac60f45044203
|
| 3 |
+
size 28431323
|
SoccerTwos/SoccerTwos-8499972.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5ff0ad631cfffb5d4596df166ca581c26db398e10ea850be1003a1e6012496c3
|
| 3 |
+
size 1768747
|
SoccerTwos/SoccerTwos-8499972.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f19afbe92d227e56b8e5bd862402248be8d561845ffbdb3bd94e876e6a78878b
|
| 3 |
+
size 28431567
|
SoccerTwos/SoccerTwos-8999992.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:17051fae5b3cf3cab577d100db0df9014ab904593420347f85bc105734dffee9
|
| 3 |
+
size 1768747
|
SoccerTwos/SoccerTwos-8999992.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d28a9a44ce7f5651503e49e1c74247a198e2c0f4efa8f7cf77054eda9ee5ae98
|
| 3 |
+
size 28431567
|
SoccerTwos/SoccerTwos-9499927.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6292f1ebcf2098e338747574bd1347f7ebf9b775dbecdc5013bbf4a7e0f8ae5f
|
| 3 |
+
size 1768747
|
SoccerTwos/SoccerTwos-9499927.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3a5699cae5e96deaf80ef9ffbdeaeab55cde3ce7e02267ebfdfd2e3aff40398b
|
| 3 |
+
size 28431567
|
SoccerTwos/SoccerTwos-9999966.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1aa5143ecdf1ed803ab060164382857c958396d6e44251a14e3d2d44b2e43ad7
|
| 3 |
+
size 1768747
|
SoccerTwos/SoccerTwos-9999966.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:859d30c83946697511f6462e56fbb50ab1fe603df3d32e1992fb620055db2087
|
| 3 |
+
size 28431567
|
SoccerTwos/checkpoint.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dbccdfeeff8fdb79ba2a3371681af8f36a75e64c423fde2612be15051677ed8a
|
| 3 |
+
size 28430591
|
SoccerTwos/events.out.tfevents.1675468472.DESKTOP-CG26U2K.16928.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ccc2e8bc8fb49552b76fe3d979ad94e00f8b9f67b1bc8c826acc6c136fc5ad57
|
| 3 |
+
size 10014885
|
config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"default_settings": null, "behaviors": {"SoccerTwos": {"trainer_type": "poca", "hyperparameters": {"batch_size": 4096, "buffer_size": 40960, "learning_rate": 0.00025, "beta": 0.005, "epsilon": 0.15, "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": 0.995, "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": 10000000, "time_horizon": 1500, "summary_freq": 10000, "threaded": false, "self_play": {"save_steps": 70000, "team_change": 350000, "swap_steps": 3000, "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": "SoccerTwosGPU", "initialize_from": null, "load_model": false, "resume": false, "force": true, "train_model": false, "inference": false, "results_dir": "results"}, "torch_settings": {"device": "cuda"}, "debug": false}
|
configuration.yaml
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
default_settings: null
|
| 2 |
+
behaviors:
|
| 3 |
+
SoccerTwos:
|
| 4 |
+
trainer_type: poca
|
| 5 |
+
hyperparameters:
|
| 6 |
+
batch_size: 4096
|
