Hugging Face's logo

brettgoehre
/
poca-SoccerTwosMemory

Reinforcement Learning
ml-agents
TensorBoard
ONNX
SoccerTwos
deep-reinforcement-learning
ML-Agents-SoccerTwos
Model card Files Metrics Community
brettgoehre commited on
Commit
471ead9
·
verified ·
1 Parent(s): 5f8d9bc

21th Push

Browse files
Files changed (11) hide show
  1. SoccerTwos.onnx +1 -1
  2. SoccerTwos/SoccerTwos-137499987.onnx +3 -0
  3. SoccerTwos/SoccerTwos-137499987.pt +3 -0
  4. SoccerTwos/checkpoint.pt +1 -1
  5. SoccerTwos/events.out.tfevents.1726768788.Bretts-MBP.home.45043.0 +2 -2
  6. SoccerTwos/events.out.tfevents.1726804282.Bretts-MBP.home.27094.0 +3 -0
  7. SoccerTwos/events.out.tfevents.1726852029.Bretts-MBP.home.31082.0 +3 -0
  8. SoccerTwos/events.out.tfevents.1726856751.Bretts-MBP.home.93804.0 +3 -0
  9. run_logs/Player-0.log +2 -2
  10. run_logs/timers.json +190 -190
  11. run_logs/training_status.json +8 -8
SoccerTwos.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:78f977c50b1c06932b41e0682c26e7c6bcdb983b4a14e29bb811d0b906cb3549
3
  size 2351062
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe445853fd261a72561815d3f0df34585061e80f258877a46e6cfb872503241e
3
  size 2351062
SoccerTwos/SoccerTwos-137499987.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce2c361e2db3c42615f60488856221b0351edd2f3a64e9c4b35544bac8151973
3
+ size 2351062
SoccerTwos/SoccerTwos-137499987.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c508a049711f35e6e8fc7ca5de5dd6c8b539af3f01fb2de404a6a70c076883a
3
+ size 36303358
SoccerTwos/checkpoint.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f039352616289da615097520d58a69402f5ac0851da7cb3fd3a6ac91c0bf5f84
3
  size 36301318
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87dfdf0a17b7f2ee2263a29472450daaa8afc206f05fb7b44411ead182ab141d
3
  size 36301318
SoccerTwos/events.out.tfevents.1726768788.Bretts-MBP.home.45043.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:be2cd16d8b7364d899b5a2c287ff74a4e67df8fc028c804a7bb55b36e9438816
3
- size 7366145
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:943aabf89e198a43c9ee9000062727307a218f798333859393b0b550cc100c1e
3
+ size 8378741
SoccerTwos/events.out.tfevents.1726804282.Bretts-MBP.home.27094.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f4aced31962226360b38509e419267057bdc88a10b3311db550bac68d53e988
3
+ size 9001235
SoccerTwos/events.out.tfevents.1726852029.Bretts-MBP.home.31082.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:acf5c99b9e085609c2103a2b5a99c893c1978457059a05799c936ec6d444786b
3
+ size 771589
SoccerTwos/events.out.tfevents.1726856751.Bretts-MBP.home.93804.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d81b38d6302e02ae3a43ada162d92bc8e85dd8107eb49222a78ead5406e0d27d
3
+ size 613337
run_logs/Player-0.log CHANGED
@@ -10,7 +10,7 @@ NullGfxDevice:
10
  Renderer: Null Device
11
  Vendor: Unity Technologies
12
  Begin MonoManager ReloadAssembly
13
- - Completed reload, in 0.053 seconds
14
  ERROR: Shader Sprites/Default shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
15
  ERROR: Shader Sprites/Mask shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
16
  ERROR: Shader Legacy Shaders/VertexLit shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
@@ -35,7 +35,7 @@ ERROR: Shader Standard shader is not supported on this GPU (none of subshaders/f
35
  WARNING: Shader Unsupported: 'Standard' - All subshaders removed
36
  WARNING: Shader Did you use #pragma only_renderers and omit this platform?
37
  WARNING: Shader If subshaders removal was intentional, you may have forgotten turning Fallback off?
38
- UnloadTime: 0.244708 ms
39
  Registered Communicator in Agent.
40
  ERROR: Shader UI/Default shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
41
  Default audio device was changed, but the audio system failed to initialize it. Attempting to reset sound system.
 
10
  Renderer: Null Device
11
  Vendor: Unity Technologies
12
  Begin MonoManager ReloadAssembly
13
+ - Completed reload, in 0.049 seconds
14
  ERROR: Shader Sprites/Default shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
15
  ERROR: Shader Sprites/Mask shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
16
  ERROR: Shader Legacy Shaders/VertexLit shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
 
35
  WARNING: Shader Unsupported: 'Standard' - All subshaders removed
36
  WARNING: Shader Did you use #pragma only_renderers and omit this platform?
