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
1f43bf7
•
1 Parent(s): a3c2e28

17th Push

Browse files
Files changed (13) hide show
  1. SoccerTwos.onnx +1 -1
  2. SoccerTwos/SoccerTwos-70913316.onnx +3 -0
  3. SoccerTwos/SoccerTwos-70913316.pt +3 -0
  4. SoccerTwos/SoccerTwos-80999976.onnx +3 -0
  5. SoccerTwos/SoccerTwos-80999976.pt +3 -0
  6. SoccerTwos/checkpoint.pt +1 -1
  7. SoccerTwos/events.out.tfevents.1725907664.Bretts-MacBook-Pro.local.17177.0 +2 -2
  8. SoccerTwos/events.out.tfevents.1725935564.Bretts-MBP.home.11827.0 +3 -0
  9. SoccerTwos/events.out.tfevents.1725978590.Bretts-MacBook-Pro.local.83048.0 +3 -0
  10. SoccerTwos/events.out.tfevents.1725991963.Bretts-MacBook-Pro.local.9961.0 +3 -0
  11. run_logs/Player-0.log +2 -2
  12. run_logs/timers.json +244 -244
  13. 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 sha256:6a95661fd6afe7ef254c9998667140cf35f8fda731bb604b573002d761ebf39d
3
  size 2351062
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e548b98cec70c6be0928cb54d3e302bf0b6f36f3c196795d13eaa5e5ccd0de79
3
  size 2351062
SoccerTwos/SoccerTwos-70913316.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:585aae44922094d5cd5376c7c164a42862b4c5ab91f28ee3e8616f7e142ce41e
3
+ size 2351062
SoccerTwos/SoccerTwos-70913316.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ada69007c34a9d1bfa28339a33469bb33c6f4ef3009f685ccca3a1d6f917887
3
+ size 36303154
SoccerTwos/SoccerTwos-80999976.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ab2bbb471add6bc7826149c27246f7077c58edd5ed951617e59bce70458d3d7
3
+ size 2351062
SoccerTwos/SoccerTwos-80999976.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aff09021ad6df126d01e8ee4aa50bd0bb1283f66a22b264881240a21edf6e069
3
+ size 36303154
SoccerTwos/checkpoint.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1728916d9b1a257cf3d7e63307e275159ffb69043dc9d347e1de5bcff30f9d1f
3
  size 36301318
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1a57ee2f8c6f84015679e2c1d67cbf1eae1bb1db78acc31c034024bf437bc30
3
  size 36301318
SoccerTwos/events.out.tfevents.1725907664.Bretts-MacBook-Pro.local.17177.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:66f2f73f602a6276602b0fbf288041776a7eee484cf6e7c2b5744927ff2b305e
3
- size 1415113
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41da532e7fc5627a8b06077311b0524a554fb2817fc97dc91b0cbc22ee775100
3
+ size 2903007
SoccerTwos/events.out.tfevents.1725935564.Bretts-MBP.home.11827.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e703e4e5b1d45ed93dbd170d2bed41ac29f67093915da3cdd1eb0798e8ca419
3
+ size 7429231
SoccerTwos/events.out.tfevents.1725978590.Bretts-MacBook-Pro.local.83048.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8316eec72063b838bebeb13fef3c6e3283a6241b157c296b2a86bd6733ab8ac0
3
+ size 2164317
SoccerTwos/events.out.tfevents.1725991963.Bretts-MacBook-Pro.local.9961.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c0984d5c1e2b06455906e92152c1e225f8a84074f0ec416ba3869e394dd85b13
3
+ size 1225321
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 0.049 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)
@@ -35,6 +35,6 @@ 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: 0.215542 ms
39
  Registered Communicator in Agent.
40
  ERROR: Shader UI/Default shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
 
10
  Renderer: Null Device
11
  Vendor: Unity Technologies
12
  Begin MonoManager ReloadAssembly
13
+ - Completed reload, in 0.298 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: 1.378958 ms
39
  Registered Communicator in Agent.
