Arijit-07 commited on
Commit
b777b7b
·
verified ·
1 Parent(s): d0f112c

Upload training_log.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. training_log.json +1282 -0
training_log.json ADDED
@@ -0,0 +1,1282 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "episode": 1,
4
+ "task_id": "easy",
5
+ "score": 0.54,
6
+ "rolling_avg": 0.54,
7
+ "loss": -0.12143471837043762,
8
+ "elapsed_min": 0.2
9
+ },
10
+ {
11
+ "episode": 2,
12
+ "task_id": "easy",
13
+ "score": 1.6400000000000001,
14
+ "rolling_avg": 1.09,
15
+ "loss": -0.12805303931236267,
16
+ "elapsed_min": 1.2
17
+ },
18
+ {
19
+ "episode": 3,
20
+ "task_id": "easy",
21
+ "score": 0.44,
22
+ "rolling_avg": 0.8733333333333334,
23
+ "loss": -0.06693907082080841,
24
+ "elapsed_min": 1.4
25
+ },
26
+ {
27
+ "episode": 4,
28
+ "task_id": "easy",
29
+ "score": 0.78,
30
+ "rolling_avg": 0.8500000000000001,
31
+ "loss": -0.16535498201847076,
32
+ "elapsed_min": 1.7
33
+ },
34
+ {
35
+ "episode": 5,
36
+ "task_id": "easy",
37
+ "score": 0.88,
38
+ "rolling_avg": 0.8560000000000001,
39
+ "loss": -0.11092036217451096,
40
+ "elapsed_min": 2.1
41
+ },
42
+ {
43
+ "episode": 6,
44
+ "task_id": "easy",
45
+ "score": 0.37,
46
+ "rolling_avg": 0.775,
47
+ "loss": -0.18090753257274628,
48
+ "elapsed_min": 2.2
49
+ },
50
+ {
51
+ "episode": 7,
52
+ "task_id": "easy",
53
+ "score": 0.37,
54
+ "rolling_avg": 0.7171428571428572,
55
+ "loss": -0.18480174243450165,
56
+ "elapsed_min": 2.3
57
+ },
58
+ {
59
+ "episode": 8,
60
+ "task_id": "easy",
61
+ "score": 0.54,
62
+ "rolling_avg": 0.6950000000000001,
63
+ "loss": -0.1352842003107071,
64
+ "elapsed_min": 2.4
65
+ },
66
+ {
67
+ "episode": 9,
68
+ "task_id": "easy",
69
+ "score": 0.37,
70
+ "rolling_avg": 0.658888888888889,
71
+ "loss": -0.10219474881887436,
72
+ "elapsed_min": 2.5
73
+ },
74
+ {
75
+ "episode": 10,
76
+ "task_id": "easy",
77
+ "score": 1.74,
78
+ "rolling_avg": 0.7670000000000001,
79
+ "loss": -0.12895138561725616,
80
+ "elapsed_min": 3.5
81
+ },
82
+ {
83
+ "episode": 11,
84
+ "task_id": "easy",
85
+ "score": 0.44,
86
+ "rolling_avg": 0.7570000000000001,
87
+ "loss": -0.1572495400905609,
88
+ "elapsed_min": 3.7
89
+ },
90
+ {
91
+ "episode": 12,
92
+ "task_id": "easy",
93
+ "score": 0.39,
94
+ "rolling_avg": 0.632,
95
+ "loss": -0.08590440452098846,
96
+ "elapsed_min": 3.9
97
+ },
98
+ {
99
+ "episode": 13,
100
+ "task_id": "easy",
101
+ "score": 0.5800000000000001,
102
+ "rolling_avg": 0.6460000000000001,
103
+ "loss": -0.07493701577186584,
104
+ "elapsed_min": 4.3
105
+ },
106
+ {
107
+ "episode": 14,
108
+ "task_id": "easy",
109
+ "score": 0.46,
110
+ "rolling_avg": 0.6140000000000001,
111