37
  WARNING: Shader If subshaders removal was intentional, you may have forgotten turning Fallback off?
38
+ UnloadTime: 0.294958 ms
39
  Registered Communicator in Agent.
40
  ERROR: Shader UI/Default shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
41
  Default audio device was changed, but the audio system failed to initialize it. Attempting to reset sound system.
run_logs/timers.json CHANGED
@@ -2,123 +2,123 @@
2
  "name": "root",
3
  "gauges": {
4
  "SoccerTwos.Policy.Entropy.mean": {
5
- "value": 2.255478620529175,
6
- "min": 2.230623483657837,
7
- "max": 2.276839017868042,
8
  "count": 73
9
  },
10
  "SoccerTwos.Policy.Entropy.sum": {
11
- "value": 48357.4609375,
12
- "min": 32327.259765625,
13
- "max": 53696.97265625,
14
  "count": 73
15
  },
16
  "SoccerTwos.Environment.EpisodeLength.mean": {
17
- "value": 62.35,
18
- "min": 50.5360824742268,
19
- "max": 73.58208955223881,
20
  "count": 73
21
  },
22
  "SoccerTwos.Environment.EpisodeLength.sum": {
23
- "value": 19952.0,
24
- "min": 12480.0,
25
- "max": 20628.0,
26
  "count": 73
27
  },
28
  "SoccerTwos.Self-play.ELO.mean": {
29
- "value": 200.00727787930185,
30
- "min": 160.59296511354657,
31
- "max": 203.92215340488602,
32
  "count": 73
33
  },
34
  "SoccerTwos.Self-play.ELO.sum": {
35
- "value": 32001.1644606883,
36
- "min": 18902.286914257813,
37
- "max": 35633.72897126356,
38
  "count": 73
39
  },
40
  "SoccerTwos.Step.mean": {
41
- "value": 120069898.0,
42
- "min": 119349896.0,
43
- "max": 120069898.0,
44
  "count": 73
45
  },
46
  "SoccerTwos.Step.sum": {
47
- "value": 120069898.0,
48
- "min": 119349896.0,
49
- "max": 120069898.0,
50
  "count": 73
51
  },
52
  "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": {
53
- "value": -0.1278150975704193,
54
- "min": -0.170237734913826,
55
- "max": 0.08623138070106506,
56
  "count": 73
57
  },
58
  "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": {
59
- "value": -20.322601318359375,
60
- "min": -24.869155883789062,
61
- "max": 14.400640487670898,
62
  "count": 73
63
  },
64
  "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": {
65
- "value": -0.1259104162454605,
66
- "min": -0.16831789910793304,
67
- "max": 0.08471225947141647,
68
  "count": 73
69
  },
70
  "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": {
71
- "value": -20.019756317138672,
72
- "min": -25.439720153808594,
73
- "max": 14.146947860717773,
74
  "count": 73
75
  },
76
  "SoccerTwos.Policy.CuriosityBaselineEstimate.mean": {
77
- "value": 0.02077561430633068,
78
- "min": 0.008352533914148808,
79
- "max": 0.02189120277762413,
80
  "count": 73
81
  },
82
  "SoccerTwos.Policy.CuriosityBaselineEstimate.sum": {
83
- "value": 3.3033227920532227,
84
- "min": 1.1812084913253784,
85
- "max": 3.915364980697632,
86
  "count": 73
87
  },
88
  "SoccerTwos.Policy.CuriosityValueEstimate.mean": {
89
- "value": 0.020814398303627968,
90
- "min": 0.00840409193187952,
91
- "max": 0.0219256691634655,
92
  "count": 73
93
  },
94
  "SoccerTwos.Policy.CuriosityValueEstimate.sum": {
95
- "value": 3.3094892501831055,
96
- "min": 1.1851625442504883,
97
- "max": 3.9218106269836426,
98
  "count": 73
99
  },
100
  "SoccerTwos.Policy.RndBaselineEstimate.mean": {
101
- "value": 0.0068628787994384766,
102
- "min": 0.006545525975525379,
103
- "max": 0.007221938576549292,
104
  "count": 73
105
  },
106
  "SoccerTwos.Policy.RndBaselineEstimate.sum": {
107
- "value": 1.0911977291107178,
108
- "min": 0.6152794361114502,
109
- "max": 1.3019368648529053,
110
  "count": 73
111
  },
112
  "SoccerTwos.Policy.RndValueEstimate.mean": {
113
- "value": 0.006880544126033783,
114
- "min": 0.0065576122142374516,
115
- "max": 0.007235044613480568,
116