40
  ERROR: Shader UI/Default shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
run_logs/timers.json CHANGED
@@ -2,309 +2,309 @@
2
  "name": "root",
3
  "gauges": {
4
  "SoccerTwos.Policy.Entropy.mean": {
5
- "value": 1.7686567306518555,
6
- "min": 1.7510119676589966,
7
- "max": 1.8440067768096924,
8
- "count": 327
9
  },
10
  "SoccerTwos.Policy.Entropy.sum": {
11
- "value": 31298.150390625,
12
- "min": 24316.234375,
13
- "max": 41254.28515625,
14
- "count": 327
15
  },
16
  "SoccerTwos.Environment.EpisodeLength.mean": {
17
- "value": 74.109375,
18
- "min": 48.009803921568626,
19
- "max": 118.73170731707317,
20
- "count": 327
21
  },
22
  "SoccerTwos.Environment.EpisodeLength.sum": {
23
- "value": 18972.0,
24
- "min": 11700.0,
25
- "max": 20688.0,
26
- "count": 327
27
  },
28
  "SoccerTwos.Self-play.ELO.mean": {
29
- "value": 266.0829627429584,
30
- "min": 225.53563294232023,
31
- "max": 276.5450906214763,
32
- "count": 327
33
  },
34
  "SoccerTwos.Self-play.ELO.sum": {
35
- "value": 34058.61923109867,
36
- "min": 19173.664054367902,
37
- "max": 52610.192182673396,
38
- "count": 327
39
  },
40
  "SoccerTwos.Step.mean": {
41
- "value": 68159944.0,
42
- "min": 64899972.0,
43
- "max": 68159944.0,
44
- "count": 327
45
  },
46
  "SoccerTwos.Step.sum": {
47
- "value": 68159944.0,
48
- "min": 64899972.0,
49
- "max": 68159944.0,
50
- "count": 327
51
  },
52
  "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": {
53
- "value": -0.1408787965774536,
54
- "min": -0.17371758818626404,
55
- "max": 0.10241176933050156,
56
- "count": 327
57
  },
58
  "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": {
59
- "value": -18.032485961914062,
60
- "min": -26.852680206298828,
61
- "max": 17.205177307128906,
62
- "count": 327
63
  },
64
  "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": {
65
- "value": -0.14279958605766296,
66
- "min": -0.1761186122894287,
67
- "max": 0.1031416580080986,
68
- "count": 327
69
  },
70
  "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": {
71
- "value": -18.27834701538086,
72
- "min": -27.158761978149414,
73
- "max": 17.32779884338379,
74
- "count": 327
75
  },
76
  "SoccerTwos.Policy.CuriosityBaselineEstimate.mean": {
77
- "value": 0.0003857061092276126,
78
- "min": 6.33984527667053e-05,
79
- "max": 0.0034244852140545845,
80
- "count": 327
81
  },
82
  "SoccerTwos.Policy.CuriosityBaselineEstimate.sum": {
83
- "value": 0.049370381981134415,
84
- "min": 0.00760781392455101,
85
- "max": 0.5821624994277954,
86
- "count": 327
87
  },
88
  "SoccerTwos.Policy.CuriosityValueEstimate.mean": {
89
- "value": 0.000361915968824178,
90
- "min": 2.3119073375710286e-05,
91
- "max": 0.0034281888511031866,
92
- "count": 327
93
  },
94
  "SoccerTwos.Policy.CuriosityValueEstimate.sum": {
95
- "value": 0.04632524400949478,
96
- "min": 0.002751169726252556,
97
- "max": 0.5755681991577148,
98
- "count": 327
99
  },
100
  "SoccerTwos.Policy.RndBaselineEstimate.mean": {
101
- "value": 4.1172832425218076e-05,
102
- "min": -0.00035084373666904867,
103
- "max": 0.00048200145829468966,
104
- "count": 327
105
  },
106
  "SoccerTwos.Policy.RndBaselineEstimate.sum": {
107
- "value": 0.005270122550427914,
108
- "min": -0.049976252019405365,
109
- "max": 0.0657184049487114,
110
- "count": 327
111
  },
112
  "SoccerTwos.Policy.RndValueEstimate.mean": {
113
- "value": 3.166253009112552e-05,
114
- "min": -0.00035453695454634726,
115
- "max": 0.0004717961710412055,
116
- "count": 327
117
  },
118
  "SoccerTwos.Policy.RndValueEstimate.sum": {
119
- "value": 0.004052803851664066,
120
- "min": -0.04941067472100258,
121
- "max": 0.06547817587852478,
122
- "count": 327
123
  },
124
  "SoccerTwos.Environment.CumulativeReward.mean": {
125
  "value": 0.0,
126
  "min": 0.0,
127
  "max": 0.0,
128
- "count": 327
129
  },
130