+ "loss": -0.1000489741563797,
112
+ "elapsed_min": 4.5
113
+ },
114
+ {
115
+ "episode": 15,
116
+ "task_id": "easy",
117
+ "score": 0.37,
118
+ "rolling_avg": 0.563,
119
+ "loss": -0.24669808149337769,
120
+ "elapsed_min": 4.6
121
+ },
122
+ {
123
+ "episode": 16,
124
+ "task_id": "easy",
125
+ "score": 1.6899999999999997,
126
+ "rolling_avg": 0.695,
127
+ "loss": -0.13417300581932068,
128
+ "elapsed_min": 5.6
129
+ },
130
+ {
131
+ "episode": 17,
132
+ "task_id": "easy",
133
+ "score": 0.7400000000000002,
134
+ "rolling_avg": 0.732,
135
+ "loss": -0.02531382441520691,
136
+ "elapsed_min": 6.7
137
+ },
138
+ {
139
+ "episode": 18,
140
+ "task_id": "easy",
141
+ "score": 0.8900000000000001,
142
+ "rolling_avg": 0.767,
143
+ "loss": -0.02539382129907608,
144
+ "elapsed_min": 7.7
145
+ },
146
+ {
147
+ "episode": 19,
148
+ "task_id": "easy",
149
+ "score": 0.7900000000000003,
150
+ "rolling_avg": 0.8090000000000002,
151
+ "loss": -0.024298425763845444,
152
+ "elapsed_min": 8.8
153
+ },
154
+ {
155
+ "episode": 20,
156
+ "task_id": "easy",
157
+ "score": 0.7400000000000002,
158
+ "rolling_avg": 0.7090000000000001,
159
+ "loss": -0.05224194377660751,
160
+ "elapsed_min": 9.8
161
+ },
162
+ {
163
+ "episode": 21,
164
+ "task_id": "easy",
165
+ "score": 0.7400000000000002,
166
+ "rolling_avg": 0.7390000000000001,
167
+ "loss": -0.058588918298482895,
168
+ "elapsed_min": 10.9
169
+ },
170
+ {
171
+ "episode": 22,
172
+ "task_id": "easy",
173
+ "score": 0.8900000000000001,
174
+ "rolling_avg": 0.789,
175
+ "loss": -0.10121776908636093,
176
+ "elapsed_min": 11.9
177
+ },
178
+ {
179
+ "episode": 23,
180
+ "task_id": "easy",
181
+ "score": 0.7400000000000002,
182
+ "rolling_avg": 0.805,
183
+ "loss": -0.05905468389391899,
184
+ "elapsed_min": 13.0
185
+ },
186
+ {
187
+ "episode": 24,
188
+ "task_id": "easy",
189
+ "score": 0.94,
190
+ "rolling_avg": 0.8530000000000001,
191
+ "loss": -0.11711429059505463,
192
+ "elapsed_min": 14.0
193
+ },
194
+ {
195
+ "episode": 25,
196
+ "task_id": "easy",
197
+ "score": 0.8400000000000001,
198
+ "rolling_avg": 0.9,
199
+ "loss": -0.09598871320486069,
200
+ "elapsed_min": 15.1
201
+ },
202
+ {
203
+ "episode": 26,
204
+ "task_id": "easy",
205
+ "score": 0.6900000000000002,
206
+ "rolling_avg": 0.8,
207
+ "loss": -0.0566844567656517,
208
+ "elapsed_min": 16.1
209
+ },
210
+ {
211
+ "episode": 27,
212
+ "task_id": "easy",
213
+ "score": 0.6400000000000001,
214
+ "rolling_avg": 0.79,
215
+ "loss": -0.05957688018679619,
216
+ "elapsed_min": 17.2
217
+ },
218
+ {
219
+ "episode": 28,
220
+ "task_id": "easy",
221
+ "score": 0.29,
222
+ "rolling_avg": 0.7300000000000002,
223
+ "loss": 0.0405953973531723,