  "count": 73
117
  },
118
  "SoccerTwos.Policy.RndValueEstimate.sum": {
119
- "value": 1.0940065383911133,
120
- "min": 0.6164155602455139,
121
- "max": 1.3072426319122314,
122
  "count": 73
123
  },
124
  "SoccerTwos.Environment.CumulativeReward.mean": {
@@ -134,51 +134,51 @@
134
  "count": 73
135
  },
136
  "SoccerTwos.Policy.ExtrinsicReward.mean": {
137
- "value": -0.14068931003786483,
138
- "min": -0.25714838889337355,
139
- "max": 0.2380424672610139,
140
  "count": 73
141
  },
142
  "SoccerTwos.Policy.ExtrinsicReward.sum": {
143
- "value": -22.369600296020508,
144
- "min": -39.98679995536804,
145
- "max": 34.75420022010803,
146
  "count": 73
147
  },
148
  "SoccerTwos.Policy.CuriosityReward.mean": {
149
- "value": 0.549969781143298,
150
  "min": 0.0,
151
- "max": 0.9045167695183087,
152
  "count": 73
153
  },
154
  "SoccerTwos.Policy.CuriosityReward.sum": {
155
- "value": 87.44519520178437,
156
  "min": 0.0,
157
- "max": 125.58684200793505,
158
  "count": 73
159
  },
160
  "SoccerTwos.Policy.RndReward.mean": {
161
- "value": 0.220354071941297,
162
- "min": 0.1871111254860676,
163
- "max": 0.27255435835625696,
164
  "count": 73
165
  },
166
  "SoccerTwos.Policy.RndReward.sum": {
167
- "value": 35.036297438666224,
168
- "min": 20.47128874901682,
169
- "max": 37.64491491857916,
170
  "count": 73
171
  },
172
  "SoccerTwos.Environment.GroupCumulativeReward.mean": {
173
- "value": -0.14068931003786483,
174
- "min": -0.25714838889337355,
175
- "max": 0.2380424672610139,
176
  "count": 73
177
  },
178
  "SoccerTwos.Environment.GroupCumulativeReward.sum": {
179
- "value": -22.369600296020508,
180
- "min": -39.98679995536804,
181
- "max": 34.75420022010803,
182
  "count": 73
183
  },
184
  "SoccerTwos.IsTraining.mean": {
@@ -194,39 +194,39 @@
194
  "count": 73
195
  },
196
  "SoccerTwos.Losses.PolicyLoss.mean": {
197
- "value": 0.013634561997605488,
198
- "min": 0.00939689805575957,
199
- "max": 0.017807006215055782,
200
  "count": 4
201
  },
202
  "SoccerTwos.Losses.PolicyLoss.sum": {
203
- "value": 0.013634561997605488,
204
- "min": 0.00939689805575957,
205
- "max": 0.017807006215055782,
206
  "count": 4
207
  },
208
  "SoccerTwos.Losses.ValueLoss.mean": {
209
- "value": 0.019645538957168657,
210
- "min": 0.019196889828890562,
211
- "max": 0.019724952677885692,
212
  "count": 4
213
  },
214
  "SoccerTwos.Losses.ValueLoss.sum": {
215
- "value": 0.019645538957168657,
216
- "min": 0.019196889828890562,
217
- "max": 0.019724952677885692,
218
  "count": 4
219
  },
220
  "SoccerTwos.Losses.BaselineLoss.mean": {
221
- "value": 0.02021182266374429,
222
- "min": 0.01983766940732797,
223
- "max": 0.02021182266374429,
224
  "count": 4
225
  },
226
  "SoccerTwos.Losses.BaselineLoss.sum": {
227
- "value": 0.02021182266374429,
228
- "min": 0.01983766940732797,
229
- "max": 0.02021182266374429,
230
  "count": 4
231
  },
232
  "SoccerTwos.Policy.LearningRate.mean": {
@@ -254,57 +254,57 @@
254
  "count": 4
255
  },
256
  "SoccerTwos.Policy.Beta.mean": {
257
- "value": 0.004940154883094001,
258
- "min": 0.004940154883094001,
259
- "max": 0.004940374242496,
260
  "count": 4
261
  },
262
  "SoccerTwos.Policy.Beta.sum": {
263
- "value": 0.004940154883094001,
264
- "min": 0.004940154883094001,
265
- "max": 0.004940374242496,
266
  "count": 4
267
  },
268
  "SoccerTwos.Losses.CuriosityForwardLoss.mean": {
269
- "value": 0.04491656149427096,
270
- "min": 0.04491656149427096,
271
- "max": 0.09600517495224874,
272
  "count": 4
273
  },
274
  "SoccerTwos.Losses.CuriosityForwardLoss.sum": {
275
- "value": 0.04491656149427096,
276
- "min": 0.04491656149427096,
277
- "max": 0.09600517495224874,
278
  "count": 4
279
  },
280