  "SoccerTwos.Environment.CumulativeReward.sum": {
131
  "value": 0.0,
132
  "min": 0.0,
133
  "max": 0.0,
134
- "count": 327
135
  },
136
  "SoccerTwos.Policy.ExtrinsicReward.mean": {
137
- "value": -0.22408437635749578,
138
- "min": -0.47111408475419164,
139
- "max": 0.2896025639313918,
140
- "count": 327
141
  },
142
  "SoccerTwos.Policy.ExtrinsicReward.sum": {
143
- "value": -28.68280017375946,
144
- "min": -66.89820003509521,
145
- "max": 45.17799997329712,
146
- "count": 327
147
  },
148
  "SoccerTwos.Policy.CuriosityReward.mean": {
149
- "value": 3.367126811482102e-06,
150
  "min": 0.0,
151
- "max": 5.515648820693418e-06,
152
- "count": 327
153
  },
154
  "SoccerTwos.Policy.CuriosityReward.sum": {
155
- "value": 0.00043099223186970903,
156
  "min": 0.0,
157
- "max": 0.000711518697869451,
158
- "count": 327
159
  },
160
  "SoccerTwos.Policy.RndReward.mean": {
161
- "value": 1.4164559778429009e-06,
162
- "min": 8.796986468423745e-07,
163
- "max": 2.3542934765070793e-06,
164
- "count": 327
165
  },
166
  "SoccerTwos.Policy.RndReward.sum": {
167
- "value": 0.00018130636516389131,
168
- "min": 0.00010200310338603913,
169
- "max": 0.0001954063585500876,
170
- "count": 327
171
  },
172
  "SoccerTwos.Environment.GroupCumulativeReward.mean": {
173
- "value": -0.22408437635749578,
174
- "min": -0.47111408475419164,
175
- "max": 0.2896025639313918,
176
- "count": 327
177
  },
178
  "SoccerTwos.Environment.GroupCumulativeReward.sum": {
179
- "value": -28.68280017375946,
180
- "min": -66.89820003509521,
181
- "max": 45.17799997329712,
182
- "count": 327
183
  },
184
  "SoccerTwos.IsTraining.mean": {
185
  "value": 1.0,
186
  "min": 1.0,
187
  "max": 1.0,
188
- "count": 327
189
  },
190
  "SoccerTwos.IsTraining.sum": {
191
  "value": 1.0,
192
  "min": 1.0,
193
  "max": 1.0,
194
- "count": 327
195
  },
196
  "SoccerTwos.Losses.PolicyLoss.mean": {
197
- "value": 0.0075759488851569285,
198
- "min": 0.005062628969123277,
199
- "max": 0.01822049516583017,
200
- "count": 22
201
  },
202
  "SoccerTwos.Losses.PolicyLoss.sum": {
203
- "value": 0.0075759488851569285,
204
- "min": 0.005062628969123277,
205
- "max": 0.01822049516583017,
206
- "count": 22
207
  },
208
  "SoccerTwos.Losses.ValueLoss.mean": {
209
- "value": 0.01655620491753022,
210
- "min": 0.01596076476077239,
211
- "max": 0.017545947494606178,
212
- "count": 22
213
  },
214
  "SoccerTwos.Losses.ValueLoss.sum": {
215
- "value": 0.01655620491753022,
216
- "min": 0.01596076476077239,
217
- "max": 0.017545947494606178,
218
- "count": 22
219
  },
220
  "SoccerTwos.Losses.BaselineLoss.mean": {
221
- "value": 0.016997046923885744,
222
- "min": 0.016401407153656084,
223
- "max": 0.017796412110328674,
224
- "count": 22
225
  },
226
  "SoccerTwos.Losses.BaselineLoss.sum": {
227
- "value": 0.016997046923885744,
228
- "min": 0.016401407153656084,
229
- "max": 0.017796412110328674,
230
- "count": 22
231
  },
232
  "SoccerTwos.Policy.LearningRate.mean": {
233
  "value": 0.0003,
234
  "min": 0.0003,
235
  "max": 0.0003,
236
- "count": 22
237
  },
238
  "SoccerTwos.Policy.LearningRate.sum": {
239
  "value": 0.0003,
240
  "min": 0.0003,
241
  "max": 0.0003,
242
- "count": 22
243
  },
244
  "SoccerTwos.Policy.Epsilon.mean": {
245
  "value": 0.20000000000000004,
246
  "min": 0.20000000000000004,
247
  "max": 0.20000000000000004,
248
- "count": 22
249
  },
250
  "SoccerTwos.Policy.Epsilon.sum": {
251
  "value": 0.20000000000000004,
252
  "min": 0.20000000000000004,
253
  "max": 0.20000000000000004,
254
- "count": 22
255
  },
256
  "SoccerTwos.Policy.Beta.mean": {
257
- "value": 0.0014885509690000006,
258
- "min": 0.0014885509690000006,
259
- "max": 0.0018779282020000003,
260
- "count": 22
261
  },
262
  "SoccerTwos.Policy.Beta.sum": {
263
- "value": 0.0014885509690000006,
264
- "min": 0.0014885509690000006,
265