| 7 |
+
buffer_size: 40960
|
| 8 |
+
learning_rate: 0.00025
|
| 9 |
+
beta: 0.005
|
| 10 |
+
epsilon: 0.15
|
| 11 |
+
lambd: 0.95
|
| 12 |
+
num_epoch: 3
|
| 13 |
+
learning_rate_schedule: constant
|
| 14 |
+
beta_schedule: constant
|
| 15 |
+
epsilon_schedule: constant
|
| 16 |
+
checkpoint_interval: 500000
|
| 17 |
+
network_settings:
|
| 18 |
+
normalize: false
|
| 19 |
+
hidden_units: 512
|
| 20 |
+
num_layers: 2
|
| 21 |
+
vis_encode_type: simple
|
| 22 |
+
memory: null
|
| 23 |
+
goal_conditioning_type: hyper
|
| 24 |
+
deterministic: false
|
| 25 |
+
reward_signals:
|
| 26 |
+
extrinsic:
|
| 27 |
+
gamma: 0.995
|
| 28 |
+
strength: 1.0
|
| 29 |
+
network_settings:
|
| 30 |
+
normalize: false
|
| 31 |
+
hidden_units: 128
|
| 32 |
+
num_layers: 2
|
| 33 |
+
vis_encode_type: simple
|
| 34 |
+
memory: null
|
| 35 |
+
goal_conditioning_type: hyper
|
| 36 |
+
deterministic: false
|
| 37 |
+
init_path: null
|
| 38 |
+
keep_checkpoints: 5
|
| 39 |
+
even_checkpoints: false
|
| 40 |
+
max_steps: 10000000
|
| 41 |
+
time_horizon: 1500
|
| 42 |
+
summary_freq: 10000
|
| 43 |
+
threaded: false
|
| 44 |
+
self_play:
|
| 45 |
+
save_steps: 70000
|
| 46 |
+
team_change: 350000
|
| 47 |
+
swap_steps: 3000
|
| 48 |
+
window: 10
|
| 49 |
+
play_against_latest_model_ratio: 0.5
|
| 50 |
+
initial_elo: 1200.0
|
| 51 |
+
behavioral_cloning: null
|
| 52 |
+
env_settings:
|
| 53 |
+
env_path: ./training-envs-executables/SoccerTwos.exe
|
| 54 |
+
env_args: null
|
| 55 |
+
base_port: 5005
|
| 56 |
+
num_envs: 1
|
| 57 |
+
num_areas: 1
|
| 58 |
+
seed: -1
|
| 59 |
+
max_lifetime_restarts: 10
|
| 60 |
+
restarts_rate_limit_n: 1
|
| 61 |
+
restarts_rate_limit_period_s: 60
|
| 62 |
+
engine_settings:
|
| 63 |
+
width: 84
|
| 64 |
+
height: 84
|
| 65 |
+
quality_level: 5
|
| 66 |
+
time_scale: 20
|
| 67 |
+
target_frame_rate: -1
|
| 68 |
+
capture_frame_rate: 60
|
| 69 |
+
no_graphics: true
|
| 70 |
+
environment_parameters: null
|
| 71 |
+
checkpoint_settings:
|
| 72 |
+
run_id: SoccerTwosGPU
|
| 73 |
+
initialize_from: null
|
| 74 |
+
load_model: false
|
| 75 |
+
resume: false
|
| 76 |
+
force: true
|
| 77 |
+
train_model: false
|
| 78 |
+
inference: false
|
| 79 |
+
results_dir: results
|
| 80 |
+
torch_settings:
|
| 81 |
+
device: cuda
|
| 82 |
+
debug: false
|
run_logs/Player-0.log
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
run_logs/timers.json
ADDED
|
@@ -0,0 +1,368 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"name": "root",
|
| 3 |
+
"gauges": {
|
| 4 |
+
"SoccerTwos.Policy.Entropy.mean": {
|
| 5 |
+
"value": 2.104137897491455,
|
| 6 |
+
"min": 2.104137897491455,
|
| 7 |
+
"max": 3.295746088027954,
|
| 8 |
+
"count": 1000
|
| 9 |
+
},
|
| 10 |
+
"SoccerTwos.Policy.Entropy.sum": {
|
| 11 |
+
"value": 39591.45703125,
|
| 12 |
+
"min": 28661.078125,
|
| 13 |
+
"max": 105463.796875,
|
