Hugging Face's logo

soonchang
/
poca-SoccerTwos

Reinforcement Learning
ml-agents
TensorBoard
ONNX
SoccerTwos
deep-reinforcement-learning
ML-Agents-SoccerTwos
Model card Files
xet
 Metrics Community
soonchang commited on
Commit
b4a800a
·
1 Parent(s): 7c4a682

Update model

Browse files
Files changed (11) hide show
  1. SoccerTwos/SoccerTwos-130712.pt +3 -0
  2. SoccerTwos/SoccerTwos-14160.pt +3 -0
  3. SoccerTwos/SoccerTwos-470490.onnx +3 -0
  4. SoccerTwos/SoccerTwos-470490.pt +3 -0
  5. SoccerTwos/checkpoint.pt +2 -2
  6. SoccerTwos/events.out.tfevents.1703591227.cybertron.29738.0 +3 -0
  7. config.json +1 -1
  8. configuration.yaml +7 -7
  9. run_logs/Player-0.log +53 -14
  10. run_logs/timers.json +220 -155
  11. run_logs/training_status.json +1 -30
SoccerTwos/SoccerTwos-130712.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc46e7c3a8ae976987e0d8263aea774c2d1ba3e7a92552de55cc4ca70b15efb3
3
+ size 646709386
SoccerTwos/SoccerTwos-14160.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d41435150bf8066294d97f42bdbcc4daaac71c9f472c02ee463ce20a8f81e9a
3
+ size 215565130
SoccerTwos/SoccerTwos-470490.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a0e840234d497bfa7e8d15f51e624526c42d36b28a548b110a16a04aa67090b
3
+ size 1768747
SoccerTwos/SoccerTwos-470490.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f763411ef7db6202a6b94fdf8070499d7e2d87a7b42649d254e574c957010553
3
+ size 28431898
SoccerTwos/checkpoint.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b3d6f8efb9422465ad2b4af990732935d6abe98d4231c252f04fe48c52940ad5
3
- size 28431030
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e47db37c9c05736730ca607b9cc6fd6da24c20f38eca97b7bfd2e35fb0b199f
3
+ size 646697542
SoccerTwos/events.out.tfevents.1703591227.cybertron.29738.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:063d8ad9f924e7e5d433747d4ddf8906d5a9323d010354a3a977613d499ee81a
3
+ size 113286
config.json CHANGED
@@ -1 +1 @@
1
- {"default_settings": null, "behaviors": {"SoccerTwos": {"trainer_type": "poca", "hyperparameters": {"batch_size": 9048, "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.x86_64", "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}, "environment_parameters": null, "checkpoint_settings": {"run_id": "SoccerTwos", "initialize_from": null, "load_model": false, "resume": true, "force": false, "train_model": false, "inference": false, "results_dir": "results"}, "torch_settings": {"device": null}, "debug": false}
 
1
+ {"default_settings": null, "behaviors": {"SoccerTwos": {"trainer_type": "poca", "hyperparameters": {"batch_size": 4096, "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": 100, "vis_encode_type": "simple", "memory": null, "goal_conditioning_type": "hyper", "deterministic": false}, "reward_signals": {"extrinsic": {"gamma": 0.95, "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": 40000000, "time_horizon": 1000, "summary_freq": 10000, "threaded": false, "self_play": {"save_steps": 50000, "team_change": 200000, "swap_steps": 2000, "window": 100, "play_against_latest_model_ratio": 0.5, "initial_elo": 1200.0}, "behavioral_cloning": null}}, "env_settings": {"env_path": "training-envs-executables/SoccerTwos/SoccerTwos.x86_64", "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}, "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 CHANGED
@@ -3,7 +3,7 @@ behaviors:
3
  SoccerTwos:
4
  trainer_type: poca
5
  hyperparameters:
6
- batch_size: 9048
7
  buffer_size: 20480
8
  learning_rate: 0.0003
9
  beta: 0.005
@@ -17,14 +17,14 @@ behaviors:
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: 0.99
28
  strength: 1.0
29
  network_settings:
30
  normalize: false
@@ -37,7 +37,7 @@ behaviors:
37
  init_path: null
38
  keep_checkpoints: 5
39
  even_checkpoints: false
40
- max_steps: 50000000
41
  time_horizon: 1000
42
  summary_freq: 10000
43
  threaded: false
@@ -45,7 +45,7 @@ behaviors:
45
  save_steps: 50000
46
  team_change: 200000
47
  swap_steps: 2000
48
- window: 10
49
  play_against_latest_model_ratio: 0.5
50
  initial_elo: 1200.0
51
  behavioral_cloning: null
@@ -73,8 +73,8 @@ checkpoint_settings:
73
  run_id: SoccerTwos
74
  initialize_from: null
75
  load_model: false
76
- resume: true
77
- force: false
78
  train_model: false
79
  inference: false
80
  results_dir: results
 
3
  SoccerTwos:
4
  trainer_type: poca
5
  hyperparameters:
6
+ batch_size: 4096
7
  buffer_size: 20480
8
  learning_rate: 0.0003
9
  beta: 0.005
 
17
  network_settings:
18
  normalize: false
19
  hidden_units: 512
20
+ num_layers: 100
21
  vis_encode_type: simple
22
  memory: null
23
  goal_conditioning_type: hyper
24
  deterministic: false
25
  reward_signals:
26
  extrinsic:
27
+ gamma: 0.95
28
  strength: 1.0
29
  network_settings:
30
  normalize: false
 