- "max": 0.0018779282020000003,
266
- "count": 22
267
  },
268
  "SoccerTwos.Losses.CuriosityForwardLoss.mean": {
269
- "value": 0.04267202224582434,
270
- "min": 0.04210249986499548,
271
- "max": 0.1752277029057344,
272
- "count": 22
273
  },
274
  "SoccerTwos.Losses.CuriosityForwardLoss.sum": {
275
- "value": 0.04267202224582434,
276
- "min": 0.04210249986499548,
277
- "max": 0.1752277029057344,
278
- "count": 22
279
  },
280
  "SoccerTwos.Losses.CuriosityInverseLoss.mean": {
281
- "value": 1.2906148433685303,
282
- "min": 1.2906148433685303,
283
- "max": 1.5635768075784047,
284
- "count": 22
285
  },
286
  "SoccerTwos.Losses.CuriosityInverseLoss.sum": {
287
- "value": 1.2906148433685303,
288
- "min": 1.2906148433685303,
289
- "max": 1.5635768075784047,
290
- "count": 22
291
  },
292
  "SoccerTwos.Losses.RNDLoss.mean": {
293
- "value": 0.016484828665852547,
294
- "min": 0.01626814529299736,
295
- "max": 0.017836589366197586,
296
- "count": 22
297
  },
298
  "SoccerTwos.Losses.RNDLoss.sum": {
299
- "value": 0.016484828665852547,
300
- "min": 0.01626814529299736,
301
- "max": 0.017836589366197586,
302
- "count": 22
303
  }
304
  },
305
  "metadata": {
306
  "timer_format_version": "0.1.0",
307
- "start_time_seconds": "1725893434",
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": "1725906456"
316
  },
317
- "total": 13021.444170292001,
318
  "count": 1,
319
- "self": 0.14744037599302828,
320
  "children": {
321
  "run_training.setup": {
322
- "total": 0.022143291018437594,
323
  "count": 1,
324
- "self": 0.022143291018437594
325
  },
326
  "TrainerController.start_learning": {
327
- "total": 13021.27458662499,
328
  "count": 1,
329
- "self": 2.3865584495360963,
330
  "children": {
331
  "TrainerController._reset_env": {
332
- "total": 3.743250749015715,
333
- "count": 10,
334
- "self": 3.743250749015715
335
  },
336
  "TrainerController.advance": {
337
- "total": 13014.947715634422,
338
- "count": 223599,
339
- "self": 1.990279897057917,
340
  "children": {
341
  "env_step": {
342
- "total": 9167.291288381675,
343
- "count": 223599,
344
- "self": 8836.760780083598,
345
  "children": {
346
  "SubprocessEnvManager._take_step": {
347
- "total": 329.17745976604056,
348
- "count": 223599,
349
- "self": 13.062646454607602,
350
  "children": {
351
  "TorchPolicy.evaluate": {
352
- "total": 316.11481331143295,
353
- "count": 410286,
354
- "self": 316.11481331143295
355
  }
356
  }
357
  },
358
  "workers": {
359
- "total": 1.3530485320370644,
360
- "count": 223599,
361
  "self": 0.0,
362
  "children": {
363
  "worker_root": {
364
- "total": 13014.626546319574,
365
- "count": 223599,
366
  "is_parallel": true,
367
- "self": 4427.421425099019,
368
  "children": {
369
  "steps_from_proto": {
370
- "total": 0.013593417999800295,
371
- "count": 20,
372
  "is_parallel": true,
373
- "self": 0.0017637503915466368,
374
  "children": {
375
  "_process_rank_one_or_two_observation": {
376
- "total": 0.011829667608253658,
377
- "count": 80,
378
  "is_parallel": true,
379
- "self": 0.011829667608253658
380
  }
381
  }
382
  },
383
  "UnityEnvironment.step": {
384
- "total": 8587.191527802555,
385
- "count": 223599,
386
  "is_parallel": true,
387
- "self": 22.658694755868055,
388
  "children": {
389
  "UnityEnvironment._generate_step_input": {
390
- "total": 170.9808094984619,
391
- "count": 223599,
392
  "is_parallel": true,
393
- "self": 170.9808094984619
394
  },
395
  "communicator.exchange": {
396
- "total": 8116.417533892614,
397
- "count": 223599,
398
  "is_parallel": true,
399
- "self": 8116.417533892614
400
  },
401
  "steps_from_proto": {
402
- "total": 277.1344896556111,
403
- "count": 447198,
404
  "is_parallel": true,
405
- "self": 33.63658828905318,
406
  "children": {
407
  "_process_rank_one_or_two_observation": {
408
- "total": 243.49790136655793,