224
+ "elapsed_min": 18.2
225
+ },
226
+ {
227
+ "episode": 29,
228
+ "task_id": "easy",
229
+ "score": 0.23999999999999996,
230
+ "rolling_avg": 0.675,
231
+ "loss": 0.0662434920668602,
232
+ "elapsed_min": 19.3
233
+ },
234
+ {
235
+ "episode": 30,
236
+ "task_id": "easy",
237
+ "score": 0.29000000000000004,
238
+ "rolling_avg": 0.6300000000000001,
239
+ "loss": 0.04871468245983124,
240
+ "elapsed_min": 20.3
241
+ },
242
+ {
243
+ "episode": 31,
244
+ "task_id": "easy",
245
+ "score": 0.3400000000000001,
246
+ "rolling_avg": 0.5900000000000001,
247
+ "loss": 0.0225782822817564,
248
+ "elapsed_min": 21.3
249
+ },
250
+ {
251
+ "episode": 32,
252
+ "task_id": "easy",
253
+ "score": 0.39,
254
+ "rolling_avg": 0.54,
255
+ "loss": 0.0005533260991796851,
256
+ "elapsed_min": 22.3
257
+ },
258
+ {
259
+ "episode": 33,
260
+ "task_id": "easy",
261
+ "score": 0.39000000000000007,
262
+ "rolling_avg": 0.505,
263
+ "loss": -0.00933866761624813,
264
+ "elapsed_min": 23.4
265
+ },
266
+ {
267
+ "episode": 34,
268
+ "task_id": "easy",
269
+ "score": 0.29,
270
+ "rolling_avg": 0.44000000000000006,
271
+ "loss": 0.025673745200037956,
272
+ "elapsed_min": 24.4
273
+ },
274
+ {
275
+ "episode": 35,
276
+ "task_id": "easy",
277
+ "score": 0.6900000000000002,
278
+ "rolling_avg": 0.4250000000000001,
279
+ "loss": -0.084134541451931,
280
+ "elapsed_min": 25.4
281
+ },
282
+ {
283
+ "episode": 36,
284
+ "task_id": "easy",
285
+ "score": 0.7400000000000002,
286
+ "rolling_avg": 0.43000000000000005,
287
+ "loss": -0.053947098553180695,
288
+ "elapsed_min": 26.4
289
+ },
290
+ {
291
+ "episode": 37,
292
+ "task_id": "easy",
293
+ "score": 0.9400000000000002,
294
+ "rolling_avg": 0.46000000000000013,
295
+ "loss": -0.1245078295469284,
296
+ "elapsed_min": 27.4
297
+ },
298
+ {
299
+ "episode": 38,
300
+ "task_id": "easy",
301
+ "score": 1.09,
302
+ "rolling_avg": 0.5400000000000001,
303
+ "loss": -0.1214509904384613,
304
+ "elapsed_min": 28.5
305
+ },
306
+ {
307
+ "episode": 39,
308
+ "task_id": "easy",
309
+ "score": 0.6900000000000002,
310
+ "rolling_avg": 0.5850000000000002,
311
+ "loss": -0.07190271466970444,
312
+ "elapsed_min": 29.5
313
+ },
314
+ {
315
+ "episode": 40,
316
+ "task_id": "easy",
317
+ "score": 0.78,
318
+ "rolling_avg": 0.6340000000000001,
319
+ "loss": -0.11039736866950989,
320
+ "elapsed_min": 30.3
321
+ },
322
+ {
323
+ "episode": 41,
324
+ "task_id": "medium",
325
+ "score": 0.49000000000000005,
326
+ "rolling_avg": 0.49000000000000005,
327
+ "loss": -0.09235180914402008,
328
+ "elapsed_min": 31.4
329
+ },
330
+ {
331
+ "episode": 42,
332
+ "task_id": "medium",
333
+ "score": 0.8400000000000001,
334