37
  init_path: null
38
  keep_checkpoints: 5
39
  even_checkpoints: false
40
+ max_steps: 40000000
41
  time_horizon: 1000
42
  summary_freq: 10000
43
  threaded: false
 
45
  save_steps: 50000
46
  team_change: 200000
47
  swap_steps: 2000
48
+ window: 100
49
  play_against_latest_model_ratio: 0.5
50
  initial_elo: 1200.0
51
  behavioral_cloning: null
 
73
  run_id: SoccerTwos
74
  initialize_from: null
75
  load_model: false
76
+ resume: false
77
+ force: true
78
  train_model: false
79
  inference: false
80
  results_dir: results
run_logs/Player-0.log CHANGED
@@ -11,7 +11,7 @@ NullGfxDevice:
11
  Renderer: Null Device
12
  Vendor: Unity Technologies
13
  Begin MonoManager ReloadAssembly
14
- - Completed reload, in 0.072 seconds
15
  ERROR: Shader Sprites/Default shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
16
  ERROR: Shader Sprites/Mask shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
17
  ERROR: Shader Legacy Shaders/VertexLit shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
@@ -36,7 +36,7 @@ ERROR: Shader Standard shader is not supported on this GPU (none of subshaders/f
36
  WARNING: Shader Unsupported: 'Standard' - All subshaders removed
37
  WARNING: Shader Did you use #pragma only_renderers and omit this platform?
38
  WARNING: Shader If subshaders removal was intentional, you may have forgotten turning Fallback off?
39
- UnloadTime: 0.448741 ms
40
  Registered Communicator in Agent.
41
  ERROR: Shader UI/Default shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
42
  requesting resize 84 x 84
@@ -45,12 +45,51 @@ TOO LONG
45
  TOO LONG
46
  TOO LONG
47
  TOO LONG
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
48
  Setting up 6 worker threads for Enlighten.
49
  Memory Statistics:
50
  [ALLOC_TEMP_TLS] TLS Allocator
51
  StackAllocators :
52
  [ALLOC_TEMP_MAIN]
53
- Peak usage frame count: [8.0 KB-16.0 KB]: 436 frames, [2.0 MB-4.0 MB]: 1 frames
54
  Initial Block Size 4.0 MB
55
  Current Block Size 4.0 MB
56
  Peak Allocated Bytes 2.1 MB
@@ -158,7 +197,7 @@ Memory Statistics:
158
  [ALLOC_TEMP_Job.Worker 9]
159
  Initial Block Size 256.0 KB
160
  Current Block Size 256.0 KB
161
- Peak Allocated Bytes 0 B
162
  Overflow Count 0
163
  [ALLOC_TEMP_Background Job.Worker 13]
164
  Initial Block Size 32.0 KB
@@ -173,7 +212,7 @@ Memory Statistics:
173
  [ALLOC_TEMP_Job.Worker 7]
174
  Initial Block Size 256.0 KB
175
  Current Block Size 256.0 KB
176
- Peak Allocated Bytes 0.7 KB
177
  Overflow Count 0
178
  [ALLOC_TEMP_Job.Worker 1]
179
  Initial Block Size 256.0 KB
@@ -216,19 +255,19 @@ Memory Statistics:
216
  Peak Allocated Bytes 0 B
217
  Overflow Count 0
218
  [ALLOC_DEFAULT] Dual Thread Allocator
219
- Peak main deferred allocation count 4
220
  [ALLOC_BUCKET]
221
  Large Block size 4.0 MB
222
  Used Block count 1
223
  Peak Allocated bytes 1.0 MB
224
  [ALLOC_DEFAULT_MAIN]
225
- Peak usage frame count: [4.0 MB-8.0 MB]: 437 frames
226
  Requested Block Size 16.0 MB
227
  Peak Block count 1
228
  Peak Allocated memory 4.7 MB
229
  Peak Large allocation bytes 0 B
230
  [ALLOC_DEFAULT_THREAD]
231
- Peak usage frame count: [16.0 MB-32.0 MB]: 437 frames
232
  Requested Block Size 16.0 MB
233
  Peak Block count 1
234
  Peak Allocated memory 17.5 MB
@@ -260,13 +299,13 @@ Memory Statistics:
260
  Used Block count 1
261
  Peak Allocated bytes 1.0 MB
262
  [ALLOC_GFX_MAIN]
263
- Peak usage frame count: [32.0 KB-64.0 KB]: 436 frames, [64.0 KB-128.0 KB]: 1 frames
264
  Requested Block Size 16.0 MB
265
  Peak Block count 1
266
  Peak Allocated memory 65.6 KB
267
  Peak Large allocation bytes 0 B
268
  [ALLOC_GFX_THREAD]
269
- Peak usage frame count: [128.0 KB-256.0 KB]: 437 frames
270
  Requested Block Size 16.0 MB
271
  Peak Block count 1
272
  Peak Allocated memory 173.5 KB
@@ -278,13 +317,13 @@ Memory Statistics:
278
  Used Block count 1
279
  Peak Allocated bytes 1.0 MB
280
  [ALLOC_CACHEOBJECTS_MAIN]
281
- Peak usage frame count: [0.5 MB-1.0 MB]: 437 frames
282
  Requested Block Size 4.0 MB
283
  Peak Block count 1
284
  Peak Allocated memory 0.6 MB
285
  Peak Large allocation bytes 0 B
286
  [ALLOC_CACHEOBJECTS_THREAD]
287
- Peak usage frame count: [0.5 MB-1.0 MB]: 436 frames, [4.0 MB-8.0 MB]: 1 frames
288
  Requested Block Size 4.0 MB
289
  Peak Block count 2
290
  Peak Allocated memory 4.5 MB
@@ -296,13 +335,13 @@ Memory Statistics:
296
  Used Block count 1
297
  Peak Allocated bytes 1.0 MB
298
  [ALLOC_TYPETREE_MAIN]
299
- Peak usage frame count: [0-1.0 KB]: 437 frames
300
  Requested Block Size 2.0 MB
301
  Peak Block count 1
302
  Peak Allocated memory 1.0 KB
303
  Peak Large allocation bytes 0 B
304
  [ALLOC_TYPETREE_THREAD]
305
- Peak usage frame count: [1.0 KB-2.0 KB]: 436 frames, [2.0 KB-4.0 KB]: 1 frames
306
  Requested Block Size 2.0 MB
307
  Peak Block count 1
308
  Peak Allocated memory 2.2 KB
 
