Hugging Face's logo

ccnets
/
causal-gpt-rl-unity

Reinforcement Learning
PyTorch
ONNX
Safetensors
causal-gpt-rl
unity-ml-agents
Model card Files
xet
 Community
kissin42 commited on
Commit
b8eaaea
·
verified ·
1 Parent(s): 35a9ea2

Publish Crawler 60k continuous policy

Browse files

Add the first continuous-action Unity policy: Crawler 60k safetensors and fixed-batch-10 ctx16/ctx32/ctx64 ONNX files. Over seeds 100..103, ctx32 averaged 1721.88 return and is the recommended deployment variant.

Files changed (7) hide show
  1. README.md +30 -1
  2. config.json +2 -1
  3. crawler/config.json +813 -0
  4. crawler/crawler-b10-ctx16.onnx +3 -0
  5. crawler/crawler-b10-ctx32.onnx +3 -0
  6. crawler/crawler-b10-ctx64.onnx +3 -0
  7. crawler/model.safetensors +3 -0
README.md CHANGED
@@ -81,7 +81,7 @@ from causal_gpt_rl.inference import load_runner_from_hub
81
 
82
  runner = load_runner_from_hub(
83
  repo_id="ccnets/causal-gpt-rl-unity",
84
- subfolder="pyramids", # or "dungeon-escape" / "soccer-twos"
85
  )
86
  ```
87
 
@@ -121,12 +121,14 @@ Notebook versions:
121
  | `dungeon-escape` | 32 | 36 | 60.19% agent / 94.44% group success |
122
  | `soccer-twos` | 32 | 16 | 40.00% win rate vs stock |
123
  | `pyramids` | 32 | 16 | 1.7503 mean return (`ctx16`) |
 
124
 
125
  | Bundle | Observation (ego) | Action (ego) |
126
  |---|---|---|
127
  | `dungeon-escape` | `Tuple(Box(10), Box(360), Box(1))` | `Discrete(7)` |
128
  | `soccer-twos` | `Tuple(Box(264), Box(72))` | `MultiDiscrete([3,3,3])` |
129
  | `pyramids` | `Tuple(Box(56), Box(56), Box(56), Box(4))` | `Discrete(5)` |
 
130
 
131
  The result column is the **ONNX at context length 32**, the recommended Unity
132
  deployment artifact: a self-contained, fixed-batch rolling-window graph derived
@@ -254,6 +256,27 @@ All three graphs use fixed batch 16. The context dimension is 16, 32, or 64:
254
  `action [16, 5]`. Select the discrete action with `argmax` and retain one
255
  independent rolling context per agent.
256
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
257
  </details>
258
 
259
  ### Reproduction runtime
@@ -321,6 +344,12 @@ ccnets/causal-gpt-rl-unity/
321
  pyramids-b16-ctx16.onnx
322
  pyramids-b16-ctx32.onnx
323
  pyramids-b16-ctx64.onnx
 
 
 
 
 
 
324
  LICENSE
325
  ```
326
 