| 14 |
+
"count": 1000
|
| 15 |
+
},
|
| 16 |
+
"SoccerTwos.Environment.EpisodeLength.mean": {
|
| 17 |
+
"value": 53.0989010989011,
|
| 18 |
+
"min": 39.95867768595041,
|
| 19 |
+
"max": 999.0,
|
| 20 |
+
"count": 1000
|
| 21 |
+
},
|
| 22 |
+
"SoccerTwos.Environment.EpisodeLength.sum": {
|
| 23 |
+
"value": 19328.0,
|
| 24 |
+
"min": 16684.0,
|
| 25 |
+
"max": 23208.0,
|
| 26 |
+
"count": 1000
|
| 27 |
+
},
|
| 28 |
+
"SoccerTwos.Self-play.ELO.mean": {
|
| 29 |
+
"value": 1687.9152761926134,
|
| 30 |
+
"min": 1190.6857026828052,
|
| 31 |
+
"max": 1698.9798879194325,
|
| 32 |
+
"count": 997
|
| 33 |
+
},
|
| 34 |
+
"SoccerTwos.Self-play.ELO.sum": {
|
| 35 |
+
"value": 307200.58026705566,
|
| 36 |
+
"min": 2401.677687384432,
|
| 37 |
+
"max": 401313.1546064704,
|
| 38 |
+
"count": 997
|
| 39 |
+
},
|
| 40 |
+
"SoccerTwos.Step.mean": {
|
| 41 |
+
"value": 9999966.0,
|
| 42 |
+
"min": 9620.0,
|
| 43 |
+
"max": 9999966.0,
|
| 44 |
+
"count": 1000
|
| 45 |
+
},
|
| 46 |
+
"SoccerTwos.Step.sum": {
|
| 47 |
+
"value": 9999966.0,
|
| 48 |
+
"min": 9620.0,
|
| 49 |
+
"max": 9999966.0,
|
| 50 |
+
"count": 1000
|
| 51 |
+
},
|
| 52 |
+
"SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": {
|
| 53 |
+
"value": -0.04598451405763626,
|
| 54 |
+
"min": -0.12628822028636932,
|
| 55 |
+
"max": 0.18047218024730682,
|
| 56 |
+
"count": 1000
|
| 57 |
+
},
|
| 58 |
+
"SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": {
|
| 59 |
+
"value": -8.369181632995605,
|
| 60 |
+
"min": -22.226726531982422,
|
| 61 |
+
"max": 28.187450408935547,
|
| 62 |
+
"count": 1000
|
| 63 |
+
},
|
| 64 |
+
"SoccerTwos.Policy.ExtrinsicValueEstimate.mean": {
|
| 65 |
+
"value": -0.04881046712398529,
|
| 66 |
+
"min": -0.12806162238121033,
|
| 67 |
+
"max": 0.178554505109787,
|
| 68 |
+
"count": 1000
|
| 69 |
+
},
|
| 70 |
+
"SoccerTwos.Policy.ExtrinsicValueEstimate.sum": {
|
| 71 |
+
"value": -8.883504867553711,
|
| 72 |
+
"min": -22.53884506225586,
|
| 73 |
+
"max": 28.47722625732422,
|
| 74 |
+
"count": 1000
|
| 75 |
+
},
|
| 76 |
+
"SoccerTwos.Environment.CumulativeReward.mean": {
|
| 77 |
+
"value": 0.0,
|
| 78 |
+
"min": 0.0,
|
| 79 |
+
"max": 0.0,
|
| 80 |
+
"count": 1000
|
| 81 |
+
},
|
| 82 |
+
"SoccerTwos.Environment.CumulativeReward.sum": {
|
| 83 |
+
"value": 0.0,
|
| 84 |
+
"min": 0.0,
|
| 85 |
+
"max": 0.0,
|
| 86 |
+
"count": 1000
|
| 87 |
+
},
|
| 88 |
+
"SoccerTwos.Policy.ExtrinsicReward.mean": {
|
| 89 |
+
"value": 0.1185000018104092,
|
| 90 |
+
"min": -0.7692307692307693,
|
| 91 |
+
"max": 0.5266690904443915,
|
| 92 |
+
"count": 1000
|
| 93 |
+
},
|
| 94 |
+
"SoccerTwos.Policy.ExtrinsicReward.sum": {
|
| 95 |
+
"value": 21.567000329494476,
|
| 96 |
+
"min": -56.49059993028641,
|
| 97 |
+
"max": 60.974000334739685,
|
| 98 |
+
"count": 1000
|
| 99 |
+
},
|
| 100 |
+
"SoccerTwos.Environment.GroupCumulativeReward.mean": {
