akanametov commited on
Commit
4530712
1 Parent(s): b3c5f95

Trained SoccerTwos agent upload

Browse files
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: akanametov/MLAgents-SoccerTwos
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 sha256:631237d1d2ef4a263077f627c8aa1e4c014440d7fff6b2043cf566a0a97b120c
3
+ size 1781689
SoccerTwos/SoccerTwos-1026228.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:631237d1d2ef4a263077f627c8aa1e4c014440d7fff6b2043cf566a0a97b120c
3
+ size 1781689
SoccerTwos/SoccerTwos-1026228.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d79fa8f9fa863c75c209f6986229f3a249d7416be71d67f0f5da6e513f7eb903
3
+ size 28590523
SoccerTwos/SoccerTwos-499114.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5184ff67e42c7034c5a75cd224d284dd6d4434e842c5214f3baca6daf6e6798c
3
+ size 1781689
SoccerTwos/SoccerTwos-499114.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:01301570baea05e0c9b9b890f44f59af5238fdf1848529481c109190cc54119a
3
+ size 28590523
SoccerTwos/SoccerTwos-999766.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d15b056d110fcbda462a2876bbe095d9b25036082e138e1083d3cd4ee1190db2
3
+ size 1781689
SoccerTwos/SoccerTwos-999766.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5956f88fb001eabb7c4065189f621611f00b03a93a53ae86e099b7d050bbe745
3
+ size 28590523
SoccerTwos/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d79fa8f9fa863c75c209f6986229f3a249d7416be71d67f0f5da6e513f7eb903
3
+ size 28590523
SoccerTwos/events.out.tfevents.1673807871.697baef2d0c0.6167.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b2b22976af17e534828fa4100c11331e0eacbef85a84e20639b64d17ac9fa2f7
3
+ size 545780
config.json ADDED
@@ -0,0 +1 @@
 
 
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"}, "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, "checkpoint_interval": 500000, "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/linux/linux-x86_64/soccer-twos/soccer-twos", "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 ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ default_settings: null
2
+ behaviors:
3
+ SoccerTwos:
4
+ trainer_type: poca
5
+ hyperparameters:
6
+ batch_size: 2048
7
+ buffer_size: 20480
8
+ learning_rate: 0.0003
9
+ beta: 0.005
10
+ epsilon: 0.2
11
+ lambd: 0.95
12
+ num_epoch: 3
13
+ learning_rate_schedule: constant
14
+ beta_schedule: constant
15
+ epsilon_schedule: constant
16
+ network_settings:
17
+ normalize: false
18
+ hidden_units: 512
19
+ num_layers: 2
20
+ vis_encode_type: simple
21
+ memory: null
22
+ goal_conditioning_type: hyper
23
+ deterministic: false
24
+ reward_signals:
25
+ extrinsic:
26
+ gamma: 0.99
27
+ strength: 1.0
28
+ network_settings:
29
+ normalize: false
30
+ hidden_units: 128
31
+ num_layers: 2
32
+ vis_encode_type: simple
33
+ memory: null
34
+ goal_conditioning_type: hyper
35
+ deterministic: false
36
+ init_path: null
37
+ keep_checkpoints: 5
38
+ checkpoint_interval: 500000
39
+ max_steps: 50000000
40
+ time_horizon: 1000
41
+ summary_freq: 10000
42
+ threaded: false
43
+ self_play:
44
+ save_steps: 50000
45
+ team_change: 200000
46
+ swap_steps: 2000