409
- "count": 1788792,
410
  "is_parallel": true,
411
- "self": 243.49790136655793
412
  }
413
  }
414
  }
@@ -421,31 +421,31 @@
421
  }
422
  },
423
  "trainer_advance": {
424
- "total": 3845.6661473556887,
425
- "count": 223599,
426
- "self": 19.21496674482478,
427
  "children": {
428
  "process_trajectory": {
429
- "total": 708.6676588628325,
430
- "count": 223599,
431
- "self": 707.5820961118443,
432
  "children": {
433
  "RLTrainer._checkpoint": {
434
- "total": 1.0855627509881742,
435
- "count": 7,
436
- "self": 1.0855627509881742
437
  }
438
  }
439
  },
440
  "_update_policy": {
441
- "total": 3117.7835217480315,
442
- "count": 22,
443
- "self": 316.2076580732246,
444
  "children": {
445
  "TorchPOCAOptimizer.update": {
446
- "total": 2801.575863674807,
447
- "count": 264,
448
- "self": 2801.575863674807
449
  }
450
  }
451
  }
@@ -454,19 +454,19 @@
454
  }
455
  },
456
  "trainer_threads": {
457
- "total": 4.5902561396360397e-07,
458
  "count": 1,
459
- "self": 4.5902561396360397e-07
460
  },
461
  "TrainerController._save_models": {
462
- "total": 0.1970613329904154,
463
  "count": 1,
464
- "self": 0.0020279160235077143,
465
  "children": {
466
  "RLTrainer._checkpoint": {
467
- "total": 0.19503341696690768,
468
  "count": 1,
469
- "self": 0.19503341696690768
470
  }
471
  }
472
  }
 
2
  "name": "root",
3
  "gauges": {
4
  "SoccerTwos.Policy.Entropy.mean": {
5
+ "value": 1.5399205684661865,
6
+ "min": 1.512854814529419,
7
+ "max": 1.568915843963623,
8
+ "count": 205
9
  },
10
  "SoccerTwos.Policy.Entropy.sum": {
11
+ "value": 33410.1171875,
12
+ "min": 16442.71484375,
13
+ "max": 34939.3828125,
14
+ "count": 205
15
  },
16
  "SoccerTwos.Environment.EpisodeLength.mean": {
17
+ "value": 68.73611111111111,
18
+ "min": 50.714285714285715,
19
+ "max": 97.9,
20
+ "count": 205
21
  },
22
  "SoccerTwos.Environment.EpisodeLength.sum": {
23
+ "value": 19796.0,
24
+ "min": 8296.0,
25
+ "max": 20428.0,
26
+ "count": 205
27
  },
28
  "SoccerTwos.Self-play.ELO.mean": {
29
+ "value": 272.7278779503926,
30
+ "min": 233.8647348324593,
31
+ "max": 300.1237163905635,
32
+ "count": 205
33
  },
34
  "SoccerTwos.Self-play.ELO.sum": {
35
+ "value": 39272.81442485654,
36
+ "min": 19555.710165917008,
37
+ "max": 53458.99614819349,
38
+ "count": 205
39
  },
40
  "SoccerTwos.Step.mean": {
41
+ "value": 79999942.0,
42
+ "min": 77959890.0,
43
+ "max": 79999942.0,
44
+ "count": 205
45
  },
46
  "SoccerTwos.Step.sum": {
47
+ "value": 79999942.0,
48
+ "min": 77959890.0,
49
+ "max": 79999942.0,
50
+ "count": 205
51
  },
52
  "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": {
53
+ "value": -0.13813956081867218,
54
+ "min": -0.20680981874465942,
55
+ "max": 0.12538133561611176,
56
+ "count": 205
57
  },
58
  "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": {
59
+ "value": -19.89209747314453,
60
+ "min": -29.84649658203125,
61
+ "max": 23.822452545166016,
62
+ "count": 205
63
  },
64
  "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": {
65
+ "value": -0.1384831964969635,
66
+ "min": -0.2039821445941925,
67
+ "max": 0.12654316425323486,
68
+ "count": 205
69
  },
70
  "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": {
71
+ "value": -19.941579818725586,
72
+ "min": -29.398372650146484,
73
+ "max": 24.043201446533203,
74
+ "count": 205
75
  },
76
  "SoccerTwos.Policy.CuriosityBaselineEstimate.mean": {
77
+ "value": 0.00017744496290106326,
78
+ "min": -1.280401829717448e-05,
79
+ "max": 0.00034355284878984094,
80
+ "count": 205
81
  },
82
  "SoccerTwos.Policy.CuriosityBaselineEstimate.sum": {
83
+ "value": 0.02555207535624504,
84
+ "min": -0.000870673218742013,
85
+ "max": 0.05909109115600586,
86
+ "count": 205
87
  },
88
  "SoccerTwos.Policy.CuriosityValueEstimate.mean": {
89
+ "value": 0.00018491833179723471,
90