  "SoccerTwos.Losses.CuriosityInverseLoss.mean": {
281
- "value": 1.5401246945063274,
282
- "min": 1.5401246945063274,
283
- "max": 1.7537684241930644,
284
  "count": 4
285
  },
286
  "SoccerTwos.Losses.CuriosityInverseLoss.sum": {
287
- "value": 1.5401246945063274,
288
- "min": 1.5401246945063274,
289
- "max": 1.7537684241930644,
290
  "count": 4
291
  },
292
  "SoccerTwos.Losses.RNDLoss.mean": {
293
- "value": 0.01586015336215496,
294
- "min": 0.01586015336215496,
295
- "max": 0.017172401770949364,
296
  "count": 4
297
  },
298
  "SoccerTwos.Losses.RNDLoss.sum": {
299
- "value": 0.01586015336215496,
300
- "min": 0.01586015336215496,
301
- "max": 0.017172401770949364,
302
  "count": 4
303
  }
304
  },
305
  "metadata": {
306
  "timer_format_version": "0.1.0",
307
- "start_time_seconds": "1726760363",
308
  "python_version": "3.10.12 (main, Aug 23 2024, 16:29:11) [Clang 15.0.0 (clang-1500.3.9.4)]",
309
  "command_line_arguments": "/Users/brettgohre/Desktop/huggingface/ml-agents/games/bin/mlagents-learn ./config/poca/sept/three.yaml --env=./SoccerTwos/SoccerTwos.app --run-id=sept-run-01 --resume --no-graphics",
310
  "mlagents_version": "1.1.0.dev0",
@@ -312,103 +312,103 @@
312
  "communication_protocol_version": "1.5.0",
313
  "pytorch_version": "2.4.0",
314
  "numpy_version": "1.23.5",
315
- "end_time_seconds": "1726763505"
316
  },
317
- "total": 3141.775703000021,
318
  "count": 1,
319
- "self": 0.019297625054605305,
320
  "children": {
321
  "run_training.setup": {
322
- "total": 0.019264457980170846,
323
  "count": 1,
324
- "self": 0.019264457980170846
325
  },
326
  "TrainerController.start_learning": {
327
- "total": 3141.7371409169864,
328
  "count": 1,
329
- "self": 0.616179323522374,
330
  "children": {
331
  "TrainerController._reset_env": {
332
- "total": 6.17303312709555,
333
  "count": 74,
334
- "self": 6.17303312709555
335
  },
336
  "TrainerController.advance": {
337
- "total": 3134.5881778413896,
338
- "count": 54375,
339
- "self": 0.46005123551003635,
340
  "children": {
341
  "env_step": {
342
- "total": 2235.4445955763804,
343
- "count": 54375,
344
- "self": 2155.006646992173,
345
  "children": {
346
  "SubprocessEnvManager._take_step": {
347
- "total": 80.10751251596957,
348
- "count": 54375,
349
- "self": 3.1288182924035937,
350
  "children": {
351
  "TorchPolicy.evaluate": {
352
- "total": 76.97869422356598,
353
- "count": 99600,
354
- "self": 76.97869422356598
355
  }
356
  }
357
  },
358
  "workers": {
359
- "total": 0.3304360682377592,
360
- "count": 54375,
361
  "self": 0.0,
362
  "children": {
363
  "worker_root": {
364
- "total": 3000.0345093504293,
365
- "count": 54375,
366
  "is_parallel": true,
367
- "self": 905.6446920702001,
368
  "children": {
369
  "steps_from_proto": {
370
- "total": 0.1013583323219791,
371
  "count": 148,
372
  "is_parallel": true,
373
- "self": 0.013186299125663936,
374
  "children": {
375
  "_process_rank_one_or_two_observation": {
376
- "total": 0.08817203319631517,
377
  "count": 592,
378
  "is_parallel": true,
379
- "self": 0.08817203319631517
380
  }
381
  }
382
  },
383
  "UnityEnvironment.step": {
384
- "total": 2094.288458947907,
385
- "count": 54375,
386
  "is_parallel": true,
387
- "self": 5.375599776511081,
388
  "children": {
389
  "UnityEnvironment._generate_step_input": {
390
- "total": 41.997091523138806,
391
- "count": 54375,
392
  "is_parallel": true,
393
- "self": 41.997091523138806
394
  },
395
  "communicator.exchange": {
396
- "total": 1978.6759033377748,
397
- "count": 54375,
398
  "is_parallel": true,
399
- "self": 1978.6759033377748
400
  },
401
  "steps_from_proto": {
402
- "total": 68.23986431048252,
403
- "count": 108750,
404
  "is_parallel": true,
405