|
| 101 |
+
"value": 0.1185000018104092,
|
| 102 |
+
"min": -0.7692307692307693,
|
| 103 |
+
"max": 0.5266690904443915,
|
| 104 |
+
"count": 1000
|
| 105 |
+
},
|
| 106 |
+
"SoccerTwos.Environment.GroupCumulativeReward.sum": {
|
| 107 |
+
"value": 21.567000329494476,
|
| 108 |
+
"min": -56.49059993028641,
|
| 109 |
+
"max": 60.974000334739685,
|
| 110 |
+
"count": 1000
|
| 111 |
+
},
|
| 112 |
+
"SoccerTwos.IsTraining.mean": {
|
| 113 |
+
"value": 1.0,
|
| 114 |
+
"min": 1.0,
|
| 115 |
+
"max": 1.0,
|
| 116 |
+
"count": 1000
|
| 117 |
+
},
|
| 118 |
+
"SoccerTwos.IsTraining.sum": {
|
| 119 |
+
"value": 1.0,
|
| 120 |
+
"min": 1.0,
|
| 121 |
+
"max": 1.0,
|
| 122 |
+
"count": 1000
|
| 123 |
+
},
|
| 124 |
+
"SoccerTwos.Losses.PolicyLoss.mean": {
|
| 125 |
+
"value": 0.012351672106888145,
|
| 126 |
+
"min": 0.006377983589967092,
|
| 127 |
+
"max": 0.01650880670834643,
|
| 128 |
+
"count": 242
|
| 129 |
+
},
|
| 130 |
+
"SoccerTwos.Losses.PolicyLoss.sum": {
|
| 131 |
+
"value": 0.012351672106888145,
|
| 132 |
+
"min": 0.006377983589967092,
|
| 133 |
+
"max": 0.01650880670834643,
|
| 134 |
+
"count": 242
|
| 135 |
+
},
|
| 136 |
+
"SoccerTwos.Losses.ValueLoss.mean": {
|
| 137 |
+
"value": 0.12223331481218339,
|
| 138 |
+
"min": 0.0028801917796954513,
|
| 139 |
+
"max": 0.13559552580118178,
|
| 140 |
+
"count": 242
|
| 141 |
+
},
|
| 142 |
+
"SoccerTwos.Losses.ValueLoss.sum": {
|
| 143 |
+
"value": 0.12223331481218339,
|
| 144 |
+
"min": 0.0028801917796954513,
|
| 145 |
+
"max": 0.13559552580118178,
|
| 146 |
+
"count": 242
|
| 147 |
+
},
|
| 148 |
+
"SoccerTwos.Losses.BaselineLoss.mean": {
|
| 149 |
+
"value": 0.12333853219946225,
|
| 150 |
+
"min": 0.002953560114838183,
|
| 151 |
+
"max": 0.137561371922493,
|
| 152 |
+
"count": 242
|
| 153 |
+
},
|
| 154 |
+
"SoccerTwos.Losses.BaselineLoss.sum": {
|
| 155 |
+
"value": 0.12333853219946225,
|
| 156 |
+
"min": 0.002953560114838183,
|
| 157 |
+
"max": 0.137561371922493,
|
| 158 |
+
"count": 242
|
| 159 |
+
},
|
| 160 |
+
"SoccerTwos.Policy.LearningRate.mean": {
|
| 161 |
+
"value": 0.00025000000000000006,
|
| 162 |
+
"min": 0.00025000000000000006,
|
| 163 |
+
"max": 0.00025000000000000006,
|
| 164 |
+
"count": 242
|
| 165 |
+
},
|
| 166 |
+
"SoccerTwos.Policy.LearningRate.sum": {
|
| 167 |
+
"value": 0.00025000000000000006,
|
| 168 |
+
"min": 0.00025000000000000006,
|
| 169 |
+
"max": 0.00025000000000000006,
|
| 170 |
+
"count": 242
|
| 171 |
+
},
|
| 172 |
+
"SoccerTwos.Policy.Epsilon.mean": {
|
| 173 |
+
"value": 0.15000000000000002,
|
| 174 |
+
"min": 0.15000000000000002,
|
| 175 |
+
"max": 0.15000000000000002,
|
| 176 |
+
"count": 242
|
| 177 |
+
},
|
| 178 |
+
"SoccerTwos.Policy.Epsilon.sum": {
|
| 179 |
+
"value": 0.15000000000000002,
|
| 180 |
+
"min": 0.15000000000000002,
|
| 181 |
+
"max": 0.15000000000000002,
|
| 182 |