 
81
 
82
  runner = load_runner_from_hub(
83
  repo_id="ccnets/causal-gpt-rl-unity",
84
+ subfolder="crawler", # or "pyramids" / "dungeon-escape" / "soccer-twos"
85
  )
86
  ```
87
 
 
121
  | `dungeon-escape` | 32 | 36 | 60.19% agent / 94.44% group success |
122
  | `soccer-twos` | 32 | 16 | 40.00% win rate vs stock |
123
  | `pyramids` | 32 | 16 | 1.7503 mean return (`ctx16`) |
124
+ | `crawler` | 32 | 10 | 1721.88 mean return (ONNX `ctx32`) |
125
 
126
  | Bundle | Observation (ego) | Action (ego) |
127
  |---|---|---|
128
  | `dungeon-escape` | `Tuple(Box(10), Box(360), Box(1))` | `Discrete(7)` |
129
  | `soccer-twos` | `Tuple(Box(264), Box(72))` | `MultiDiscrete([3,3,3])` |
130
  | `pyramids` | `Tuple(Box(56), Box(56), Box(56), Box(4))` | `Discrete(5)` |
131
+ | `crawler` | `Tuple(Box(126), Box(32))` | `Box(20, [-1, 1])` |
132
 
133
  The result column is the **ONNX at context length 32**, the recommended Unity
134
  deployment artifact: a self-contained, fixed-batch rolling-window graph derived
 
256
  `action [16, 5]`. Select the discrete action with `argmax` and retain one
257
  independent rolling context per agent.
258
 
259
+ **Crawler** — the repository's first continuous-action Causal GPT-RL
260
+ policy. This is the 60k checkpoint. Closed-loop ONNX validation used the matching
261
+ ML-Agents release-23 model-removed build, seeds `100..103`, 10 agents per launch,
262
+ BOS discard, and 40 first episodes per context variant.
263
+
264
+ | Length | Runtime | Mean return | Std | HF simple mean |
265
+ |---:|---|---:|---:|---:|
266
+ | 16 | ONNX (windowed) | 1135.54 | 1096.65 | 70.73% |
267
+ | **32** | **ONNX (windowed)** | **1721.88** | **1213.81** | **107.25%** |
268
+ | 64 | ONNX (windowed) | 1469.44 | 1178.60 | 91.53% |
269
+
270
+ Context 32 is the recommended/default ONNX variant. It has the highest aggregate
271
+ return and exceeds the published `unity/crawler/simple-v0` dataset mean of
272
+ 1605.4283 by 7.25%. Dataset landmarks are simple 1605.4283, medium 2141.5175,
273
+ and expert 2575.8405.
274
+
275
+ The graphs use fixed batch 10. Inputs are `states [10, context, 158]`,
276
+ `actions [10, context, 20]`, `is_bos [10, context, 1]`, and
277
+ `mask [10, context]`; output is `action [10, 20]` in the environment's
278
+ `[-1, 1]` action range.
279
+
280
  </details>
281
 
282
  ### Reproduction runtime
 
344
  pyramids-b16-ctx16.onnx
345
  pyramids-b16-ctx32.onnx
346
  pyramids-b16-ctx64.onnx
347
+ crawler/
348
+ config.json
349
+ model.safetensors
350
+ crawler-b10-ctx16.onnx
351
+ crawler-b10-ctx32.onnx
352
+ crawler-b10-ctx64.onnx
353
  LICENSE
354
  ```
355
 
config.json CHANGED
@@ -3,6 +3,7 @@
3
  "bundles": [
4
  "dungeon-escape",
5
  "soccer-twos",
6
- "pyramids"
 