+ "rolling_avg": 0.665,
335
+ "loss": -0.176436185836792,
336
+ "elapsed_min": 32.4
337
+ },
338
+ {
339
+ "episode": 43,
340
+ "task_id": "medium",
341
+ "score": 0.39000000000000007,
342
+ "rolling_avg": 0.5733333333333334,
343
+ "loss": -0.047593407332897186,
344
+ "elapsed_min": 33.6
345
+ },
346
+ {
347
+ "episode": 44,
348
+ "task_id": "medium",
349
+ "score": 0.3400000000000001,
350
+ "rolling_avg": 0.5150000000000001,
351
+ "loss": -0.040770307183265686,
352
+ "elapsed_min": 34.7
353
+ },
354
+ {
355
+ "episode": 45,
356
+ "task_id": "medium",
357
+ "score": 0.29,
358
+ "rolling_avg": 0.4700000000000001,
359
+ "loss": -0.02725624106824398,
360
+ "elapsed_min": 35.7
361
+ },
362
+ {
363
+ "episode": 46,
364
+ "task_id": "medium",
365
+ "score": 0.39000000000000007,
366
+ "rolling_avg": 0.4566666666666668,
367
+ "loss": -0.046115946024656296,
368
+ "elapsed_min": 36.8
369
+ },
370
+ {
371
+ "episode": 47,
372
+ "task_id": "medium",
373
+ "score": 0.23999999999999996,
374
+ "rolling_avg": 0.42571428571428577,
375
+ "loss": 0.00903060007840395,
376
+ "elapsed_min": 37.9
377
+ },
378
+ {
379
+ "episode": 48,
380
+ "task_id": "medium",
381
+ "score": 0.23999999999999996,
382
+ "rolling_avg": 0.4025,
383
+ "loss": 0.01689656637609005,
384
+ "elapsed_min": 39.0
385
+ },
386
+ {
387
+ "episode": 49,
388
+ "task_id": "medium",
389
+ "score": 0.34,
390
+ "rolling_avg": 0.39555555555555555,
391
+ "loss": -0.028721345588564873,
392
+ "elapsed_min": 40.1
393
+ },
394
+ {
395
+ "episode": 50,
396
+ "task_id": "medium",
397
+ "score": 0.39000000000000007,
398
+ "rolling_avg": 0.395,
399
+ "loss": -0.07144424319267273,
400
+ "elapsed_min": 41.2
401
+ },
402
+ {
403
+ "episode": 51,
404
+ "task_id": "medium",
405
+ "score": 0.5400000000000001,
406
+ "rolling_avg": 0.4,
407
+ "loss": -0.10756971687078476,
408
+ "elapsed_min": 42.2
409
+ },
410
+ {
411
+ "episode": 52,
412
+ "task_id": "medium",
413
+ "score": 0.4400000000000001,
414
+ "rolling_avg": 0.36000000000000004,
415
+ "loss": -0.10540562868118286,
416
+ "elapsed_min": 43.3
417
+ },
418
+ {
419
+ "episode": 53,
420
+ "task_id": "medium",
421
+ "score": 0.3400000000000001,
422
+ "rolling_avg": 0.3550000000000001,
423
+ "loss": -0.01990152895450592,
424
+ "elapsed_min": 44.4
425
+ },
426
+ {
427
+ "episode": 54,
428
+ "task_id": "medium",
429
+ "score": 0.23999999999999996,
430
+ "rolling_avg": 0.345,
431
+ "loss": 0.011199951171875,
432
+ "elapsed_min": 45.4
433
+ },
434
+ {
435
+ "episode": 55,
436
+ "task_id": "medium",
437
+ "score": 0.34,
438
+ "rolling_avg": 0.35,
439
+ "loss": -0.0704483762383461,
440
+ "elapsed_min": 46.5
441
+ },
442
+ {
443
+ "episode": 56,
444