+
"count": 242
|
| 183 |
+
},
|
| 184 |
+
"SoccerTwos.Policy.Beta.mean": {
|
| 185 |
+
"value": 0.005000000000000001,
|
| 186 |
+
"min": 0.005000000000000001,
|
| 187 |
+
"max": 0.005000000000000001,
|
| 188 |
+
"count": 242
|
| 189 |
+
},
|
| 190 |
+
"SoccerTwos.Policy.Beta.sum": {
|
| 191 |
+
"value": 0.005000000000000001,
|
| 192 |
+
"min": 0.005000000000000001,
|
| 193 |
+
"max": 0.005000000000000001,
|
| 194 |
+
"count": 242
|
| 195 |
+
}
|
| 196 |
+
},
|
| 197 |
+
"metadata": {
|
| 198 |
+
"timer_format_version": "0.1.0",
|
| 199 |
+
"start_time_seconds": "1675468435",
|
| 200 |
+
"python_version": "3.8.16 (default, Jan 17 2023, 22:25:28) [MSC v.1916 64 bit (AMD64)]",
|
| 201 |
+
"command_line_arguments": "C:\\Users\\ramon\\anaconda3\\envs\\rl_football_gpu\\Scripts\\mlagents-learn ./config/poca/SoccerTwos.yaml --env=./training-envs-executables/SoccerTwos.exe --run-id=SoccerTwosGPU --no-graphics --force --torch-device=cuda",
|
| 202 |
+
"mlagents_version": "0.31.0.dev0",
|
| 203 |
+
"mlagents_envs_version": "0.31.0.dev0",
|
| 204 |
+
"communication_protocol_version": "1.5.0",
|
| 205 |
+
"pytorch_version": "1.13.1+cu117",
|
| 206 |
+
"numpy_version": "1.21.2",
|
| 207 |
+
"end_time_seconds": "1675573353"
|
| 208 |
+
},
|
| 209 |
+
"total": 104917.1378512,
|
| 210 |
+
"count": 1,
|
| 211 |
+
"self": 2.371245499991346,
|
| 212 |
+
"children": {
|
| 213 |
+
"run_training.setup": {
|
| 214 |
+
"total": 0.7410259000000003,
|
| 215 |
+
"count": 1,
|
| 216 |
+
"self": 0.7410259000000003
|
| 217 |
+
},
|
| 218 |
+
"TrainerController.start_learning": {
|
| 219 |
+
"total": 104914.0255798,
|
| 220 |
+
"count": 1,
|
| 221 |
+
"self": 74.59980279373121,
|
| 222 |
+
"children": {
|
| 223 |
+
"TrainerController._reset_env": {
|
| 224 |
+
"total": 47.93823570001701,
|
| 225 |
+
"count": 29,
|
| 226 |
+
"self": 47.93823570001701
|
| 227 |
+
},
|
| 228 |
+
"TrainerController.advance": {
|
| 229 |
+
"total": 104790.52446330624,
|
| 230 |
+
"count": 681647,
|
| 231 |
+
"self": 78.4749788112822,
|
| 232 |
+
"children": {
|
| 233 |
+
"env_step": {
|
| 234 |
+
"total": 77107.23927279649,
|
| 235 |
+
"count": 681647,
|
| 236 |
+
"self": 49394.6467882015,
|
| 237 |
+
"children": {
|
| 238 |
+
"SubprocessEnvManager._take_step": {
|
| 239 |
+
"total": 27664.324471095126,
|
| 240 |
+
"count": 681647,
|
| 241 |
+
"self": 501.3633953925819,
|
| 242 |
+
"children": {
|
| 243 |
+
"TorchPolicy.evaluate": {
|
| 244 |
+
"total": 27162.961075702544,
|
| 245 |
+
"count": 1255672,
|
| 246 |
+
"self": 27162.961075702544
|
| 247 |
+
}
|
| 248 |
+
}
|
| 249 |
+
},
|
| 250 |
+
"workers": {
|
| 251 |
+
"total": 48.26801349986546,
|
| 252 |
+
"count": 681647,
|
| 253 |
+
"self": 0.0,
|
| 254 |
+
"children": {
|
| 255 |
+
"worker_root": {
|
| 256 |
+
"total": 104787.02264969976,
|
| 257 |
+
"count": 681647,
|
| 258 |
+