- "self": 8.25137705472298,
406
  "children": {
407
  "_process_rank_one_or_two_observation": {
408
- "total": 59.98848725575954,
409
- "count": 435000,
410
  "is_parallel": true,
411
- "self": 59.98848725575954
412
  }
413
  }
414
  }
@@ -421,31 +421,31 @@
421
  }
422
  },
423
  "trainer_advance": {
424
- "total": 898.6835310294991,
425
- "count": 54375,
426
- "self": 5.210828961106017,
427
  "children": {
428
  "process_trajectory": {
429
- "total": 163.99640706938226,
430
- "count": 54375,
431
- "self": 163.62267173535656,
432
  "children": {
433
  "RLTrainer._checkpoint": {
434
- "total": 0.3737353340256959,
435
- "count": 2,
436
- "self": 0.3737353340256959
437
  }
438
  }
439
  },
440
  "_update_policy": {
441
- "total": 729.4762949990109,
442
  "count": 5,
443
- "self": 71.07644628593698,
444
  "children": {
445
  "TorchPOCAOptimizer.update": {
446
- "total": 658.3998487130739,
447
- "count": 59,
448
- "self": 658.3998487130739
449
  }
450
  }
451
  }
@@ -454,19 +454,19 @@
454
  }
455
  },
456
  "trainer_threads": {
457
- "total": 8.749775588512421e-07,
458
  "count": 1,
459
- "self": 8.749775588512421e-07
460
  },
461
  "TrainerController._save_models": {
462
- "total": 0.359749750001356,
463
  "count": 1,
464
- "self": 0.007718999986536801,
465
  "children": {
466
  "RLTrainer._checkpoint": {
467
- "total": 0.3520307500148192,
468
  "count": 1,
469
- "self": 0.3520307500148192
470
  }
471
  }
472
  }
 
2
  "name": "root",
3
  "gauges": {
4
  "SoccerTwos.Policy.Entropy.mean": {
5
+ "value": 2.013183355331421,
6
+ "min": 1.9969563484191895,
7
+ "max": 2.0372118949890137,
8
  "count": 73
9
  },
10
  "SoccerTwos.Policy.Entropy.sum": {
11
+ "value": 44515.51171875,
12
+ "min": 36500.25,
13
+ "max": 49285.63671875,
14
  "count": 73
15
  },
16
  "SoccerTwos.Environment.EpisodeLength.mean": {
17
+ "value": 67.79452054794521,
18
+ "min": 53.31521739130435,
19
+ "max": 84.3,
20
  "count": 73
21
  },
22
  "SoccerTwos.Environment.EpisodeLength.sum": {
23
+ "value": 19796.0,
24
+ "min": 15476.0,
25
+ "max": 20316.0,
26
  "count": 73
27
  },
28
  "SoccerTwos.Self-play.ELO.mean": {
29
+ "value": 226.28569760195467,
30
+ "min": 218.50462034121867,
31
+ "max": 246.13217966828475,
32
  "count": 73
33
  },
34
  "SoccerTwos.Self-play.ELO.sum": {
35
+ "value": 33037.71184988538,
36
+ "min": 25960.237905448797,
37
+ "max": 43074.66286109567,
38
  "count": 73
39
  },
40
  "SoccerTwos.Step.mean": {
41
+ "value": 137269836.0,
42
+ "min": 136549984.0,
43
+ "max": 137269836.0,
44
  "count": 73
45
  },
46
  "SoccerTwos.Step.sum": {
47
+ "value": 137269836.0,
48
+ "min": 136549984.0,
49
+ "max": 137269836.0,
50
  "count": 73
51
  },
52
  "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": {
53
+ "value": -0.03741655498743057,
54
+ "min": -0.1345166116952896,
55
+ "max": 0.07524154335260391,
56
  "count": 73
57
  },
58
  "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": {
59
+ "value": -5.425400733947754,
60
+ "min": -21.791690826416016,
61
+ "max": 12.529707908630371,
62
  "count": 73
63
  },
64
  "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": {
65
+ "value": -0.03817114606499672,
66
+ "min": -0.1369057297706604,
67
+ "max": 0.07655324786901474,
68
  "count": 73
69
  },
70
  "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": {
71
+ "value": -5.534816265106201,
72
+ "min": -22.178728103637695,
73
+ "max": 12.63769817352295,
74
  "count": 73
75
  },
76
  "SoccerTwos.Policy.CuriosityBaselineEstimate.mean": {
77
+ "value": 0.022628309205174446,
78
+ "min": 0.009245038963854313,
79
+ "max": 0.02489888109266758,
80
  "count": 73
81
  },
82
  "SoccerTwos.Policy.CuriosityBaselineEstimate.sum": {
83