+ "task_id": "medium",
445
+ "score": 0.38999999999999996,
446
+ "rolling_avg": 0.35,
447
+ "loss": -0.08460726588964462,
448
+ "elapsed_min": 47.6
449
+ },
450
+ {
451
+ "episode": 57,
452
+ "task_id": "medium",
453
+ "score": 0.4900000000000001,
454
+ "rolling_avg": 0.37500000000000006,
455
+ "loss": -0.09814709424972534,
456
+ "elapsed_min": 48.7
457
+ },
458
+ {
459
+ "episode": 58,
460
+ "task_id": "medium",
461
+ "score": 0.39,
462
+ "rolling_avg": 0.39000000000000007,
463
+ "loss": -0.040027916431427,
464
+ "elapsed_min": 49.7
465
+ },
466
+ {
467
+ "episode": 59,
468
+ "task_id": "medium",
469
+ "score": 0.38999999999999996,
470
+ "rolling_avg": 0.39500000000000013,
471
+ "loss": -0.08392232656478882,
472
+ "elapsed_min": 50.8
473
+ },
474
+ {
475
+ "episode": 60,
476
+ "task_id": "medium",
477
+ "score": 0.23999999999999996,
478
+ "rolling_avg": 0.38000000000000006,
479
+ "loss": -0.009092062711715698,
480
+ "elapsed_min": 51.9
481
+ },
482
+ {
483
+ "episode": 61,
484
+ "task_id": "medium",
485
+ "score": 0.44000000000000006,
486
+ "rolling_avg": 0.37,
487
+ "loss": -0.058374952524900436,
488
+ "elapsed_min": 53.0
489
+ },
490
+ {
491
+ "episode": 62,
492
+ "task_id": "medium",
493
+ "score": 0.43999999999999995,
494
+ "rolling_avg": 0.37,
495
+ "loss": -0.14105059206485748,
496
+ "elapsed_min": 54.1
497
+ },
498
+ {
499
+ "episode": 63,
500
+ "task_id": "medium",
501
+ "score": 0.4400000000000001,
502
+ "rolling_avg": 0.38,
503
+ "loss": -0.08380116522312164,
504
+ "elapsed_min": 55.1
505
+ },
506
+ {
507
+ "episode": 64,
508
+ "task_id": "medium",
509
+ "score": 0.49000000000000005,
510
+ "rolling_avg": 0.40499999999999997,
511
+ "loss": -0.11519451439380646,
512
+ "elapsed_min": 56.2
513
+ },
514
+ {
515
+ "episode": 65,
516
+ "task_id": "medium",
517
+ "score": 0.29,
518
+ "rolling_avg": 0.4,
519
+ "loss": -0.017708923667669296,
520
+ "elapsed_min": 57.2
521
+ },
522
+ {
523
+ "episode": 66,
524
+ "task_id": "medium",
525
+ "score": 0.23999999999999996,
526
+ "rolling_avg": 0.385,
527
+ "loss": -0.007058671675622463,
528
+ "elapsed_min": 58.3
529
+ },
530
+ {
531
+ "episode": 67,
532
+ "task_id": "medium",
533
+ "score": 0.29000000000000004,
534
+ "rolling_avg": 0.365,
535
+ "loss": -0.028480907902121544,
536
+ "elapsed_min": 59.4
537
+ },
538
+ {
539
+ "episode": 68,
540
+ "task_id": "medium",
541
+ "score": 0.23999999999999996,
542
+ "rolling_avg": 0.35,
543
+ "loss": 0.0003400370478630066,
544
+ "elapsed_min": 60.4
545
+ },
546
+ {
547
+ "episode": 69,
548
+ "task_id": "medium",
549
+ "score": 0.3400000000000001,
550
+ "rolling_avg": 0.34500000000000003,
551
+ "loss": -0.034816063940525055,
552