7
  ]
8
  }
 
3
  "bundles": [
4
  "dungeon-escape",
5
  "soccer-twos",
6
+ "pyramids",
7
+ "crawler"
8
  ]
9
  }
crawler/config.json ADDED
@@ -0,0 +1,813 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bundle_format_version": 2,
3
+ "package_version": "0.15.0",
4
+ "model_config": {
5
+ "network_name": "Llama",
6
+ "num_layers": 4,
7
+ "d_model": 256,
8
+ "num_heads": 8,
9
+ "dropout": 0.05,
10
+ "use_eos": true,
11
+ "use_bos_action_gate": true,
12
+ "rope_theta": 1000.0,
13
+ "intermediate_size": 1024,
14
+ "max_position_embeddings": 256,
15
+ "context_length": 32
16
+ },
17
+ "state_specs": [
18
+ {
19
+ "type": "continuous",
20
+ "size": 126,
21
+ "dtype": "float32",
22
+ "low": [
23
+ null,
24
+ null,
25
+ null,
26
+ null,
27
+ null,
28
+ null,
29
+ null,
30
+ null,
31
+ null,
32
+ null,
33
+ null,
34
+ null,
35
+ null,
36
+ null,
37
+ null,
38
+ null,
39
+ null,
40
+ null,
41
+ null,
42
+ null,
43
+ null,
44
+ null,
45
+ null,
46
+ null,
47
+ null,
48
+ null,
49
+ null,
50
+ null,
51
+ null,
52
+ null,
53
+ null,
54
+ null,
55
+ null,
56
+ null,
57
+ null,
58
+ null,
59
+ null,
60
+ null,
61
+ null,
62
+ null,
63
+ null,
64
+ null,
65
+ null,
66
+ null,
67
+ null,
68
+ null,
69
+ null,
70
+ null,
71
+ null,
72
+ null,
73
+ null,
74
+ null,
75
+ null,
76
+ null,
77
+ null,
78
+ null,
79
+ null,
80
+ null,
81
+ null,
82
+ null,
83
+ null,
84
+ null,
85
+ null,
86
+ null,
87
+ null,
88
+ null,
89
+ null,
90
+ null,
91
+ null,
92
+ null,
93
+ null,
94
+ null,
95
+ null,
96
+ null,
97
+ null,
98
+ null,
99
+ null,
100
+ null,
101
+ null,
102
+ null,
103
+ null,
104
+ null,
105
+ null,
106
+ null,
107
+ null,
108
+ null,
109
+ null,
110
+ null,
111
+ null,
112
+ null,
113
+ null,
114
+ null,
115
+ null,
116
+ null,
117
+ null,
118
+ null,
119
+ null,
120
+ null,
121
+ null,
122
+ null,
123
+ null,
124
+ null,
125
+ null,
126
+ null,
127
+ null,
128
+ null,
129
+ null,
130
+ null,
131
+ null,
132
+ null,
133
+ null,
134
+ null,
135
+ null,
136
+ null,
137
+ null,
138
+ null,
139
+ null,
140
+ null,
141
+ null,
142
+ null,
143
+ null,
144
+ null,
145
+ null,
146
+ null,
147
+ null,
148
+ null
149
+ ],
150
+ "high": [
151
+ null,
152
+ null,
153
+ null,
154
+ null,
155
+ null,
156
+ null,
157
+ null,
158
+ null,
159
+ null,
160
+ null,
161
+ null,
162
+ null,
163
+ null,
164
+ null,
165
+ null,
166
+ null,
167
+ null,
168
+ null,
169
+ null,
170
+ null,
171
+ null,
172
+ null,
173
+ null,
174
+ null,
175
+ null,
176
+ null,
177
+ null,
178
+ null,
179
+ null,
180
+ null,
181
+ null,
182
+ null,
183
+ null,
184
+ null,
185
+ null,
186
+ null,
187
+ null,
188
+ null,
189
+ null,
190
+ null,
191
+ null,
192
+ null,
193
+ null,
194
+ null,
195
+ null,
196
+ null,
197
+ null,
198
+ null,
199
+ null,
200
+ null,
201
+ null,
202
+ null,
203
+ null,
204
+ null,
205
+ null,
206
+ null,
207
+ null,
208
+ null,
209
+ null,