+ "min": -8.266243639809545e-06,
91
+ "max": 0.00034810250508598983,
92
+ "count": 205
93
  },
94
  "SoccerTwos.Policy.CuriosityValueEstimate.sum": {
95
+ "value": 0.026628239080309868,
96
+ "min": -0.0005621045711450279,
97
+ "max": 0.05987362936139107,
98
+ "count": 205
99
  },
100
  "SoccerTwos.Policy.RndBaselineEstimate.mean": {
101
+ "value": 1.5067198546603322e-05,
102
+ "min": -0.0004950298462063074,
103
+ "max": 9.06805507838726e-05,
104
+ "count": 205
105
  },
106
  "SoccerTwos.Policy.RndBaselineEstimate.sum": {
107
+ "value": 0.0021696765907108784,
108
+ "min": -0.03429080545902252,
109
+ "max": 0.015687735751271248,
110
+ "count": 205
111
  },
112
  "SoccerTwos.Policy.RndValueEstimate.mean": {
113
+ "value": 1.4539382391376421e-05,
114
+ "min": -0.0005132480873726308,
115
+ "max": 8.886260911822319e-05,
116
+ "count": 205
117
  },
118
  "SoccerTwos.Policy.RndValueEstimate.sum": {
119
+ "value": 0.0020936711225658655,
120
+ "min": -0.03490087017416954,
121
+ "max": 0.0153732318431139,
122
+ "count": 205
123
  },
124
  "SoccerTwos.Environment.CumulativeReward.mean": {
125
  "value": 0.0,
126
  "min": 0.0,
127
  "max": 0.0,
128
+ "count": 205
129
  },
130
  "SoccerTwos.Environment.CumulativeReward.sum": {
131
  "value": 0.0,
132
  "min": 0.0,
133
  "max": 0.0,
134
+ "count": 205
135
  },
136
  "SoccerTwos.Policy.ExtrinsicReward.mean": {
137
+ "value": -0.11527777794334623,
138
+ "min": -0.3677196840601643,
139
+ "max": 0.27121262989546124,
140
+ "count": 205
141
  },
142
  "SoccerTwos.Policy.ExtrinsicReward.sum": {
143
+ "value": -16.600000023841858,
144
+ "min": -60.78120017051697,
145
+ "max": 51.530399680137634,
146
+ "count": 205
147
  },
148
  "SoccerTwos.Policy.CuriosityReward.mean": {
149
+ "value": 3.116804232582303e-06,
150
  "min": 0.0,
151
+ "max": 5.3372595253514096e-06,
152
+ "count": 205
153
  },
154
  "SoccerTwos.Policy.CuriosityReward.sum": {
155
+ "value": 0.00044881980949185163,
156
  "min": 0.0,
157
+ "max": 0.0007159653786175113,
158
+ "count": 205
159
  },
160
  "SoccerTwos.Policy.RndReward.mean": {
161
+ "value": 1.269147985717546e-06,
162
+ "min": 8.895423654938608e-07,
163
+ "max": 1.867236133108463e-06,
164
+ "count": 205
165
  },
166
  "SoccerTwos.Policy.RndReward.sum": {
167
+ "value": 0.00018275730994332662,
168
+ "min": 7.041936916607483e-05,
169
+ "max": 0.00019470762803308617,
170
+ "count": 205
171
  },
172
  "SoccerTwos.Environment.GroupCumulativeReward.mean": {
173
+ "value": -0.11527777794334623,
174
+ "min": -0.3677196840601643,
175
+ "max": 0.27121262989546124,
176
+ "count": 205
177
  },
178
  "SoccerTwos.Environment.GroupCumulativeReward.sum": {
179
+ "value": -16.600000023841858,
180
+ "min": -60.78120017051697,
181
+ "max": 51.530399680137634,
182
+ "count": 205
183
  },
184
  "SoccerTwos.IsTraining.mean": {
185
  "value": 1.0,
186
  "min": 1.0,
187
  "max": 1.0,
188
+ "count": 205
189
  },
190
  "SoccerTwos.IsTraining.sum": {
191
  "value": 1.0,
192
  "min": 1.0,
193
  "max": 1.0,
194
+ "count": 205
195
  },
196
  "SoccerTwos.Losses.PolicyLoss.mean": {
197
+ "value": 0.012351180444966303,
198
+ "min": 0.0067968835305691755,
199
+ "max": 0.015325144719099626,
200
+ "count": 13
201
  },
202
  "SoccerTwos.Losses.PolicyLoss.sum": {
203
+ "value": 0.012351180444966303,
204
+ "min": 0.0067968835305691755,
205
+ "max": 0.015325144719099626,
206
+ "count": 13
207
  },
208
  "SoccerTwos.Losses.ValueLoss.mean": {
209
+ "value": 0.0175623488612473,
210
+ "min": 0.016617532974729936,
211
+ "max": 0.0175623488612473,
212
+ "count": 13
213
  },
214
  "SoccerTwos.Losses.ValueLoss.sum": {
215
+ "value": 0.0175623488612473,
216
+ "min": 0.016617532974729936,
217
+ "max": 0.0175623488612473,
218
+ "count": 13
219
  },
220
  "SoccerTwos.Losses.BaselineLoss.mean": {
221
+ "value": 0.018047905527055264,
222