+ "elapsed_min": 61.5
553
+ },
554
+ {
555
+ "episode": 70,
556
+ "task_id": "medium",
557
+ "score": 0.4400000000000001,
558
+ "rolling_avg": 0.365,
559
+ "loss": -0.12672019004821777,
560
+ "elapsed_min": 62.5
561
+ },
562
+ {
563
+ "episode": 71,
564
+ "task_id": "medium",
565
+ "score": 0.54,
566
+ "rolling_avg": 0.37500000000000006,
567
+ "loss": -0.1321611851453781,
568
+ "elapsed_min": 63.6
569
+ },
570
+ {
571
+ "episode": 72,
572
+ "task_id": "medium",
573
+ "score": 0.4900000000000001,
574
+ "rolling_avg": 0.38,
575
+ "loss": -0.11640733480453491,
576
+ "elapsed_min": 64.7
577
+ },
578
+ {
579
+ "episode": 73,
580
+ "task_id": "medium",
581
+ "score": 0.39000000000000007,
582
+ "rolling_avg": 0.37500000000000006,
583
+ "loss": -0.08983750641345978,
584
+ "elapsed_min": 65.7
585
+ },
586
+ {
587
+ "episode": 74,
588
+ "task_id": "medium",
589
+ "score": 0.39000000000000007,
590
+ "rolling_avg": 0.36500000000000005,
591
+ "loss": -0.06033878028392792,
592
+ "elapsed_min": 66.8
593
+ },
594
+ {
595
+ "episode": 75,
596
+ "task_id": "medium",
597
+ "score": 0.33999999999999997,
598
+ "rolling_avg": 0.37000000000000005,
599
+ "loss": -0.046499669551849365,
600
+ "elapsed_min": 67.9
601
+ },
602
+ {
603
+ "episode": 76,
604
+ "task_id": "medium",
605
+ "score": 0.3400000000000001,
606
+ "rolling_avg": 0.38000000000000006,
607
+ "loss": -0.029506457969546318,
608
+ "elapsed_min": 68.9
609
+ },
610
+ {
611
+ "episode": 77,
612
+ "task_id": "medium",
613
+ "score": 0.39000000000000007,
614
+ "rolling_avg": 0.39000000000000007,
615
+ "loss": -0.08039389550685883,
616
+ "elapsed_min": 70.0
617
+ },
618
+ {
619
+ "episode": 78,
620
+ "task_id": "medium",
621
+ "score": 0.34,
622
+ "rolling_avg": 0.4000000000000001,
623
+ "loss": -0.0734604224562645,
624
+ "elapsed_min": 71.0
625
+ },
626
+ {
627
+ "episode": 79,
628
+ "task_id": "medium",
629
+ "score": 0.23999999999999996,
630
+ "rolling_avg": 0.39,
631
+ "loss": -0.020788073539733887,
632
+ "elapsed_min": 72.1
633
+ },
634
+ {
635
+ "episode": 80,
636
+ "task_id": "medium",
637
+ "score": 0.23999999999999996,
638
+ "rolling_avg": 0.37,
639
+ "loss": -0.009029777720570564,
640
+ "elapsed_min": 73.2
641
+ },
642
+ {
643
+ "episode": 81,
644
+ "task_id": "hard",
645
+ "score": 1.5899999999999999,
646
+ "rolling_avg": 1.5899999999999999,
647
+ "loss": -0.10761424154043198,
648
+ "elapsed_min": 74.3
649
+ },
650
+ {
651
+ "episode": 82,
652
+ "task_id": "hard",
653
+ "score": 1.6399999999999997,
654
+ "rolling_avg": 1.6149999999999998,
655
+ "loss": -0.07337230443954468,
656
+ "elapsed_min": 75.3
657
+ },
658
+ {
659
+ "episode": 83,
660
+ "task_id": "hard",
661