+ "value": 3.281104803085327,
84
+ "min": 1.175850749015808,
85
+ "max": 4.356385231018066,
86
  "count": 73
87
  },
88
  "SoccerTwos.Policy.CuriosityValueEstimate.mean": {
89
+ "value": 0.02262844890356064,
90
+ "min": 0.009272996336221695,
91
+ "max": 0.024900028482079506,
92
  "count": 73
93
  },
94
  "SoccerTwos.Policy.CuriosityValueEstimate.sum": {
95
+ "value": 3.281125068664551,
96
+ "min": 1.1755244731903076,
97
+ "max": 4.360075950622559,
98
  "count": 73
99
  },
100
  "SoccerTwos.Policy.RndBaselineEstimate.mean": {
101
+ "value": 0.006809944752603769,
102
+ "min": 0.006517637521028519,
103
+ "max": 0.007117425091564655,
104
  "count": 73
105
  },
106
  "SoccerTwos.Policy.RndBaselineEstimate.sum": {
107
+ "value": 0.9874419569969177,
108
+ "min": 0.7655903697013855,
109
+ "max": 1.253800392150879,
110
  "count": 73
111
  },
112
  "SoccerTwos.Policy.RndValueEstimate.mean": {
113
+ "value": 0.006798093672841787,
114
+ "min": 0.006493216380476952,
115
+ "max": 0.0071043046191334724,
116
  "count": 73
117
  },
118
  "SoccerTwos.Policy.RndValueEstimate.sum": {
119
+ "value": 0.985723614692688,
120
+ "min": 0.7624234557151794,
121
+ "max": 1.2502329349517822,
122
  "count": 73
123
  },
124
  "SoccerTwos.Environment.CumulativeReward.mean": {
 
134
  "count": 73
135
  },
136
  "SoccerTwos.Policy.ExtrinsicReward.mean": {
137
+ "value": -0.10613517309057302,
138
+ "min": -0.2378641970363664,
139
+ "max": 0.25296296190332485,
140
  "count": 73
141
  },
142
  "SoccerTwos.Policy.ExtrinsicReward.sum": {
143
+ "value": -15.389600098133087,
144
+ "min": -38.53399991989136,
145
+ "max": 34.14999985694885,
146
  "count": 73
147
  },
148
  "SoccerTwos.Policy.CuriosityReward.mean": {
149
+ "value": 0.6438479568038521,
150
  "min": 0.0,
151
+ "max": 1.4505344783266385,
152
  "count": 73
153
  },
154
  "SoccerTwos.Policy.CuriosityReward.sum": {
155
+ "value": 93.35795373655856,
156
  "min": 0.0,
157
+ "max": 175.77997627854347,
158
  "count": 73
159
  },
160
  "SoccerTwos.Policy.RndReward.mean": {
161
+ "value": 0.24667795498951756,
162
+ "min": 0.1884528696688043,
163
+ "max": 0.30395433399050187,
164
  "count": 73
165
  },
166
  "SoccerTwos.Policy.RndReward.sum": {
167
+ "value": 35.768303473480046,
168
+ "min": 27.970972408540547,
169
+ "max": 37.355501416139305,
170
  "count": 73
171
  },
172
  "SoccerTwos.Environment.GroupCumulativeReward.mean": {
173
+ "value": -0.10613517309057302,
174
+ "min": -0.2378641970363664,
175
+ "max": 0.25296296190332485,
176
  "count": 73
177
  },
178
  "SoccerTwos.Environment.GroupCumulativeReward.sum": {
179
+ "value": -15.389600098133087,
180
+ "min": -38.53399991989136,
181
+ "max": 34.14999985694885,
182
  "count": 73
183
  },
184
  "SoccerTwos.IsTraining.mean": {
 
194
  "count": 73
195
  },
196
  "SoccerTwos.Losses.PolicyLoss.mean": {
197
+ "value": 0.005562455854184615,
198
+ "min": 0.005562455854184615,
199
+ "max": 0.014488776432699524,
200
  "count": 4
201
  },
202
  "SoccerTwos.Losses.PolicyLoss.sum": {
203
+ "value": 0.005562455854184615,
204
+ "min": 0.005562455854184615,
205
+ "max": 0.014488776432699524,
206
  "count": 4
207
  },
208
  "SoccerTwos.Losses.ValueLoss.mean": {
209
+ "value": 0.019290074861298006,
210
+ "min": 0.01834545625994603,
211
+ "max": 0.019290074861298006,
212
  "count": 4
213
  },
214
  "SoccerTwos.Losses.ValueLoss.sum": {
215
+ "value": 0.019290074861298006,
216
+ "min": 0.01834545625994603,
217
+ "max": 0.019290074861298006,
218
  "count": 4
219
  },
220
  "SoccerTwos.Losses.BaselineLoss.mean": {
221
+ "value": 0.019729879529525835,
222
+ "min": 0.018973414786159992,
223
+ "max": 0.019729879529525835,
224
  "count": 4
225
  },
226
  "SoccerTwos.Losses.BaselineLoss.sum": {