11
  Renderer: Null Device
12
  Vendor: Unity Technologies
13
  Begin MonoManager ReloadAssembly
14
+ - Completed reload, in 0.059 seconds
15
  ERROR: Shader Sprites/Default shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
16
  ERROR: Shader Sprites/Mask shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
17
  ERROR: Shader Legacy Shaders/VertexLit shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
 
36
  WARNING: Shader Unsupported: 'Standard' - All subshaders removed
37
  WARNING: Shader Did you use #pragma only_renderers and omit this platform?
38
  WARNING: Shader If subshaders removal was intentional, you may have forgotten turning Fallback off?
39
+ UnloadTime: 1.112902 ms
40
  Registered Communicator in Agent.
41
  ERROR: Shader UI/Default shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
42
  requesting resize 84 x 84
 
45
  TOO LONG
46
  TOO LONG
47
  TOO LONG
48
+ TOO LONG
49
+ TOO LONG
50
+ TOO LONG
51
+ TOO LONG
52
+ TOO LONG
53
+ TOO LONG
54
+ TOO LONG
55
+ TOO LONG
56
+ TOO LONG
57
+ TOO LONG
58
+ TOO LONG
59
+ TOO LONG
60
+ TOO LONG
61
+ TOO LONG
62
+ TOO LONG
63
+ TOO LONG
64
+ TOO LONG
65
+ TOO LONG
66
+ TOO LONG
67
+ TOO LONG
68
+ TOO LONG
69
+ TOO LONG
70
+ TOO LONG
71
+ TOO LONG
72
+ TOO LONG
73
+ TOO LONG
74
+ TOO LONG
75
+ TOO LONG
76
+ TOO LONG
77
+ TOO LONG
78
+ TOO LONG
79
+ TOO LONG
80
+ TOO LONG
81
+ TOO LONG
82
+ TOO LONG
83
+ TOO LONG
84
+ TOO LONG
85
+ TOO LONG
86
+ TOO LONG
87
  Setting up 6 worker threads for Enlighten.
88
  Memory Statistics:
89
  [ALLOC_TEMP_TLS] TLS Allocator
90
  StackAllocators :
91
  [ALLOC_TEMP_MAIN]
92
+ Peak usage frame count: [8.0 KB-16.0 KB]: 2555 frames, [2.0 MB-4.0 MB]: 1 frames
93
  Initial Block Size 4.0 MB
94
  Current Block Size 4.0 MB
95
  Peak Allocated Bytes 2.1 MB
 
197
  [ALLOC_TEMP_Job.Worker 9]
198
  Initial Block Size 256.0 KB
199
  Current Block Size 256.0 KB
200
+ Peak Allocated Bytes 0.7 KB
201
  Overflow Count 0
202
  [ALLOC_TEMP_Background Job.Worker 13]
203
  Initial Block Size 32.0 KB
 
212
  [ALLOC_TEMP_Job.Worker 7]
213
  Initial Block Size 256.0 KB
214
  Current Block Size 256.0 KB
215
+ Peak Allocated Bytes 0 B
216
  Overflow Count 0
217
  [ALLOC_TEMP_Job.Worker 1]
218
  Initial Block Size 256.0 KB
 
255
  Peak Allocated Bytes 0 B
256
  Overflow Count 0
257
  [ALLOC_DEFAULT] Dual Thread Allocator
258
+ Peak main deferred allocation count 2
259
  [ALLOC_BUCKET]
260
  Large Block size 4.0 MB
261
  Used Block count 1
262
  Peak Allocated bytes 1.0 MB
263
  [ALLOC_DEFAULT_MAIN]
264
+ Peak usage frame count: [4.0 MB-8.0 MB]: 2556 frames
265
  Requested Block Size 16.0 MB
266
  Peak Block count 1
267
  Peak Allocated memory 4.7 MB
268
  Peak Large allocation bytes 0 B
269
  [ALLOC_DEFAULT_THREAD]
270
+ Peak usage frame count: [16.0 MB-32.0 MB]: 2556 frames
271
  Requested Block Size 16.0 MB
272
  Peak Block count 1
273
  Peak Allocated memory 17.5 MB
 
299
  Used Block count 1
300
  Peak Allocated bytes 1.0 MB
301
  [ALLOC_GFX_MAIN]
302
+ Peak usage frame count: [32.0 KB-64.0 KB]: 2555 frames, [64.0 KB-128.0 KB]: 1 frames
303
  Requested Block Size 16.0 MB
304
  Peak Block count 1
305
  Peak Allocated memory 65.6 KB
306
  Peak Large allocation bytes 0 B
307
  [ALLOC_GFX_THREAD]
308
+ Peak usage frame count: [128.0 KB-256.0 KB]: 2556 frames
309
  Requested Block Size 16.0 MB
310
  Peak Block count 1
311
  Peak Allocated memory 173.5 KB
 