210
+ null,
211
+ null,
212
+ null,
213
+ null,
214
+ null,
215
+ null,
216
+ null,
217
+ null,
218
+ null,
219
+ null,
220
+ null,
221
+ null,
222
+ null,
223
+ null,
224
+ null,
225
+ null,
226
+ null,
227
+ null,
228
+ null,
229
+ null,
230
+ null,
231
+ null,
232
+ null,
233
+ null,
234
+ null,
235
+ null,
236
+ null,
237
+ null,
238
+ null,
239
+ null,
240
+ null,
241
+ null,
242
+ null,
243
+ null,
244
+ null,
245
+ null,
246
+ null,
247
+ null,
248
+ null,
249
+ null,
250
+ null,
251
+ null,
252
+ null,
253
+ null,
254
+ null,
255
+ null,
256
+ null,
257
+ null,
258
+ null,
259
+ null,
260
+ null,
261
+ null,
262
+ null,
263
+ null,
264
+ null,
265
+ null,
266
+ null,
267
+ null,
268
+ null,
269
+ null,
270
+ null,
271
+ null,
272
+ null,
273
+ null,
274
+ null,
275
+ null,
276
+ null
277
+ ],
278
+ "squash": null
279
+ },
280
+ {
281
+ "type": "continuous",
282
+ "size": 32,
283
+ "dtype": "float32",
284
+ "low": [
285
+ null,
286
+ null,
287
+ null,
288
+ null,
289
+ null,
290
+ null,
291
+ null,
292
+ null,
293
+ null,
294
+ null,
295
+ null,
296
+ null,
297
+ null,
298
+ null,
299
+ null,
300
+ null,
301
+ null,
302
+ null,
303
+ null,
304
+ null,
305
+ null,
306
+ null,
307
+ null,
308
+ null,
309
+ null,
310
+ null,
311
+ null,
312
+ null,
313
+ null,
314
+ null,
315
+ null,
316
+ null
317
+ ],
318
+ "high": [
319
+ null,
320
+ null,
321
+ null,
322
+ null,
323
+ null,
324
+ null,
325
+ null,
326
+ null,
327
+ null,
328
+ null,
329
+ null,
330
+ null,
331
+ null,
332
+ null,
333
+ null,
334
+ null,
335
+ null,
336
+ null,
337
+ null,
338
+ null,
339
+ null,
340
+ null,
341
+ null,
342
+ null,
343
+ null,
344
+ null,
345
+ null,
346
+ null,
347
+ null,
348
+ null,
349
+ null,
350
+ null
351
+ ],
352
+ "squash": null
353
+ }
354
+ ],
355
+ "action_specs": [
356
+ {
357
+ "type": "continuous",
358
+ "size": 20,
359
+ "dtype": "float32",
360
+ "low": [
361
+ -1.0,
362
+ -1.0,
363
+ -1.0,
364
+ -1.0,
365
+ -1.0,
366
+ -1.0,
367
+ -1.0,
368
+ -1.0,
369
+ -1.0,
370
+ -1.0,
371
+ -1.0,
372
+ -1.0,
373
+ -1.0,
374
+ -1.0,
375
+ -1.0,
376
+ -1.0,
377
+ -1.0,
378
+ -1.0,
379
+ -1.0,
380
+ -1.0
381
+ ],
382
+ "high": [
383
+ 1.0,
384
+ 1.0,
385
+ 1.0,
386
+ 1.0,
387
+ 1.0,
388
+ 1.0,
389
+ 1.0,
390
+ 1.0,
391
+ 1.0,
392
+ 1.0,
393
+ 1.0,
394
+ 1.0,
395
+ 1.0,
396
+ 1.0,
397
+ 1.0,
398
+ 1.0,
399
+ 1.0,
400
+ 1.0,
401
+ 1.0,
402
+ 1.0
403
+ ],
404
+ "squash": "tanh"
405
+ }
406
+ ],
407
+ "context_length": 32,
408
+ "requires_capabilities": [
409
+ "hybrid_state"
410
+ ],
411
+ "state_container": {
412
+ "type": "Tuple",
413
+ "spaces": [
414
+ {
415
+ "type": "Box",
416
+ "shape": [
417
+ 126
418
+ ],
419
+ "dtype": "float32",
420
+ "low": [
421
+ "-inf",
422
+ "-inf",
423
+ "-inf",
424
+ "-inf",
425
+ "-inf",
426
+ "-inf",
427
+ "-inf",
428
+ "-inf",
429
+ "-inf",
430
+ "-inf",
431
+ "-inf",
432
+ "-inf",
433