227
+ "value": 0.019729879529525835,
228
+ "min": 0.018973414786159992,
229
+ "max": 0.019729879529525835,
230
  "count": 4
231
  },
232
  "SoccerTwos.Policy.LearningRate.mean": {
 
254
  "count": 4
255
  },
256
  "SoccerTwos.Policy.Beta.mean": {
257
+ "value": 0.004931572211836001,
258
+ "min": 0.004931572211836001,
259
+ "max": 0.0049317921091599996,
260
  "count": 4
261
  },
262
  "SoccerTwos.Policy.Beta.sum": {
263
+ "value": 0.004931572211836001,
264
+ "min": 0.004931572211836001,
265
+ "max": 0.0049317921091599996,
266
  "count": 4
267
  },
268
  "SoccerTwos.Losses.CuriosityForwardLoss.mean": {
269
+ "value": 0.0483112009242177,
270
+ "min": 0.0483112009242177,
271
+ "max": 0.1390404033785065,
272
  "count": 4
273
  },
274
  "SoccerTwos.Losses.CuriosityForwardLoss.sum": {
275
+ "value": 0.0483112009242177,
276
+ "min": 0.0483112009242177,
277
+ "max": 0.1390404033785065,
278
  "count": 4
279
  },
280
  "SoccerTwos.Losses.CuriosityInverseLoss.mean": {
281
+ "value": 1.3768685559431713,
282
+ "min": 1.3768685559431713,
283
+ "max": 1.6764498154322307,
284
  "count": 4
285
  },
286
  "SoccerTwos.Losses.CuriosityInverseLoss.sum": {
287
+ "value": 1.3768685559431713,
288
+ "min": 1.3768685559431713,
289
+ "max": 1.6764498154322307,
290
  "count": 4
291
  },
292
  "SoccerTwos.Losses.RNDLoss.mean": {
293
+ "value": 0.015932753682136536,
294
+ "min": 0.01591380685567856,
295
+ "max": 0.017608152702450752,
296
  "count": 4
297
  },
298
  "SoccerTwos.Losses.RNDLoss.sum": {
299
+ "value": 0.015932753682136536,
300
+ "min": 0.01591380685567856,
301
+ "max": 0.017608152702450752,
302
  "count": 4
303
  }
304
  },
305
  "metadata": {
306
  "timer_format_version": "0.1.0",
307
+ "start_time_seconds": "1726852025",
308
  "python_version": "3.10.12 (main, Aug 23 2024, 16:29:11) [Clang 15.0.0 (clang-1500.3.9.4)]",
309
  "command_line_arguments": "/Users/brettgohre/Desktop/huggingface/ml-agents/games/bin/mlagents-learn ./config/poca/sept/three.yaml --env=./SoccerTwos/SoccerTwos.app --run-id=sept-run-01 --resume --no-graphics",
310
  "mlagents_version": "1.1.0.dev0",
 
312
  "communication_protocol_version": "1.5.0",
313
  "pytorch_version": "2.4.0",
314
  "numpy_version": "1.23.5",
315
+ "end_time_seconds": "1726855117"
316
  },
317
+ "total": 3091.2703225830337,
318
  "count": 1,
319
+ "self": 0.013066916028037667,
320
  "children": {
321
  "run_training.setup": {
322
+ "total": 0.019244292052462697,
323
  "count": 1,
324
+ "self": 0.019244292052462697
325
  },
326
  "TrainerController.start_learning": {
327
+ "total": 3091.238011374953,
328
  "count": 1,
329
+ "self": 0.6549470070749521,
330
  "children": {
331
  "TrainerController._reset_env": {
332
+ "total": 6.774538544355892,
333
  "count": 74,
334
+ "self": 6.774538544355892
335
  },
336
  "TrainerController.advance": {
337
+ "total": 3083.530861656647,
338
+ "count": 54443,
339
+ "self": 0.5134963177843019,
340
  "children": {
341
  "env_step": {
342
+ "total": 2252.5134539136197,
343
+ "count": 54443,
344
+ "self": 2168.886104057776,
345
  "children": {
346
  "SubprocessEnvManager._take_step": {
347
+ "total": 83.2792097517522,
348
+ "count": 54443,
349
+ "self": 3.171278899302706,
350
  "children": {
351
  "TorchPolicy.evaluate": {
352
+ "total": 80.1079308524495,
353
+ "count": 100012,
354
+ "self": 80.1079308524495
355
  }
356
  }
357
  },
358
  "workers": {
359
+ "total": 0.34814010409172624,
360
+ "count": 54443,
361
  "self": 0.0,
362
  "children": {
363
  "worker_root": {
364
+ "total": 3024.2247854277957,
365
+ "count": 54443,
366
  "is_parallel": true,
367
+ "self": 915.7552356258966,
368
  "children": {
369
  "steps_from_proto": {
370