317
  Used Block count 1
318
  Peak Allocated bytes 1.0 MB
319
  [ALLOC_CACHEOBJECTS_MAIN]
320
+ Peak usage frame count: [0.5 MB-1.0 MB]: 2556 frames
321
  Requested Block Size 4.0 MB
322
  Peak Block count 1
323
  Peak Allocated memory 0.6 MB
324
  Peak Large allocation bytes 0 B
325
  [ALLOC_CACHEOBJECTS_THREAD]
326
+ Peak usage frame count: [0.5 MB-1.0 MB]: 2555 frames, [4.0 MB-8.0 MB]: 1 frames
327
  Requested Block Size 4.0 MB
328
  Peak Block count 2
329
  Peak Allocated memory 4.5 MB
 
335
  Used Block count 1
336
  Peak Allocated bytes 1.0 MB
337
  [ALLOC_TYPETREE_MAIN]
338
+ Peak usage frame count: [0-1.0 KB]: 2556 frames
339
  Requested Block Size 2.0 MB
340
  Peak Block count 1
341
  Peak Allocated memory 1.0 KB
342
  Peak Large allocation bytes 0 B
343
  [ALLOC_TYPETREE_THREAD]
344
+ Peak usage frame count: [1.0 KB-2.0 KB]: 2555 frames, [2.0 KB-4.0 KB]: 1 frames
345
  Requested Block Size 2.0 MB
346
  Peak Block count 1
347
  Peak Allocated memory 2.2 KB
run_logs/timers.json CHANGED
@@ -2,189 +2,261 @@
2
  "name": "root",
3
  "gauges": {
4
  "SoccerTwos.Policy.Entropy.mean": {
5
- "value": 3.237346649169922,
6
- "min": 3.237346649169922,
7
- "max": 3.237346649169922,
8
- "count": 1
9
  },
10
  "SoccerTwos.Policy.Entropy.sum": {
11
- "value": 103595.09375,
12
- "min": 103595.09375,
13
- "max": 103595.09375,
14
- "count": 1
15
  },
16
  "SoccerTwos.Environment.EpisodeLength.mean": {
17
- "value": 673.2,
18
- "min": 673.2,
19
- "max": 673.2,
20
- "count": 1
21
  },
22
  "SoccerTwos.Environment.EpisodeLength.sum": {
23
- "value": 26928.0,
24
- "min": 26928.0,
25
- "max": 26928.0,
26
- "count": 1
27
  },
28
  "SoccerTwos.Self-play.ELO.mean": {
29
- "value": 1207.6368942738368,
30
- "min": 1207.6368942738368,
31
- "max": 1207.6368942738368,
32
- "count": 1
33
  },
34
  "SoccerTwos.Self-play.ELO.sum": {
35
- "value": 12076.368942738369,
36
- "min": 12076.368942738369,
37
- "max": 12076.368942738369,
38
- "count": 1
39
  },
40
  "SoccerTwos.Step.mean": {
41
- "value": 209569.0,
42
- "min": 209569.0,
43
- "max": 209569.0,
44
- "count": 1
45
  },
46
  "SoccerTwos.Step.sum": {
47
- "value": 209569.0,
48
- "min": 209569.0,
49
- "max": 209569.0,
50
- "count": 1
51
  },
52
  "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": {
53
- "value": 0.016558969393372536,
54
- "min": 0.016558969393372536,
55
- "max": 0.016558969393372536,
56
- "count": 1
57
  },
58
  "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": {
59
- "value": 0.21526660025119781,
60
- "min": 0.21526660025119781,
61
- "max": 0.21526660025119781,
62
- "count": 1
63
  },
64
  "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": {
65
- "value": 0.0151954535394907,
66
- "min": 0.0151954535394907,
67
- "max": 0.0151954535394907,
68
- "count": 1
69
  },
70
  "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": {
71
- "value": 0.19754089415073395,
72
- "min": 0.19754089415073395,
73
- "max": 0.19754089415073395,
74
- "count": 1
75
  },
76
  "SoccerTwos.Environment.CumulativeReward.mean": {
77
  "value": 0.0,
78
  "min": 0.0,
79
  "max": 0.0,
80
- "count": 1
81
  },
82
  "SoccerTwos.Environment.CumulativeReward.sum": {
83
  "value": 0.0,
84
  "min": 0.0,
85
  "max": 0.0,
86
- "count": 1
87
  },
88
  "SoccerTwos.Policy.ExtrinsicReward.mean": {
89
- "value": -0.28603077164063084,
90
- "min": -0.28603077164063084,
91
- "max": -0.28603077164063084,
92
- "count": 1
93
  },
94
  "SoccerTwos.Policy.ExtrinsicReward.sum": {
95
- "value": -3.7184000313282013,
96
- "min": -3.7184000313282013,
97
- "max": -3.7184000313282013,
98
- "count": 1
99
  },
100
  "SoccerTwos.Environment.GroupCumulativeReward.mean": {
101
- "value": -0.28603077164063084,
102
- "min": -0.28603077164063084,
103
- "max": -0.28603077164063084,
104
- "count": 1
105
  },
106
  "SoccerTwos.Environment.GroupCumulativeReward.sum": {
107
- "value": -3.7184000313282013,
108
- "min": -3.7184000313282013,
109
- "max": -3.7184000313282013,
110
- "count": 1
111
  },
112
  "SoccerTwos.IsTraining.mean": {
113
  "value": 1.0,
114
  "min": 1.0,
115
  "max": 1.0,
116
- "count": 1
117
  },
118
  "SoccerTwos.IsTraining.sum": {
119
  "value": 1.0,
120
  "min": 1.0,
121
  "max": 1.0,
122
- "count": 1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
123
  }
124
  },
125
  "metadata": {
126
  "timer_format_version": "0.1.0",
127
- "start_time_seconds": "1703589972",
128
  "python_version": "3.10.12 (main, Jul 5 2023, 18:54:27) [GCC 11.2.0]",
129
- "command_line_arguments": "/home/cybertron/anaconda3/envs/rl/bin/mlagents-learn /home/cybertron/Desktop/ml-agents/config/poca/SoccerTwos.yaml --env=training-envs-executables/SoccerTwos/SoccerTwos.x86_64 --run-id=SoccerTwos --no-graphics --resume",