+ "min": 0.017212544723103445,
223
+ "max": 0.018047905527055264,
224
+ "count": 13
225
  },
226
  "SoccerTwos.Losses.BaselineLoss.sum": {
227
+ "value": 0.018047905527055264,
228
+ "min": 0.017212544723103445,
229
+ "max": 0.018047905527055264,
230
+ "count": 13
231
  },
232
  "SoccerTwos.Policy.LearningRate.mean": {
233
  "value": 0.0003,
234
  "min": 0.0003,
235
  "max": 0.0003,
236
+ "count": 13
237
  },
238
  "SoccerTwos.Policy.LearningRate.sum": {
239
  "value": 0.0003,
240
  "min": 0.0003,
241
  "max": 0.0003,
242
+ "count": 13
243
  },
244
  "SoccerTwos.Policy.Epsilon.mean": {
245
  "value": 0.20000000000000004,
246
  "min": 0.20000000000000004,
247
  "max": 0.20000000000000004,
248
+ "count": 13
249
  },
250
  "SoccerTwos.Policy.Epsilon.sum": {
251
  "value": 0.20000000000000004,
252
  "min": 0.20000000000000004,
253
  "max": 0.20000000000000004,
254
+ "count": 13
255
  },
256
  "SoccerTwos.Policy.Beta.mean": {
257
+ "value": 2.5267467249999682e-05,
258
+ "min": 2.5267467249999682e-05,
259
+ "max": 0.0002468431697500002,
260
+ "count": 13
261
  },
262
  "SoccerTwos.Policy.Beta.sum": {
263
+ "value": 2.5267467249999682e-05,
264
+ "min": 2.5267467249999682e-05,
265
+ "max": 0.0002468431697500002,
266
+ "count": 13
267
  },
268
  "SoccerTwos.Losses.CuriosityForwardLoss.mean": {
269
+ "value": 0.04501075980563959,
270
+ "min": 0.044811712888379894,
271
+ "max": 0.12417677200088899,
272
+ "count": 13
273
  },
274
  "SoccerTwos.Losses.CuriosityForwardLoss.sum": {
275
+ "value": 0.04501075980563959,
276
+ "min": 0.044811712888379894,
277
+ "max": 0.12417677200088899,
278
+ "count": 13
279
  },
280
  "SoccerTwos.Losses.CuriosityInverseLoss.mean": {
281
+ "value": 1.1322325964768727,
282
+ "min": 1.1322325964768727,
283
+ "max": 1.3759111762046814,
284
+ "count": 13
285
  },
286
  "SoccerTwos.Losses.CuriosityInverseLoss.sum": {
287
+ "value": 1.1322325964768727,
288
+ "min": 1.1322325964768727,
289
+ "max": 1.3759111762046814,
290
+ "count": 13
291
  },
292
  "SoccerTwos.Losses.RNDLoss.mean": {
293
+ "value": 0.016439350321888924,
294
+ "min": 0.0163416750729084,
295
+ "max": 0.01815606839954853,
296
+ "count": 13
297
  },
298
  "SoccerTwos.Losses.RNDLoss.sum": {
299
+ "value": 0.016439350321888924,
300
+ "min": 0.0163416750729084,
301
+ "max": 0.01815606839954853,
302
+ "count": 13
303
  }
304
  },
305
  "metadata": {
306
  "timer_format_version": "0.1.0",
307
+ "start_time_seconds": "1725978587",
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": "1725986627"
316
  },
317
+ "total": 8040.06352700002,
318
  "count": 1,
319
+ "self": 0.23776454094331712,
320
  "children": {
321
  "run_training.setup": {
322
+ "total": 0.019753542030230165,
323
  "count": 1,
324
+ "self": 0.019753542030230165
325
  },
326
  "TrainerController.start_learning": {
327
+ "total": 8039.806008917047,
328
  "count": 1,
329
+ "self": 1.314517370716203,
330
  "children": {
331
  "TrainerController._reset_env": {
332
+ "total": 3.6344570019864477,
333
+ "count": 7,
334
+ "self": 3.6344570019864477
335
  },
336
  "TrainerController.advance": {
337
+ "total": 8034.724037919368,
338
+ "count": 140117,
339
+ "self": 1.1400749064050615,
340
  "children": {
341
  "env_step": {
342
+ "total": 5695.069258543197,
343
+ "count": 140117,
344
+ "self": 5497.789965885109,
345
  "children": {
346
  "SubprocessEnvManager._take_step": {
347
+ "total": 196.4914063619217,
348
+ "count": 140117,
349
+ "self": 8.097888264048379,
350
  "children": {
351
  "TorchPolicy.evaluate": {
352
+ "total": 188.3935180978733,
353
+ "count": 256298,
354
+ "self": 188.3935180978733
355
  }
356
  }
357
  },
358
  "workers": {
359
+ "total": 0.7878862961661071,
360
+ "count": 140117,
361
  "self": 0.0,
362
  "children": {
363
  "worker_root": {
364
+ "total": 8034.413471959007,
365
+ "count": 140117,