+ "total": 0.10174066864419729,
371
  "count": 148,
372
  "is_parallel": true,
373
+ "self": 0.013216774677857757,
374
  "children": {
375
  "_process_rank_one_or_two_observation": {
376
+ "total": 0.08852389396633953,
377
  "count": 592,
378
  "is_parallel": true,
379
+ "self": 0.08852389396633953
380
  }
381
  }
382
  },
383
  "UnityEnvironment.step": {
384
+ "total": 2108.367809133255,
385
+ "count": 54443,
386
  "is_parallel": true,
387
+ "self": 5.666770220734179,
388
  "children": {
389
  "UnityEnvironment._generate_step_input": {
390
+ "total": 40.98215689067729,
391
+ "count": 54443,
392
  "is_parallel": true,
393
+ "self": 40.98215689067729
394
  },
395
  "communicator.exchange": {
396
+ "total": 1992.8668013456045,
397
+ "count": 54443,
398
  "is_parallel": true,
399
+ "self": 1992.8668013456045
400
  },
401
  "steps_from_proto": {
402
+ "total": 68.85208067623898,
403
+ "count": 108886,
404
  "is_parallel": true,
405
+ "self": 8.394826258183457,
406
  "children": {
407
  "_process_rank_one_or_two_observation": {
408
+ "total": 60.45725441805553,
409
+ "count": 435544,
410
  "is_parallel": true,
411
+ "self": 60.45725441805553
412
  }
413
  }
414
  }
 
421
  }
422
  },
423
  "trainer_advance": {
424
+ "total": 830.5039114252431,
425
+ "count": 54443,
426
+ "self": 5.782571685384028,
427
  "children": {
428
  "process_trajectory": {
429
+ "total": 163.31070186395664,
430
+ "count": 54443,
431
+ "self": 163.1295437810477,
432
  "children": {
433
  "RLTrainer._checkpoint": {
434
+ "total": 0.18115808290895075,
435
+ "count": 1,
436
+ "self": 0.18115808290895075
437
  }
438
  }
439
  },
440
  "_update_policy": {
441
+ "total": 661.4106378759025,
442
  "count": 5,
443
+ "self": 64.14400474901777,
444
  "children": {
445
  "TorchPOCAOptimizer.update": {
446
+ "total": 597.2666331268847,
447
+ "count": 53,
448
+ "self": 597.2666331268847
449
  }
450
  }
451
  }
 
454
  }
455
  },
456
  "trainer_threads": {
457
+ "total": 1.5419209375977516e-06,
458
  "count": 1,
459
+ "self": 1.5419209375977516e-06
460
  },
461
  "TrainerController._save_models": {
462
+ "total": 0.277662624954246,
463
  "count": 1,
464
+ "self": 0.006202915916219354,
465
  "children": {
466
  "RLTrainer._checkpoint": {
467
+ "total": 0.27145970903802663,
468
  "count": 1,
469
+ "self": 0.27145970903802663
470
  }
471
  }
472
  }
run_logs/training_status.json CHANGED
@@ -1,24 +1,24 @@
1
  {
2
  "SoccerTwos": {
3
- "elo": 187.00288389830894,
4
  "checkpoints": [
5
  {
6
- "steps": 120077520,
7
- "file_path": "results/sept-run-01/SoccerTwos/SoccerTwos-120077520.onnx",
8
  "reward": null,
9
- "creation_time": 1726763505.4426508,
10
  "auxillary_file_paths": [
11
- "results/sept-run-01/SoccerTwos/SoccerTwos-120077520.pt"
12
  ]
13
  }
14
  ],
15
  "final_checkpoint": {
16
- "steps": 120077520,
17
  "file_path": "results/sept-run-01/SoccerTwos.onnx",
18
  "reward": null,
19
- "creation_time": 1726763505.4426508,
20
  "auxillary_file_paths": [
21
- "results/sept-run-01/SoccerTwos/SoccerTwos-120077520.pt"
22
  ]
23
  }
24
  },
 
1
  {
2
  "SoccerTwos": {
3
+ "elo": 217.20535211320427,
4
  "checkpoints": [
5
  {
6
+ "steps": 137276584,
7
+ "file_path": "results/sept-run-01/SoccerTwos/SoccerTwos-137276584.onnx",
8
  "reward": null,
9
+ "creation_time": 1726855117.018749,
10
  "auxillary_file_paths": [
11
+ "results/sept-run-01/SoccerTwos/SoccerTwos-137276584.pt"
12
  ]
13
  }
14
  ],
15
  "final_checkpoint": {
16
+ "steps": 137276584,
17
  "file_path": "results/sept-run-01/SoccerTwos.onnx",
18
  "reward": null,
19
+ "creation_time": 1726855117.018749,
20
  "auxillary_file_paths": [
21
+ "results/sept-run-01/SoccerTwos/SoccerTwos-137276584.pt"
22
  ]
23
  }
24
  },