"is_parallel": true,
|
| 259 |
+
"self": 65118.05426029919,
|
| 260 |
+
"children": {
|
| 261 |
+
"steps_from_proto": {
|
| 262 |
+
"total": 0.28205810001661646,
|
| 263 |
+
"count": 58,
|
| 264 |
+
"is_parallel": true,
|
| 265 |
+
"self": 0.05843879995229884,
|
| 266 |
+
"children": {
|
| 267 |
+
"_process_rank_one_or_two_observation": {
|
| 268 |
+
"total": 0.22361930006431763,
|
| 269 |
+
"count": 232,
|
| 270 |
+
"is_parallel": true,
|
| 271 |
+
"self": 0.22361930006431763
|
| 272 |
+
}
|
| 273 |
+
}
|
| 274 |
+
},
|
| 275 |
+
"UnityEnvironment.step": {
|
| 276 |
+
"total": 39668.68633130055,
|
| 277 |
+
"count": 681647,
|
| 278 |
+
"is_parallel": true,
|
| 279 |
+
"self": 2240.013187102624,
|
| 280 |
+
"children": {
|
| 281 |
+
"UnityEnvironment._generate_step_input": {
|
| 282 |
+
"total": 1663.4952165971774,
|
| 283 |
+
"count": 681647,
|
| 284 |
+
"is_parallel": true,
|
| 285 |
+
"self": 1663.4952165971774
|
| 286 |
+
},
|
| 287 |
+
"communicator.exchange": {
|
| 288 |
+
"total": 28830.686681596937,
|
| 289 |
+
"count": 681647,
|
| 290 |
+
"is_parallel": true,
|
| 291 |
+
"self": 28830.686681596937
|
| 292 |
+
},
|
| 293 |
+
"steps_from_proto": {
|
| 294 |
+
"total": 6934.491246003817,
|
| 295 |
+
"count": 1363294,
|
| 296 |
+
"is_parallel": true,
|
| 297 |
+
"self": 1478.5051235932751,
|
| 298 |
+
"children": {
|
| 299 |
+
"_process_rank_one_or_two_observation": {
|
| 300 |
+
"total": 5455.986122410542,
|
| 301 |
+
"count": 5453176,
|
| 302 |
+
"is_parallel": true,
|
| 303 |
+
"self": 5455.986122410542
|
| 304 |
+
}
|
| 305 |
+
}
|
| 306 |
+
}
|
| 307 |
+
}
|
| 308 |
+
}
|
| 309 |
+
}
|
| 310 |
+
}
|
| 311 |
+
}
|
| 312 |
+
}
|
| 313 |
+
}
|
| 314 |
+
},
|
| 315 |
+
"trainer_advance": {
|
| 316 |
+
"total": 27604.81021169846,
|
| 317 |
+
"count": 681647,
|
| 318 |
+
"self": 375.8029027978482,
|
| 319 |
+
"children": {
|
| 320 |
+
"process_trajectory": {
|
| 321 |
+
"total": 12891.255822200601,
|
| 322 |
+
"count": 681647,
|
| 323 |
+
"self": 12871.471023100607,
|
| 324 |
+
"children": {
|
| 325 |
+
"RLTrainer._checkpoint": {
|
| 326 |
+
"total": 19.784799099993506,
|
| 327 |
+
"count": 20,
|
| 328 |
+
"self": 19.784799099993506
|
| 329 |
+
}
|
| 330 |
+
}
|
| 331 |
+
},
|
| 332 |
+
"_update_policy": {
|
| 333 |
+
"total": 14337.75148670001,
|
| 334 |
+
"count": 242,
|
| 335 |
+
"self": 9861.626107899774,
|
| 336 |
+
"children": {
|
| 337 |
+
"TorchPOCAOptimizer.update": {
|
| 338 |
+
"total": 4476.125378800237,
|
| 339 |
+
"count": 7260,
|
| 340 |
+
"self": 4476.125378800237
|
| 341 |
+
}
|
| 342 |
+
}
|
| 343 |
+
}
|
| 344 |
+
}
|
| 345 |
+
}
|
| 346 |
+
}
|
| 347 |
+
},
|
| 348 |
+
"trainer_threads": {
|
| 349 |
+
"total": 2.0000006770715117e-05,
|
| 350 |
+
"count": 1,
|
| 351 |
+
"self": 2.0000006770715117e-05
|
| 352 |
+
},
|
| 353 |
+
"TrainerController._save_models": {
|
| 354 |