47
+ window: 10
48
+ play_against_latest_model_ratio: 0.5
49
+ initial_elo: 1200.0
50
+ behavioral_cloning: null
51
+ env_settings:
52
+ env_path: ./training-envs-executables/linux/linux-x86_64/soccer-twos/soccer-twos
53
+ env_args: null
54
+ base_port: 5005
55
+ num_envs: 1
56
+ num_areas: 1
57
+ seed: -1
58
+ max_lifetime_restarts: 10
59
+ restarts_rate_limit_n: 1
60
+ restarts_rate_limit_period_s: 60
61
+ engine_settings:
62
+ width: 84
63
+ height: 84
64
+ quality_level: 5
65
+ time_scale: 20
66
+ target_frame_rate: -1
67
+ capture_frame_rate: 60
68
+ no_graphics: true
69
+ environment_parameters: null
70
+ checkpoint_settings:
71
+ run_id: SoccerTwos
72
+ initialize_from: null
73
+ load_model: false
74
+ resume: false
75
+ force: false
76
+ train_model: false
77
+ inference: false
78
+ results_dir: results
79
+ torch_settings:
80
+ device: null
81
+ debug: false
run_logs/Player-0.log ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Mono path[0] = '/content/ml-agents/./training-envs-executables/linux/linux-x86_64/soccer-twos/soccer-twos_Data/Managed'
2
+ Mono config path = '/content/ml-agents/./training-envs-executables/linux/linux-x86_64/soccer-twos/soccer-twos_Data/MonoBleedingEdge/etc'
3
+ Preloaded 'lib_burst_generated.so'
4
+ Preloaded 'libgrpc_csharp_ext.x64.so'
5
+ Initialize engine version: 2019.4.30f1 (e8c891080a1f)
6
+ [Subsystems] Discovering subsystems at path /content/ml-agents/./training-envs-executables/linux/linux-x86_64/soccer-twos/soccer-twos_Data/UnitySubsystems
7
+ Forcing GfxDevice: Null
8
+ GfxDevice: creating device client; threaded=0
9
+ NullGfxDevice:
10
+ Version: NULL 1.0 [1.0]
11
+ Renderer: Null Device
12
+ Vendor: Unity Technologies
13
+ ALSA lib confmisc.c:767:(parse_card) cannot find card '0'
14
+ ALSA lib conf.c:4528:(_snd_config_evaluate) function snd_func_card_driver returned error: No such file or directory
15
+ ALSA lib confmisc.c:392:(snd_func_concat) error evaluating strings
16
+ ALSA lib conf.c:4528:(_snd_config_evaluate) function snd_func_concat returned error: No such file or directory
17
+ ALSA lib confmisc.c:1246:(snd_func_refer) error evaluating name
18
+ ALSA lib conf.c:4528:(_snd_config_evaluate) function snd_func_refer returned error: No such file or directory
19
+ ALSA lib conf.c:5007:(snd_config_expand) Evaluate error: No such file or directory
20
+ ALSA lib pcm.c:2495:(snd_pcm_open_noupdate) Unknown PCM default
21
+ FMOD failed to initialize the output device.: "Error initializing output device. " (60)
22
+ (Filename: Line: 1652)
23
+
24
+ Forced to initialize FMOD to to the device driver's system output rate 48000, this may impact performance and/or give inconsistent experiences compared to selected sample rate 48000
25
+ (Filename: Line: 1588)
26
+
27
+ ALSA lib confmisc.c:767:(parse_card) cannot find card '0'
28
+ ALSA lib conf.c:4528:(_snd_config_evaluate) function snd_func_card_driver returned error: No such file or directory
29
+ ALSA lib confmisc.c:392:(snd_func_concat) error evaluating strings
30
+ ALSA lib conf.c:4528:(_snd_config_evaluate) function snd_func_concat returned error: No such file or directory