+ "-inf",
434
+ "-inf",
435
+ "-inf",
436
+ "-inf",
437
+ "-inf",
438
+ "-inf",
439
+ "-inf",
440
+ "-inf",
441
+ "-inf",
442
+ "-inf",
443
+ "-inf",
444
+ "-inf",
445
+ "-inf",
446
+ "-inf",
447
+ "-inf",
448
+ "-inf",
449
+ "-inf",
450
+ "-inf",
451
+ "-inf",
452
+ "-inf",
453
+ "-inf",
454
+ "-inf",
455
+ "-inf",
456
+ "-inf",
457
+ "-inf",
458
+ "-inf",
459
+ "-inf",
460
+ "-inf",
461
+ "-inf",
462
+ "-inf",
463
+ "-inf",
464
+ "-inf",
465
+ "-inf",
466
+ "-inf",
467
+ "-inf",
468
+ "-inf",
469
+ "-inf",
470
+ "-inf",
471
+ "-inf",
472
+ "-inf",
473
+ "-inf",
474
+ "-inf",
475
+ "-inf",
476
+ "-inf",
477
+ "-inf",
478
+ "-inf",
479
+ "-inf",
480
+ "-inf",
481
+ "-inf",
482
+ "-inf",
483
+ "-inf",
484
+ "-inf",
485
+ "-inf",
486
+ "-inf",
487
+ "-inf",
488
+ "-inf",
489
+ "-inf",
490
+ "-inf",
491
+ "-inf",
492
+ "-inf",
493
+ "-inf",
494
+ "-inf",
495
+ "-inf",
496
+ "-inf",
497
+ "-inf",
498
+ "-inf",
499
+ "-inf",
500
+ "-inf",
501
+ "-inf",
502
+ "-inf",
503
+ "-inf",
504
+ "-inf",
505
+ "-inf",
506
+ "-inf",
507
+ "-inf",
508
+ "-inf",
509
+ "-inf",
510
+ "-inf",
511
+ "-inf",
512
+ "-inf",
513
+ "-inf",
514
+ "-inf",
515
+ "-inf",
516
+ "-inf",
517
+ "-inf",
518
+ "-inf",
519
+ "-inf",
520
+ "-inf",
521
+ "-inf",
522
+ "-inf",
523
+ "-inf",
524
+ "-inf",
525
+ "-inf",
526
+ "-inf",
527
+ "-inf",
528
+ "-inf",
529
+ "-inf",
530
+ "-inf",
531
+ "-inf",
532
+ "-inf",
533
+ "-inf",
534
+ "-inf",
535
+ "-inf",
536
+ "-inf",
537
+ "-inf",
538
+ "-inf",
539
+ "-inf",
540
+ "-inf",
541
+ "-inf",
542
+ "-inf",
543
+ "-inf",
544
+ "-inf",
545
+ "-inf",
546
+ "-inf"
547
+ ],
548
+ "high": [
549
+ "inf",
550
+ "inf",
551
+ "inf",
552
+ "inf",
553
+ "inf",
554
+ "inf",
555
+ "inf",
556
+ "inf",
557
+ "inf",
558
+ "inf",
559
+ "inf",
560
+ "inf",
561
+ "inf",
562
+ "inf",
563
+ "inf",
564
+ "inf",
565
+ "inf",
566
+ "inf",
567
+ "inf",
568
+ "inf",
569
+ "inf",
570
+ "inf",
571
+ "inf",
572
+ "inf",
573
+ "inf",
574
+ "inf",
575
+ "inf",
576
+ "inf",
577
+ "inf",
578
+ "inf",
579
+ "inf",
580
+ "inf",
581
+ "inf",
582
+ "inf",
583
+ "inf",
584
+ "inf",
585
+ "inf",
586
+ "inf",
587
+ "inf",
588
+ "inf",
589
+ "inf",
590
+ "inf",
591
+ "inf",
592
+ "inf",
593
+ "inf",
594
+ "inf",
595
+ "inf",
596
+ "inf",
597
+ "inf",
598
+ "inf",
599
+ "inf",
600
+ "inf",
601
+ "inf",
602
+ "inf",
603
+ "inf",
604
+ "inf",
605
+ "inf",
606
+ "inf",
607
+ "inf",
608
+ "inf",
609
+ "inf",
610
+ "inf",
611
+ "inf",
612
+ "inf",
613
+ "inf",
614
+ "inf",
615
+ "inf",
616
+ "inf",
617
+ "inf",
618
+ "inf",
619
+ "inf",
620
+ "inf",
621
+ "inf",
622
+ "inf",
623
+ "inf",
624
+ "inf",
625
+ "inf",
626
+ "inf",
627
+ "inf",
628
+ "inf",
629
+ "inf",
630
+ "inf",
631
+ "inf",
632
+ "inf",
633
+ "inf",
634
+ "inf",
635
+ "inf",
636
+ "inf",
637
+ "inf",
638
+ "inf",
639
+ "inf",
640
+ "inf",
641
+ "inf",
642
+ "inf",
643
+ "inf",
644
+ "inf",
645