+
"total": 0.9630580000084592,
|
| 355 |
+
"count": 1,
|
| 356 |
+
"self": 0.03351940000720788,
|
| 357 |
+
"children": {
|
| 358 |
+
"RLTrainer._checkpoint": {
|
| 359 |
+
"total": 0.9295386000012513,
|
| 360 |
+
"count": 1,
|
| 361 |
+
"self": 0.9295386000012513
|
| 362 |
+
}
|
| 363 |
+
}
|
| 364 |
+
}
|
| 365 |
+
}
|
| 366 |
+
}
|
| 367 |
+
}
|
| 368 |
+
}
|
run_logs/training_status.json
ADDED
|
@@ -0,0 +1,66 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"SoccerTwos": {
|
| 3 |
+
"checkpoints": [
|
| 4 |
+
{
|
| 5 |
+
"steps": 8499972,
|
| 6 |
+
"file_path": "results\\SoccerTwosGPU\\SoccerTwos\\SoccerTwos-8499972.onnx",
|
| 7 |
+
"reward": 0.0,
|
| 8 |
+
"creation_time": 1675557333.0228786,
|
| 9 |
+
"auxillary_file_paths": [
|
| 10 |
+
"results\\SoccerTwosGPU\\SoccerTwos\\SoccerTwos-8499972.pt"
|
| 11 |
+
]
|
| 12 |
+
},
|
| 13 |
+
{
|
| 14 |
+
"steps": 8999992,
|
| 15 |
+
"file_path": "results\\SoccerTwosGPU\\SoccerTwos\\SoccerTwos-8999992.onnx",
|
| 16 |
+
"reward": 0.0,
|
| 17 |
+
"creation_time": 1675562657.2476342,
|
| 18 |
+
"auxillary_file_paths": [
|
| 19 |
+
"results\\SoccerTwosGPU\\SoccerTwos\\SoccerTwos-8999992.pt"
|
| 20 |
+
]
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"steps": 9499927,
|
| 24 |
+
"file_path": "results\\SoccerTwosGPU\\SoccerTwos\\SoccerTwos-9499927.onnx",
|
| 25 |
+
"reward": 0.0,
|
| 26 |
+
"creation_time": 1675568045.5658274,
|
| 27 |
+
"auxillary_file_paths": [
|
| 28 |
+
"results\\SoccerTwosGPU\\SoccerTwos\\SoccerTwos-9499927.pt"
|
| 29 |
+
]
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"steps": 9999966,
|
| 33 |
+
"file_path": "results\\SoccerTwosGPU\\SoccerTwos\\SoccerTwos-9999966.onnx",
|
| 34 |
+
"reward": 0.0,
|
| 35 |
+
"creation_time": 1675573349.6780024,
|
| 36 |
+
"auxillary_file_paths": [
|
| 37 |
+
"results\\SoccerTwosGPU\\SoccerTwos\\SoccerTwos-9999966.pt"
|
| 38 |
+
]
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"steps": 10000050,
|
| 42 |
+
"file_path": "results\\SoccerTwosGPU\\SoccerTwos\\SoccerTwos-10000050.onnx",
|
| 43 |
+
"reward": 0.0,
|
| 44 |
+
"creation_time": 1675573350.7069924,
|
| 45 |
+
"auxillary_file_paths": [
|
| 46 |
+
"results\\SoccerTwosGPU\\SoccerTwos\\SoccerTwos-10000050.pt"
|
| 47 |
+
]
|
| 48 |
+
}
|
| 49 |
+
],
|
| 50 |
+
"elo": 1699.8390156339956,
|
| 51 |
+
"final_checkpoint": {
|
| 52 |
+
"steps": 10000050,
|
| 53 |
+
"file_path": "results\\SoccerTwosGPU\\SoccerTwos.onnx",
|
| 54 |
+
"reward": 0.0,
|
| 55 |
+
"creation_time": 1675573350.7069924,
|
| 56 |
+
"auxillary_file_paths": [
|
| 57 |
+
"results\\SoccerTwosGPU\\SoccerTwos\\SoccerTwos-10000050.pt"
|
| 58 |
+
]
|
| 59 |
+
}
|
| 60 |
+
},
|
| 61 |
+
"metadata": {
|
| 62 |
+
"stats_format_version": "0.3.0",
|
| 63 |
+
"mlagents_version": "0.31.0.dev0",
|
| 64 |
+
"torch_version": "1.13.1+cu117"
|
| 65 |
+
}
|
| 66 |
+
}
|