31
+ ALSA lib confmisc.c:1246:(snd_func_refer) error evaluating name
32
+ ALSA lib conf.c:4528:(_snd_config_evaluate) function snd_func_refer returned error: No such file or directory
33
+ ALSA lib conf.c:5007:(snd_config_expand) Evaluate error: No such file or directory
34
+ ALSA lib pcm.c:2495:(snd_pcm_open_noupdate) Unknown PCM default
35
+ FMOD failed to initialize the output device.: "Error initializing output device. " (60)
36
+ (Filename: Line: 1652)
37
+
38
+ FMOD initialized on nosound output
39
+ Begin MonoManager ReloadAssembly
40
+ - Completed reload, in 0.067 seconds
41
+ UnloadTime: 0.881175 ms
42
+ Fallback handler could not load library /content/ml-agents/./training-envs-executables/linux/linux-x86_64/soccer-twos/soccer-twos_Data/Mono/libcoreclr.so
43
+ Fallback handler could not load library /content/ml-agents/./training-envs-executables/linux/linux-x86_64/soccer-twos/soccer-twos_Data/Mono/libcoreclr.so
44
+ Fallback handler could not load library /content/ml-agents/./training-envs-executables/linux/linux-x86_64/soccer-twos/soccer-twos_Data/Mono/libcoreclr.so
45
+ Fallback handler could not load library /content/ml-agents/./training-envs-executables/linux/linux-x86_64/soccer-twos/soccer-twos_Data/Mono/libSystem.dylib
46
+ Fallback handler could not load library /content/ml-agents/./training-envs-executables/linux/linux-x86_64/soccer-twos/soccer-twos_Data/Mono/libSystem.dylib.so
47
+ Fallback handler could not load library /content/ml-agents/./training-envs-executables/linux/linux-x86_64/soccer-twos/soccer-twos_Data/Mono/libSystem.dylib
48
+ Fallback handler could not load library /content/ml-agents/./training-envs-executables/linux/linux-x86_64/soccer-twos/soccer-twos_Data/Mono/libcoreclr.so
49
+ Fallback handler could not load library /content/ml-agents/./training-envs-executables/linux/linux-x86_64/soccer-twos/soccer-twos_Data/Mono/libcoreclr.so
50
+ Fallback handler could not load library /content/ml-agents/./training-envs-executables/linux/linux-x86_64/soccer-twos/soccer-twos_Data/Mono/libcoreclr.so
51
+ Fallback handler could not load library /content/ml-agents/./training-envs-executables/linux/linux-x86_64/soccer-twos/soccer-twos_Data/Mono/libSystem.dylib
52
+ Fallback handler could not load library /content/ml-agents/./training-envs-executables/linux/linux-x86_64/soccer-twos/soccer-twos_Data/Mono/libSystem.dylib.so
53
+ Fallback handler could not load library /content/ml-agents/./training-envs-executables/linux/linux-x86_64/soccer-twos/soccer-twos_Data/Mono/libSystem.dylib
54
+ EnvConfigurationChannel is set up.
55
+ (Filename: ./Runtime/Export/Debug/Debug.bindings.h Line: 39)
56
+
57
+ requesting resize 84 x 84
58
+ Setting up 1 worker threads for Enlighten.
59
+ Thread -> id: 7fa63cbfa700 -> priority: 1
run_logs/timers.json ADDED
@@ -0,0 +1,431 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "root",
3
+ "gauges": {
4
+ "SoccerTwos.Policy.Entropy.mean": {
5
+ "value": 3.1996026039123535,
6
+ "min": 2.9278101921081543,
7
+ "max": 3.2722084522247314,
8
+ "count": 102
9
+ },
10
+ "SoccerTwos.Policy.Entropy.sum": {