366
  "is_parallel": true,
367
+ "self": 2685.770048620354,
368
  "children": {
369
  "steps_from_proto": {
370
+ "total": 0.009649291925597936,
371
+ "count": 14,
372
  "is_parallel": true,
373
+ "self": 0.0013223288697190583,
374
  "children": {
375
  "_process_rank_one_or_two_observation": {
376
+ "total": 0.008326963055878878,
377
+ "count": 56,
378
  "is_parallel": true,
379
+ "self": 0.008326963055878878
380
  }
381
  }
382
  },
383
  "UnityEnvironment.step": {
384
+ "total": 5348.633774046728,
385
+ "count": 140117,
386
  "is_parallel": true,
387
+ "self": 13.765971858112607,
388
  "children": {
389
  "UnityEnvironment._generate_step_input": {
390
+ "total": 104.13079777162056,
391
+ "count": 140117,
392
  "is_parallel": true,
393
+ "self": 104.13079777162056
394
  },
395
  "communicator.exchange": {
396
+ "total": 5057.198739988089,
397
+ "count": 140117,
398
  "is_parallel": true,
399
+ "self": 5057.198739988089
400
  },
401
  "steps_from_proto": {
402
+ "total": 173.53826442890568,
403
+ "count": 280234,
404
  "is_parallel": true,
405
+ "self": 20.749318457383197,
406
  "children": {
407
  "_process_rank_one_or_two_observation": {
408
+ "total": 152.78894597152248,
409
+ "count": 1120936,
410
  "is_parallel": true,
411
+ "self": 152.78894597152248
412
  }
413
  }
414
  }
 
421
  }
422
  },
423
  "trainer_advance": {
424
+ "total": 2338.514704469766,
425
+ "count": 140117,
426
+ "self": 10.135464723221958,
427
  "children": {
428
  "process_trajectory": {
429
+ "total": 436.0033031643834,
430
+ "count": 140117,
431
+ "self": 435.12472358031664,
432
  "children": {
433
  "RLTrainer._checkpoint": {
434
+ "total": 0.8785795840667561,
435
+ "count": 5,
436
+ "self": 0.8785795840667561
437
  }
438
  }
439
  },
440
  "_update_policy": {
441
+ "total": 1892.3759365821606,
442
+ "count": 13,
443
+ "self": 187.99826970574213,
444
  "children": {
445
  "TorchPOCAOptimizer.update": {
446
+ "total": 1704.3776668764185,
447
+ "count": 156,
448
+ "self": 1704.3776668764185
449
  }
450
  }
451
  }
 
454
  }
455
  },
456
  "trainer_threads": {
457
+ "total": 4.159519448876381e-07,
458
  "count": 1,
459
+ "self": 4.159519448876381e-07
460
  },
461
  "TrainerController._save_models": {
462
+ "total": 0.132996209023986,
463
  "count": 1,
464
+ "self": 0.004118792014196515,
465
  "children": {
466
  "RLTrainer._checkpoint": {
467
+ "total": 0.1288774170097895,
468
  "count": 1,
469
+ "self": 0.1288774170097895
470
  }
471
  }
472
  }
run_logs/training_status.json CHANGED
@@ -1,24 +1,24 @@
1
  {
2
  "SoccerTwos": {
3
- "elo": 262.64752300771465,
4
  "checkpoints": [
5
  {
6
- "steps": 68168382,
7
- "file_path": "results/sept-run-01/SoccerTwos/SoccerTwos-68168382.onnx",
8
  "reward": 0.0,
9
- "creation_time": 1725906456.0079901,
10
  "auxillary_file_paths": [
11
- "results/sept-run-01/SoccerTwos/SoccerTwos-68168382.pt"
12
  ]
13
  }
14
  ],
15
  "final_checkpoint": {
16
- "steps": 68168382,
17
  "file_path": "results/sept-run-01/SoccerTwos.onnx",
18
  "reward": 0.0,
19
- "creation_time": 1725906456.0079901,
20
  "auxillary_file_paths": [
21
- "results/sept-run-01/SoccerTwos/SoccerTwos-68168382.pt"
22
  ]
23
  }
24
  },
 
1
  {
2
  "SoccerTwos": {
3
+ "elo": 271.2064667443272,
4
  "checkpoints": [
5
  {
6
+ "steps": 80000070,
7
+ "file_path": "results/sept-run-01/SoccerTwos/SoccerTwos-80000070.onnx",
8
  "reward": 0.0,
9
+ "creation_time": 1725986627.517882,
10
  "auxillary_file_paths": [
11
+ "results/sept-run-01/SoccerTwos/SoccerTwos-80000070.pt"
12
  ]
13
  }
14
  ],
15
  "final_checkpoint": {
16
+ "steps": 80000070,
17
  "file_path": "results/sept-run-01/SoccerTwos.onnx",
18
  "reward": 0.0,
19
+ "creation_time": 1725986627.517882,
20
  "auxillary_file_paths": [
21
+ "results/sept-run-01/SoccerTwos/SoccerTwos-80000070.pt"
22
  ]
23
  }
24
  },