130
  "mlagents_version": "1.0.0",
131
  "mlagents_envs_version": "1.0.0",
132
  "communication_protocol_version": "1.5.0",
133
  "pytorch_version": "2.1.1",
134
  "numpy_version": "1.21.2",
135
- "end_time_seconds": "1703589997"
136
  },
137
- "total": 25.27916620600081,
138
  "count": 1,
139
- "self": 0.030008082998392638,
140
  "children": {
141
  "run_training.setup": {
142
- "total": 0.011635144001047593,
143
  "count": 1,
144
- "self": 0.011635144001047593
145
  },
146
  "TrainerController.start_learning": {
147
- "total": 25.23752297900137,
148
  "count": 1,
149
- "self": 0.02163675902193063,
150
  "children": {
151
  "TrainerController._reset_env": {
152
- "total": 1.8909534080012236,
153
- "count": 2,
154
- "self": 1.8909534080012236
155
  },
156
  "TrainerController.advance": {
157
- "total": 23.136872530980327,
158
- "count": 1461,
159
- "self": 0.021862540888832882,
160
  "children": {
161
  "env_step": {
162
- "total": 22.047475695031608,
163
- "count": 1461,
164
- "self": 16.652173617960216,
165
  "children": {
166
  "SubprocessEnvManager._take_step": {
167
- "total": 5.3811991840047995,
168
- "count": 1461,
169
- "self": 0.12078593808109872,
170
  "children": {
171
  "TorchPolicy.evaluate": {
172
- "total": 5.260413245923701,
173
- "count": 2908,
174
- "self": 5.260413245923701
175
  }
176
  }
177
  },
178
  "workers": {
179
- "total": 0.014102893066592515,
180
- "count": 1460,
181
  "self": 0.0,
182
  "children": {
183
  "worker_root": {
184
- "total": 25.02045437198103,
185
- "count": 1460,
186
  "is_parallel": true,
187
- "self": 11.028041653993569,
188
  "children": {
189
  "run_training.setup": {
190
  "total": 0.0,
@@ -193,48 +265,48 @@
193
  "self": 0.0,
194
  "children": {
195
  "steps_from_proto": {
196
- "total": 0.0017820489993027877,
197
  "count": 2,
198
  "is_parallel": true,
199
- "self": 0.00039754300451022573,
200
  "children": {
201
  "_process_rank_one_or_two_observation": {
202
- "total": 0.001384505994792562,
203
  "count": 8,
204
  "is_parallel": true,
205
- "self": 0.001384505994792562
206
  }
207
  }
208
  },
209
  "UnityEnvironment.step": {
210
- "total": 0.019492260998958955,
211
  "count": 1,
212
  "is_parallel": true,
213
- "self": 0.0004075500000908505,
214
  "children": {
215
  "UnityEnvironment._generate_step_input": {
216
- "total": 0.0003197090009052772,
217
  "count": 1,
218
  "is_parallel": true,
219
- "self": 0.0003197090009052772
220
  },
221
  "communicator.exchange": {
222
- "total": 0.017475802000262775,
223
  "count": 1,
224
  "is_parallel": true,
225
- "self": 0.017475802000262775
226
  },
227
  "steps_from_proto": {
228
- "total": 0.0012891999977000523,
229
  "count": 2,
230
  "is_parallel": true,
231
- "self": 0.0002657109980646055,
232
  "children": {
233
  "_process_rank_one_or_two_observation": {
234
- "total": 0.0010234889996354468,
235
  "count": 8,
236
  "is_parallel": true,
237
- "self": 0.0010234889996354468
238
  }
239
  }
240
  }
@@ -242,49 +314,35 @@
242
  }
243
  }
244
  },
245
- "steps_from_proto": {
246
- "total": 0.0012812650020350702,
247
- "count": 2,
248
- "is_parallel": true,
249
- "self": 0.0002530899982957635,
250
- "children": {
251
- "_process_rank_one_or_two_observation": {
252
- "total": 0.0010281750037393067,
253
- "count": 8,
254
- "is_parallel": true,
255
- "self": 0.0010281750037393067
256
- }
257
- }
258
- },
259
  "UnityEnvironment.step": {
260
- "total": 13.991131452985428,
261
- "count": 1459,
262
  "is_parallel": true,
263
- "self": 0.7237382169769262,
264
  "children": {
265
  "UnityEnvironment._generate_step_input": {
266
- "total": 0.41074541095804307,
267
- "count": 1459,
268
  "is_parallel": true,
269
- "self": 0.41074541095804307
270
  },
271
  "communicator.exchange": {
272
- "total": 10.776612414010742,
273
- "count": 1459,
274
  "is_parallel": true,
275
- "self": 10.776612414010742
276
  },
277
  "steps_from_proto": {
278
- "total": 2.0800354110397166,
279
- "count": 2918,
280
  "is_parallel": true,
281
- "self": 0.384538546029944,
282
  "children": {
283
  "_process_rank_one_or_two_observation": {
284
- "total": 1.6954968650097726,
285
- "count": 11672,
286
  "is_parallel": true,
287
- "self": 1.6954968650097726
288
  }
289
  }
290
  }
@@ -297,33 +355,40 @@
297
  }
298
  },
299
  "trainer_advance": {
300
- "total": 1.067534295059886,
301
- "count": 1460,
302
- "self": 0.10599143002400524,
303
  "children": {
304
  "process_trajectory": {
305
- "total": 0.9615428650358808,
306
- "count": 1460,
307
- "self": 0.9615428650358808
 