11
+ "value": 62328.2578125,
12
+ "min": 49968.30078125,
13
+ "max": 74948.84375,
14
+ "count": 102
15
+ },
16
+ "SoccerTwos.Environment.EpisodeLength.mean": {
17
+ "value": 810.6666666666666,
18
+ "min": 531.6666666666666,
19
+ "max": 999.0,
20
+ "count": 102
21
+ },
22
+ "SoccerTwos.Environment.EpisodeLength.sum": {
23
+ "value": 19456.0,
24
+ "min": 16516.0,
25
+ "max": 23324.0,
26
+ "count": 102
27
+ },
28
+ "SoccerTwos.Self-play.ELO.mean": {
29
+ "value": 1204.0218902996917,
30
+ "min": 1199.5028730369677,
31
+ "max": 1210.6164886930674,
32
+ "count": 69
33
+ },
34
+ "SoccerTwos.Self-play.ELO.sum": {
35
+ "value": 4816.087561198767,
36
+ "min": 2400.4997554563024,
37
+ "max": 14461.67778818551,
38
+ "count": 69
39
+ },
40
+ "SoccerTwos.Step.mean": {
41
+ "value": 1019052.0,
42
+ "min": 9478.0,
43
+ "max": 1019052.0,
44
+ "count": 102
45
+ },
46
+ "SoccerTwos.Step.sum": {
47
+ "value": 1019052.0,
48
+ "min": 9478.0,
49
+ "max": 1019052.0,
50
+ "count": 102
51
+ },
52
+ "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": {
53
+ "value": 0.020359380170702934,
54
+ "min": -0.03247058019042015,
55
+ "max": 0.0642065554857254,
56
+ "count": 102
57
+ },
58
+ "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": {
59
+ "value": 0.24431255459785461,
60
+ "min": -0.42885616421699524,
61
+ "max": 0.8988363742828369,
62
+ "count": 102
63
+ },
64
+ "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": {
65
+ "value": 0.018934523686766624,
66
+ "min": -0.044129129499197006,
67
+ "max": 0.06423702090978622,
68
+ "count": 102
69
+ },
70
+ "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": {
71
+ "value": 0.2272142916917801,
72
+ "min": -0.48542043566703796,
73
+ "max": 0.8992221355438232,
74
+ "count": 102
75
+ },
76
+ "SoccerTwos.Environment.CumulativeReward.mean": {
77
+ "value": 0.0,
78
+ "min": 0.0,
79
+ "max": 0.0,
80
+ "count": 102
81
+ },
82
+ "SoccerTwos.Environment.CumulativeReward.sum": {
83
+ "value": 0.0,
84
+ "min": 0.0,
85
+ "max": 0.0,
86
+ "count": 102
87
+ },
88
+ "SoccerTwos.Policy.ExtrinsicReward.mean": {
89
+ "value": -0.14220000058412552,
90
+ "min": -0.5333333333333333,
91
+ "max": 0.3313599983851115,
92
+ "count": 102
93
+ },
94
+ "SoccerTwos.Policy.ExtrinsicReward.sum": {
95
+ "value": -1.7064000070095062,
96
+ "min": -8.0,
97
+ "max": 5.485599935054779,
98
+ "count": 102
99
+ },
100
+ "SoccerTwos.Environment.GroupCumulativeReward.mean": {
101
+ "value": -0.14220000058412552,
102
+ "min": -0.5333333333333333,
103
+ "max": 0.3313599983851115,
104
+ "count": 102
105
+ },
106
+ "SoccerTwos.Environment.GroupCumulativeReward.sum": {
107
+ "value": -1.7064000070095062,
108
+ "min": -8.0,
109
+ "max": 5.485599935054779,
110
+ "count": 102
111
+ },
112
+ "SoccerTwos.IsTraining.mean": {
113
+ "value": 1.0,
114
+ "min": 1.0,
115
+ "max": 1.0,
116
+ "count": 102
117
+ },
118
+ "SoccerTwos.IsTraining.sum": {
119
+ "value": 1.0,
120
+ "min": 1.0,
121
+ "max": 1.0,
122
+ "count": 102
123
+ },
124
+ "SoccerTwos.Losses.PolicyLoss.mean": {
125
+ "value": 0.017422009384851358,
126
+ "min": 0.013289852785722664,
127
+ "max": 0.021624805549314865,