+ "inf",
646
+ "inf",
647
+ "inf",
648
+ "inf",
649
+ "inf",
650
+ "inf",
651
+ "inf",
652
+ "inf",
653
+ "inf",
654
+ "inf",
655
+ "inf",
656
+ "inf",
657
+ "inf",
658
+ "inf",
659
+ "inf",
660
+ "inf",
661
+ "inf",
662
+ "inf",
663
+ "inf",
664
+ "inf",
665
+ "inf",
666
+ "inf",
667
+ "inf",
668
+ "inf",
669
+ "inf",
670
+ "inf",
671
+ "inf",
672
+ "inf",
673
+ "inf",
674
+ "inf"
675
+ ]
676
+ },
677
+ {
678
+ "type": "Box",
679
+ "shape": [
680
+ 32
681
+ ],
682
+ "dtype": "float32",
683
+ "low": [
684
+ "-inf",
685
+ "-inf",
686
+ "-inf",
687
+ "-inf",
688
+ "-inf",
689
+ "-inf",
690
+ "-inf",
691
+ "-inf",
692
+ "-inf",
693
+ "-inf",
694
+ "-inf",
695
+ "-inf",
696
+ "-inf",
697
+ "-inf",
698
+ "-inf",
699
+ "-inf",
700
+ "-inf",
701
+ "-inf",
702
+ "-inf",
703
+ "-inf",
704
+ "-inf",
705
+ "-inf",
706
+ "-inf",
707
+ "-inf",
708
+ "-inf",
709
+ "-inf",
710
+ "-inf",
711
+ "-inf",
712
+ "-inf",
713
+ "-inf",
714
+ "-inf",
715
+ "-inf"
716
+ ],
717
+ "high": [
718
+ "inf",
719
+ "inf",
720
+ "inf",
721
+ "inf",
722
+ "inf",
723
+ "inf",
724
+ "inf",
725
+ "inf",
726
+ "inf",
727
+ "inf",
728
+ "inf",
729
+ "inf",
730
+ "inf",
731
+ "inf",
732
+ "inf",
733
+ "inf",
734
+ "inf",
735
+ "inf",
736
+ "inf",
737
+ "inf",
738
+ "inf",
739
+ "inf",
740
+ "inf",
741
+ "inf",
742
+ "inf",
743
+ "inf",
744
+ "inf",
745
+ "inf",
746
+ "inf",
747
+ "inf",
748
+ "inf",
749
+ "inf"
750
+ ]
751
+ }
752
+ ]
753
+ },
754
+ "action_container": {
755
+ "type": "Box",
756
+ "shape": [
757
+ 20
758
+ ],
759
+ "dtype": "float32",
760
+ "low": [
761
+ -1.0,
762
+ -1.0,
763
+ -1.0,
764
+ -1.0,
765
+ -1.0,
766
+ -1.0,
767
+ -1.0,
768
+ -1.0,
769
+ -1.0,
770
+ -1.0,
771
+ -1.0,
772
+ -1.0,
773
+ -1.0,
774
+ -1.0,
775
+ -1.0,
776
+ -1.0,
777
+ -1.0,
778
+ -1.0,
779
+ -1.0,
780
+ -1.0
781
+ ],
782
+ "high": [
783
+ 1.0,
784
+ 1.0,
785
+ 1.0,
786
+ 1.0,
787
+ 1.0,
788
+ 1.0,
789
+ 1.0,
790
+ 1.0,
791
+ 1.0,
792
+ 1.0,
793
+ 1.0,
794
+ 1.0,
795
+ 1.0,
796
+ 1.0,
797
+ 1.0,
798
+ 1.0,
799
+ 1.0,
800
+ 1.0,
801
+ 1.0,
802
+ 1.0
803
+ ]
804
+ },
805
+ "state_normalization": {
806
+ "embedded": true,
807
+ "legacy_sidecar": false
808
+ },
809
+ "env_id": "Crawler",
810
+ "serving": {
811
+ "bos_cache_mode": "discard"
812
+ }
813
+ }
crawler/crawler-b10-ctx16.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38980a287835ddb2345317a62bec797977f653e97f5c8e1e403ecd3c8c3da577
3
+ size 17799441
crawler/crawler-b10-ctx32.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa4cf0553c58ff4a492057b48de993d6fc570d004f76777a0b4ac12c4230b1b9
3
+ size 17817110
crawler/crawler-b10-ctx64.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d71268c9e76bfe599b715e83405817f2685eed104d389260893c09ad167053d
3
+ size 17907772
crawler/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:31d73a2127c85bfac6d25caec706bb7d68821eda55d834e0b70a3fe392293f58
3
+ size 17023276