 
 
 
 
 
 
 
 
 
 
 
308
  }
309
  }
310
  }
311
  }
312
  },
313
- "trainer_threads": {
314
- "total": 8.399983926210552e-07,
315
- "count": 1,
316
- "self": 8.399983926210552e-07
317
- },
318
  "TrainerController._save_models": {
319
- "total": 0.18805944099949556,
320
  "count": 1,
321
- "self": 0.0014082630004850216,
322
  "children": {
323
  "RLTrainer._checkpoint": {
324
- "total": 0.18665117799901054,
325
  "count": 1,
326
- "self": 0.18665117799901054
327
  }
328
  }
329
  }
 
2
  "name": "root",
3
  "gauges": {
4
  "SoccerTwos.Policy.Entropy.mean": {
5
+ "value": 3.2872161865234375,
6
+ "min": 3.283947467803955,
7
+ "max": 3.2958364486694336,
8
+ "count": 13
9
  },
10
  "SoccerTwos.Policy.Entropy.sum": {
11
+ "value": 43759.421875,
12
+ "min": 42781.015625,
13
+ "max": 105466.765625,
14
+ "count": 13
15
  },
16
  "SoccerTwos.Environment.EpisodeLength.mean": {
17
+ "value": 811.5,
18
+ "min": 494.7142857142857,
19
+ "max": 906.6,
20
+ "count": 13
21
  },
22
  "SoccerTwos.Environment.EpisodeLength.sum": {
23
+ "value": 19476.0,
24
+ "min": 13852.0,
25
+ "max": 27716.0,
26
+ "count": 13
27
  },
28
  "SoccerTwos.Self-play.ELO.mean": {
29
+ "value": 1202.1297790220663,
30
+ "min": 1198.6916415169974,
31
+ "max": 1202.9673173073832,
32
+ "count": 13
33
  },
34
  "SoccerTwos.Self-play.ELO.sum": {
35
+ "value": 7212.778674132397,
36
+ "min": 4798.067989990133,
37
+ "max": 14399.635556919653,
38
+ "count": 13
39
  },
40
  "SoccerTwos.Step.mean": {
41
+ "value": 129712.0,
42
+ "min": 9876.0,
43
+ "max": 129712.0,
44
+ "count": 13
45
  },
46
  "SoccerTwos.Step.sum": {
47
+ "value": 129712.0,
48
+ "min": 9876.0,
49
+ "max": 129712.0,
50
+ "count": 13
51
  },
52
  "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": {
53
+ "value": -0.0014713257551193237,
54
+ "min": -0.0014713257551193237,
55
+ "max": 0.05451347306370735,
56
+ "count": 13
57
  },
58
  "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": {
59
+ "value": -0.017655909061431885,
60
+ "min": -0.017655909061431885,
61
+ "max": 0.8722155690193176,
62
+ "count": 13
63
  },
64
  "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": {
65
+ "value": -0.0014713257551193237,
66
+ "min": -0.0014713257551193237,
67
+ "max": 0.05451347306370735,
68
+ "count": 13
69
  },
70
  "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": {
71
+ "value": -0.017655909061431885,
72
+ "min": -0.017655909061431885,
73
+ "max": 0.8722155690193176,
74
+ "count": 13
75
  },
76
  "SoccerTwos.Environment.CumulativeReward.mean": {
77
  "value": 0.0,
78
  "min": 0.0,
79
  "max": 0.0,
80
+ "count": 13
81
  },
82
  "SoccerTwos.Environment.CumulativeReward.sum": {
83
  "value": 0.0,
84
  "min": 0.0,
85
  "max": 0.0,
86
+ "count": 13
87
  },
88
  "SoccerTwos.Policy.ExtrinsicReward.mean": {
89
+ "value": -0.5,
90
+ "min": -0.5,
91
+ "max": 0.2147230735191932,
92
+ "count": 13
93
  },
94
  "SoccerTwos.Policy.ExtrinsicReward.sum": {
95
+ "value": -6.0,
96
+ "min": -6.0,
97
+ "max": 2.7913999557495117,
98
+ "count": 13
99
  },
100
  "SoccerTwos.Environment.GroupCumulativeReward.mean": {
101
+ "value": -0.5,
102
+ "min": -0.5,
103
+ "max": 0.2147230735191932,
104
+ "count": 13
105
  },
106
  "SoccerTwos.Environment.GroupCumulativeReward.sum": {
107
+ "value": -6.0,
108
+ "min": -6.0,
109
+ "max": 2.7913999557495117,
110
+ "count": 13
111
  },
112
  "SoccerTwos.IsTraining.mean": {
113
  "value": 1.0,
114
  "min": 1.0,
115
  "max": 1.0,
116
+ "count": 13
117
  },
118
  "SoccerTwos.IsTraining.sum": {
119
  "value": 1.0,
120
  "min": 1.0,
121
  "max": 1.0,
122
+ "count": 13
123
+ },
124
+ "SoccerTwos.Losses.PolicyLoss.mean": {
125
+ "value": 0.012890987443582465,
126
+ "min": 0.00523900882108137,
127
+ "max": 0.013176202323908608,
128
+ "count": 5
129
+ },
130
+ "SoccerTwos.Losses.PolicyLoss.sum": {
131