128
+ "count": 47
129
+ },
130
+ "SoccerTwos.Losses.PolicyLoss.sum": {
131
+ "value": 0.017422009384851358,
132
+ "min": 0.013289852785722664,
133
+ "max": 0.021624805549314865,
134
+ "count": 47
135
+ },
136
+ "SoccerTwos.Losses.ValueLoss.mean": {
137
+ "value": 0.0021655489845822254,
138
+ "min": 2.518433211662341e-06,
139
+ "max": 0.006047092261724174,
140
+ "count": 47
141
+ },
142
+ "SoccerTwos.Losses.ValueLoss.sum": {
143
+ "value": 0.0021655489845822254,
144
+ "min": 2.518433211662341e-06,
145
+ "max": 0.006047092261724174,
146
+ "count": 47
147
+ },
148
+ "SoccerTwos.Losses.BaselineLoss.mean": {
149
+ "value": 0.002190248357752959,
150
+ "min": 3.3172456975686753e-06,
151
+ "max": 0.00505002357531339,
152
+ "count": 47
153
+ },
154
+ "SoccerTwos.Losses.BaselineLoss.sum": {
155
+ "value": 0.002190248357752959,
156
+ "min": 3.3172456975686753e-06,
157
+ "max": 0.00505002357531339,
158
+ "count": 47
159
+ },
160
+ "SoccerTwos.Policy.LearningRate.mean": {
161
+ "value": 0.0003,
162
+ "min": 0.0003,
163
+ "max": 0.0003,
164
+ "count": 47
165
+ },
166
+ "SoccerTwos.Policy.LearningRate.sum": {
167
+ "value": 0.0003,
168
+ "min": 0.0003,
169
+ "max": 0.0003,
170
+ "count": 47
171
+ },
172
+ "SoccerTwos.Policy.Epsilon.mean": {
173
+ "value": 0.20000000000000007,
174
+ "min": 0.20000000000000007,
175
+ "max": 0.20000000000000007,
176
+ "count": 47
177
+ },
178
+ "SoccerTwos.Policy.Epsilon.sum": {
179
+ "value": 0.20000000000000007,
180
+ "min": 0.20000000000000007,
181
+ "max": 0.20000000000000007,
182
+ "count": 47
183
+ },
184
+ "SoccerTwos.Policy.Beta.mean": {
185
+ "value": 0.005000000000000001,
186
+ "min": 0.005000000000000001,
187
+ "max": 0.005000000000000001,
188
+ "count": 47
189
+ },
190
+ "SoccerTwos.Policy.Beta.sum": {
191
+ "value": 0.005000000000000001,
192
+ "min": 0.005000000000000001,
193
+ "max": 0.005000000000000001,
194
+ "count": 47
195
+ }
196
+ },
197
+ "metadata": {
198
+ "timer_format_version": "0.1.0",
199
+ "start_time_seconds": "1673807868",
200
+ "python_version": "3.8.16 (default, Dec 7 2022, 01:12:13) \n[GCC 7.5.0]",
201
+ "command_line_arguments": "/usr/local/bin/mlagents-learn ./config/poca/SoccerTwos.yaml --env=./training-envs-executables/linux/linux-x86_64/soccer-twos/soccer-twos --run-id=SoccerTwos --no-graphics",
202
+ "mlagents_version": "0.29.0.dev0",
203
+ "mlagents_envs_version": "0.29.0.dev0",
204
+ "communication_protocol_version": "1.5.0",
205
+ "pytorch_version": "1.8.1+cu102",
206
+ "numpy_version": "1.21.6",
207
+ "end_time_seconds": "1673814156"
208
+ },
209
+ "total": 6287.747221217,
210
+ "count": 1,
211
+ "self": 10.005318574998455,
212
+ "children": {
213
+ "run_training.setup": {
214
+ "total": 0.1071236590000808,
215
+ "count": 1,
216
+ "self": 0.1071236590000808
217
+ },
218
+ "TrainerController.start_learning": {
219
+ "total": 6277.634778983001,
220
+ "count": 1,
221
+ "self": 9.346853454097982,
222
+ "children": {
223
+ "TrainerController._reset_env": {
224
+ "total": 6.287176204000843,
225
+ "count": 6,
226
+ "self": 6.287176204000843
227
+ },
228
+ "TrainerController.advance": {
229