+ "value": 0.012890987443582465,
132
+ "min": 0.00523900882108137,
133
+ "max": 0.013176202323908608,
134
+ "count": 5
135
+ },
136
+ "SoccerTwos.Losses.ValueLoss.mean": {
137
+ "value": 0.001682706782594323,
138
+ "min": 0.0012276781412462394,
139
+ "max": 0.003556596456716458,
140
+ "count": 5
141
+ },
142
+ "SoccerTwos.Losses.ValueLoss.sum": {
143
+ "value": 0.001682706782594323,
144
+ "min": 0.0012276781412462394,
145
+ "max": 0.003556596456716458,
146
+ "count": 5
147
+ },
148
+ "SoccerTwos.Losses.BaselineLoss.mean": {
149
+ "value": 0.001682706782594323,
150
+ "min": 0.0012276781412462394,
151
+ "max": 0.003556596456716458,
152
+ "count": 5
153
+ },
154
+ "SoccerTwos.Losses.BaselineLoss.sum": {
155
+ "value": 0.001682706782594323,
156
+ "min": 0.0012276781412462394,
157
+ "max": 0.003556596456716458,
158
+ "count": 5
159
+ },
160
+ "SoccerTwos.Policy.LearningRate.mean": {
161
+ "value": 0.0003,
162
+ "min": 0.0003,
163
+ "max": 0.0003,
164
+ "count": 5
165
+ },
166
+ "SoccerTwos.Policy.LearningRate.sum": {
167
+ "value": 0.0003,
168
+ "min": 0.0003,
169
+ "max": 0.0003,
170
+ "count": 5
171
+ },
172
+ "SoccerTwos.Policy.Epsilon.mean": {
173
+ "value": 0.20000000000000007,
174
+ "min": 0.20000000000000007,
175
+ "max": 0.20000000000000007,
176
+ "count": 5
177
+ },
178
+ "SoccerTwos.Policy.Epsilon.sum": {
179
+ "value": 0.20000000000000007,
180
+ "min": 0.20000000000000007,
181
+ "max": 0.20000000000000007,
182
+ "count": 5
183
+ },
184
+ "SoccerTwos.Policy.Beta.mean": {
185
+ "value": 0.005,
186
+ "min": 0.005,
187
+ "max": 0.005,
188
+ "count": 5
189
+ },
190
+ "SoccerTwos.Policy.Beta.sum": {
191
+ "value": 0.005,
192
+ "min": 0.005,
193
+ "max": 0.005,
194
+ "count": 5
195
  }
196
  },
197
  "metadata": {
198
  "timer_format_version": "0.1.0",
199
+ "start_time_seconds": "1703591226",
200
  "python_version": "3.10.12 (main, Jul 5 2023, 18:54:27) [GCC 11.2.0]",
201
+ "command_line_arguments": "/home/cybertron/anaconda3/envs/rl/bin/mlagents-learn /home/cybertron/Desktop/ml-agents/config/poca/SoccerTwos.yaml --env=training-envs-executables/SoccerTwos/SoccerTwos.x86_64 --run-id=SoccerTwos --no-graphics --force",
202
  "mlagents_version": "1.0.0",
203
  "mlagents_envs_version": "1.0.0",
204
  "communication_protocol_version": "1.5.0",
205
  "pytorch_version": "2.1.1",
206
  "numpy_version": "1.21.2",
207
+ "end_time_seconds": "1703591510"
208
  },
209
+ "total": 284.3635966809961,
210
  "count": 1,
211
+ "self": 0.21882801299216226,
212
  "children": {
213
  "run_training.setup": {
214
+ "total": 0.012106276000849903,
215
  "count": 1,
216
+ "self": 0.012106276000849903
217
  },
218
  "TrainerController.start_learning": {
219
+ "total": 284.1326623920031,
220
  "count": 1,
221
+ "self": 0.13699659491248894,
222
  "children": {
223
  "TrainerController._reset_env": {
224
+ "total": 2.0738567980006337,
225
+ "count": 1,
226
+ "self": 2.0738567980006337
227
  },
228
  "TrainerController.advance": {
229
+ "total": 279.10317788109387,
230
+ "count": 8577,
231
+ "self": 0.1445215416897554,
232
  "children": {
233
  "env_step": {
234
+ "total": 198.89272247839108,
235
+ "count": 8577,
236
+ "self": 100.6482229002795,
237
  "children": {
238
  "SubprocessEnvManager._take_step": {
239
+ "total": 98.15602654993563,
240
+ "count": 8577,
241
+ "self": 0.9458562079526018,
242
  "children": {
243
  "TorchPolicy.evaluate": {
244
+ "total": 97.21017034198303,
245
+ "count": 17038,
246
+ "self": 97.21017034198303
247
  }
248
  }
249
  },
250
  "workers": {
251
+ "total": 0.08847302817594027,
252
+ "count": 8577,
253
  "self": 0.0,
254
  "children": {
255
  "worker_root": {
256
+ "total": 279.9393852657813,
257
+ "count": 8577,
258
  "is_parallel": true,
259
+ "self": 195.6091849016957,
260
  "children": {
261
  "run_training.setup": {
262
  "total": 0.0,
 