+ "total": 6261.662972145901,
230
+ "count": 514320,
231
+ "self": 9.436193643903607,
232
+ "children": {
233
+ "env_step": {
234
+ "total": 5814.310632615159,
235
+ "count": 514320,
236
+ "self": 3150.765826409681,
237
+ "children": {
238
+ "SubprocessEnvManager._take_step": {
239
+ "total": 2657.5126975569865,
240
+ "count": 514320,
241
+ "self": 64.71362542510724,
242
+ "children": {
243
+ "TorchPolicy.evaluate": {
244
+ "total": 2592.7990721318793,
245
+ "count": 1027698,
246
+ "self": 521.0321252985041,
247
+ "children": {
248
+ "TorchPolicy.sample_actions": {
249
+ "total": 2071.766946833375,
250
+ "count": 1027698,
251
+ "self": 2071.766946833375
252
+ }
253
+ }
254
+ }
255
+ }
256
+ },
257
+ "workers": {
258
+ "total": 6.032108648491885,
259
+ "count": 514319,
260
+ "self": 0.0,
261
+ "children": {
262
+ "worker_root": {
263
+ "total": 6249.836879560299,
264
+ "count": 514319,
265
+ "is_parallel": true,
266
+ "self": 3779.7106470330614,
267
+ "children": {
268
+ "run_training.setup": {
269
+ "total": 0.0,
270
+ "count": 0,
271
+ "is_parallel": true,
272
+ "self": 0.0,
273
+ "children": {
274
+ "steps_from_proto": {
275
+ "total": 0.0017015900000387774,
276
+ "count": 2,
277
+ "is_parallel": true,
278
+ "self": 0.000768551000419393,
279
+ "children": {
280
+ "_process_rank_one_or_two_observation": {
281
+ "total": 0.0009330389996193844,
282
+ "count": 12,
283
+ "is_parallel": true,
284
+ "self": 0.0009330389996193844
285
+ }
286
+ }
287
+ },
288
+ "UnityEnvironment.step": {
289
+ "total": 0.026431305999949473,
290
+ "count": 1,
291
+ "is_parallel": true,
292
+ "self": 0.0003021919997081568,
293
+ "children": {
294
+ "UnityEnvironment._generate_step_input": {
295
+ "total": 0.0002521020001040597,
296
+ "count": 1,
297
+ "is_parallel": true,
298
+ "self": 0.0002521020001040597
299
+ },
300
+ "communicator.exchange": {
301
+ "total": 0.0243987260000722,
302
+ "count": 1,
303
+ "is_parallel": true,
304
+ "self": 0.0243987260000722
305
+ },
306
+ "steps_from_proto": {
307
+ "total": 0.0014782860000650544,
308
+ "count": 2,
309
+ "is_parallel": true,
310
+ "self": 0.0006566069998825697,
311
+ "children": {
312
+ "_process_rank_one_or_two_observation": {
313
+ "total": 0.0008216790001824847,
314
+ "count": 12,
315
+ "is_parallel": true,
316
+ "self": 0.0008216790001824847
317
+ }
318
+ }
319
+ }
320
+ }
321
+ }
322
+ }
323
+ },
324
+ "UnityEnvironment.step": {
325
+ "total": 2470.121857965237,
326
+ "count": 514318,
327
+ "is_parallel": true,
328
+ "self": 122.82167207231305,
329
+ "children": {
330
+ "UnityEnvironment._generate_step_input": {
331
+ "total": 81.66833717314012,
332
+ "count": 514318,
333
+ "is_parallel": true,
334
+ "self": 81.66833717314012
335
+ },
336
+ "communicator.exchange": {
337
+ "total": 1582.744087071872,
338
+ "count": 514318,
339
+ "is_parallel": true,
340
+ "self": 1582.744087071872
341
+ },
342
+ "steps_from_proto": {
343
+ "total": 682.8877616479117,
344
+ "count": 1028636,
345
+ "is_parallel": true,
346
+ "self": 295.6551202482824,
347
+ "children": {
348
+ "_process_rank_one_or_two_observation": {
349
+ "total": 387.2326413996293,
350
+ "count": 6171816,
351
+ "is_parallel": true,
352
+ "self": 387.2326413996293
353
+ }
354
+ }
355
+ }
356
+ }
357
+ },
358
+ "steps_from_proto": {
359
+ "total": 0.004374562000975857,
360
+ "count": 10,
361
+ "is_parallel": true,
362
+ "self": 0.0019437449996075884,
363
+ "children": {
364
+ "_process_rank_one_or_two_observation": {
365
+ "total": 0.0024308170013682684,
366
+ "count": 60,
367
+ "is_parallel": true,
368
+ "self": 0.0024308170013682684
369
+ }
370
+ }
371
+ }
372
+ }
373
+ }
374
+ }
375
+ }
376
+ }
377
+ },
378
+ "trainer_advance": {
379
+ "total": 437.9161458868384,
380
+ "count": 514319,
381
+ "self": 27.845883952824124,
382
+ "children": {
383
+ "process_trajectory": {
384
+ "total": 119.94541229301785,
385
+ "count": 514319,
386
+ "self": 119.49111925601801,
387
+ "children": {
388
+ "RLTrainer._checkpoint": {
389
+ "total": 0.454293036999843,
390
+ "count": 2,
391
+ "self": 0.454293036999843
392
+ }
393
+ }
394
+ },
395
+ "_update_policy": {
396
+ "total": 290.12484964099644,
397
+ "count": 47,
398
+ "self": 180.0239040919996,
399
+ "children": {
400
+ "TorchPOCAOptimizer.update": {
401
+ "total": 110.10094554899683,
402
+ "count": 1410,
403
+ "self": 110.10094554899683
404
+ }
405
+ }
406
+ }
407
+ }
408
+ }
409
+ }
410
+ },
411
+ "trainer_threads": {
412
+ "total": 1.3760000001639128e-06,
413
+ "count": 1,
414
+ "self": 1.3760000001639128e-06
415
+ },
416
+ "TrainerController._save_models": {
417
+ "total": 0.33777580300011323,
418
+ "count": 1,
419
+ "self": 0.0021911399999225978,
420
+ "children": {
421
+ "RLTrainer._checkpoint": {
422
+ "total": 0.33558466300019063,
423
+ "count": 1,
424
+ "self": 0.33558466300019063
425
+ }
426
+ }
427
+ }
428
+ }
429
+ }
430
+ }
431
+ }
run_logs/training_status.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "SoccerTwos": {
3
+ "checkpoints": [
4
+ {
5
+ "steps": 499114,
6
+ "file_path": "results/SoccerTwos/SoccerTwos/SoccerTwos-499114.onnx",
7
+ "reward": 0.0,
8
+ "creation_time": 1673810931.2366776,
9
+ "auxillary_file_paths": [
10
+ "results/SoccerTwos/SoccerTwos/SoccerTwos-499114.pt"
11
+ ]
12
+ },
13
+ {
14
+ "steps": 999766,
15
+ "file_path": "results/SoccerTwos/SoccerTwos/SoccerTwos-999766.onnx",
16
+ "reward": 0.0,
17
+ "creation_time": 1673813988.7482765,
18
+ "auxillary_file_paths": [
19
+ "results/SoccerTwos/SoccerTwos/SoccerTwos-999766.pt"
20
+ ]
21
+ },
22
+ {
23
+ "steps": 1026228,
24
+ "file_path": "results/SoccerTwos/SoccerTwos/SoccerTwos-1026228.onnx",
25
+ "reward": 0.0,
26
+ "creation_time": 1673814146.275693,
27
+ "auxillary_file_paths": [
28
+ "results/SoccerTwos/SoccerTwos/SoccerTwos-1026228.pt"
29
+ ]
30
+ }
31
+ ],
32
+ "elo": 1204.6475008961813,
33
+ "final_checkpoint": {
34
+ "steps": 1026228,
35
+ "file_path": "results/SoccerTwos/SoccerTwos.onnx",
36
+ "reward": 0.0,
37
+ "creation_time": 1673814146.275693,
38
+ "auxillary_file_paths": [
39
+ "results/SoccerTwos/SoccerTwos/SoccerTwos-1026228.pt"
40
+ ]
41
+ }
42
+ },
43
+ "metadata": {
44
+ "stats_format_version": "0.3.0",
45
+ "mlagents_version": "0.29.0.dev0",
46
+ "torch_version": "1.8.1+cu102"
47
+ }
48
+ }