265
  "self": 0.0,
266
  "children": {
267
  "steps_from_proto": {
268
+ "total": 0.0018885289973695762,
269
  "count": 2,
270
  "is_parallel": true,
271
+ "self": 0.0004485460085561499,
272
  "children": {
273
  "_process_rank_one_or_two_observation": {
274
+ "total": 0.0014399829888134263,
275
  "count": 8,
276
  "is_parallel": true,
277
+ "self": 0.0014399829888134263
278
  }
279
  }
280
  },
281
  "UnityEnvironment.step": {
282
+ "total": 0.019859125000948552,
283
  "count": 1,
284
  "is_parallel": true,
285
+ "self": 0.00042441000550752506,
286
  "children": {
287
  "UnityEnvironment._generate_step_input": {
288
+ "total": 0.0005746079987147823,
289
  "count": 1,
290
  "is_parallel": true,
291
+ "self": 0.0005746079987147823
292
  },
293
  "communicator.exchange": {
294
+ "total": 0.017522023001220077,
295
  "count": 1,
296
  "is_parallel": true,
297
+ "self": 0.017522023001220077
298
  },
299
  "steps_from_proto": {
300
+ "total": 0.0013380839955061674,
301
  "count": 2,
302
  "is_parallel": true,
303
+ "self": 0.0002767289915936999,
304
  "children": {
305
  "_process_rank_one_or_two_observation": {
306
+ "total": 0.0010613550039124675,
307
  "count": 8,
308
  "is_parallel": true,
309
+ "self": 0.0010613550039124675
310
  }
311
  }
312
  }
 
314
  }
315
  }
316
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
317
  "UnityEnvironment.step": {
318
+ "total": 84.33020036408561,
319
+ "count": 8576,
320
  "is_parallel": true,
321
+ "self": 4.363296162802726,
322
  "children": {
323
  "UnityEnvironment._generate_step_input": {
324
+ "total": 2.5144244331968366,
325
+ "count": 8576,
326
  "is_parallel": true,
327
+ "self": 2.5144244331968366
328
  },
329
  "communicator.exchange": {
330
+ "total": 64.61790556683991,
331
+ "count": 8576,
332
  "is_parallel": true,
333
+ "self": 64.61790556683991
334
  },
335
  "steps_from_proto": {
336
+ "total": 12.834574201246141,
337
+ "count": 17152,
338
  "is_parallel": true,
339
+ "self": 2.373371579713421,
340
  "children": {
341
  "_process_rank_one_or_two_observation": {
342
+ "total": 10.46120262153272,
343
+ "count": 68608,
344
  "is_parallel": true,
345
+ "self": 10.46120262153272
346
  }
347
  }
348
  }
 
355
  }
356
  },
357
  "trainer_advance": {
358
+ "total": 80.06593386101304,
359
+ "count": 8577,
360
+ "self": 1.3821683759160805,
361
  "children": {
362
  "process_trajectory": {
363
+ "total": 11.169169155109557,
364
+ "count": 8577,
365
+ "self": 11.169169155109557
366
+ },
367
+ "_update_policy": {
368
+ "total": 67.5145963299874,
369
+ "count": 6,
370
+ "self": 12.408137330006866,
371
+ "children": {
372
+ "TorchPOCAOptimizer.update": {
373
+ "total": 55.10645899998053,
374
+ "count": 76,
375
+ "self": 55.10645899998053
376
+ }
377
+ }
378
  }
379
  }
380
  }
381
  }
382
  },
 
 
 
 
 
383
  "TrainerController._save_models": {
384
+ "total": 2.8186311179961194,
385
  "count": 1,
386
+ "self": 2.9962000553496182e-05,
387
  "children": {
388
  "RLTrainer._checkpoint": {
389
+ "total": 2.818601155995566,
390
  "count": 1,
391
+ "self": 2.818601155995566
392
  }
393
  }
394
  }
run_logs/training_status.json CHANGED
@@ -1,35 +1,6 @@
1
  {
2
  "SoccerTwos": {
3
- "elo": 1205.850037125567,
4
- "checkpoints": [
5
- {
6
- "steps": 203078,
7
- "file_path": "results/SoccerTwos/SoccerTwos/SoccerTwos-203078.onnx",
8
- "reward": 0.0,
9
- "creation_time": 1703589965.3602862,
10
- "auxillary_file_paths": [
11
- "results/SoccerTwos/SoccerTwos/SoccerTwos-203078.pt"
12
- ]
13
- },
14
- {
15
- "steps": 218994,
16
- "file_path": "results/SoccerTwos/SoccerTwos/SoccerTwos-218994.onnx",
17
- "reward": 0.0,
18
- "creation_time": 1703589997.6629596,
19
- "auxillary_file_paths": [
20
- "results/SoccerTwos/SoccerTwos/SoccerTwos-218994.pt"
21
- ]
22
- }
23
- ],
24
- "final_checkpoint": {
25
- "steps": 218994,
26
- "file_path": "results/SoccerTwos/SoccerTwos.onnx",
27
- "reward": 0.0,
28
- "creation_time": 1703589997.6629596,
29
- "auxillary_file_paths": [
30
- "results/SoccerTwos/SoccerTwos/SoccerTwos-218994.pt"
31
- ]
32
- }
33
  },
34
  "metadata": {
35
  "stats_format_version": "0.3.0",
 
1
  {
2
  "SoccerTwos": {
3
+ "elo": 1200.953350551564
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4
  },
5
  "metadata": {
6
  "stats_format_version": "0.3.0",