+ "score": 1.5399999999999996,
662
+ "rolling_avg": 1.5899999999999999,
663
+ "loss": -0.10122022032737732,
664
+ "elapsed_min": 76.4
665
+ },
666
+ {
667
+ "episode": 84,
668
+ "task_id": "hard",
669
+ "score": 1.7399999999999995,
670
+ "rolling_avg": 1.6274999999999997,
671
+ "loss": -0.05785955488681793,
672
+ "elapsed_min": 77.5
673
+ },
674
+ {
675
+ "episode": 85,
676
+ "task_id": "hard",
677
+ "score": 2.0399999999999996,
678
+ "rolling_avg": 1.7099999999999997,
679
+ "loss": -0.09229159355163574,
680
+ "elapsed_min": 78.6
681
+ },
682
+ {
683
+ "episode": 86,
684
+ "task_id": "hard",
685
+ "score": 1.7899999999999996,
686
+ "rolling_avg": 1.723333333333333,
687
+ "loss": -0.1031956821680069,
688
+ "elapsed_min": 79.6
689
+ },
690
+ {
691
+ "episode": 87,
692
+ "task_id": "hard",
693
+ "score": 1.5399999999999998,
694
+ "rolling_avg": 1.6971428571428568,
695
+ "loss": -0.11315083503723145,
696
+ "elapsed_min": 80.7
697
+ },
698
+ {
699
+ "episode": 88,
700
+ "task_id": "hard",
701
+ "score": 1.14,
702
+ "rolling_avg": 1.6274999999999997,
703
+ "loss": -0.06293876469135284,
704
+ "elapsed_min": 81.8
705
+ },
706
+ {
707
+ "episode": 89,
708
+ "task_id": "hard",
709
+ "score": 0.9900000000000002,
710
+ "rolling_avg": 1.5566666666666664,
711
+ "loss": -0.09899002313613892,
712
+ "elapsed_min": 82.9
713
+ },
714
+ {
715
+ "episode": 90,
716
+ "task_id": "hard",
717
+ "score": 1.3399999999999999,
718
+ "rolling_avg": 1.5349999999999997,
719
+ "loss": -0.07810334116220474,
720
+ "elapsed_min": 84.0
721
+ },
722
+ {
723
+ "episode": 91,
724
+ "task_id": "hard",
725
+ "score": 0.8900000000000001,
726
+ "rolling_avg": 1.4649999999999999,
727
+ "loss": -0.10680361092090607,
728
+ "elapsed_min": 85.1
729
+ },
730
+ {
731
+ "episode": 92,
732
+ "task_id": "hard",
733
+ "score": 1.4899999999999998,
734
+ "rolling_avg": 1.4499999999999997,
735
+ "loss": -0.1284235566854477,
736
+ "elapsed_min": 86.2
737
+ },
738
+ {
739
+ "episode": 93,
740
+ "task_id": "hard",
741
+ "score": 1.09,
742
+ "rolling_avg": 1.4049999999999998,
743
+ "loss": -0.11051454395055771,
744
+ "elapsed_min": 87.3
745
+ },
746
+ {
747
+ "episode": 94,
748
+ "task_id": "hard",
749
+ "score": 1.14,
750
+ "rolling_avg": 1.3450000000000002,
751
+ "loss": -0.15035489201545715,
752
+ "elapsed_min": 88.4
753
+ },
754
+ {
755
+ "episode": 95,
756
+ "task_id": "hard",
757
+ "score": 0.94,
758
+ "rolling_avg": 1.2349999999999999,
759
+ "loss": -0.11060954630374908,
760
+ "elapsed_min": 89.4
761
+ },
762
+ {
763
+ "episode": 96,
764
+ "task_id": "hard",
765
+ "score": 0.44000000000000006,
766
+ "rolling_avg": 1.1,
767
+ "loss": -0.05425233766436577,
768
+ "elapsed_min": 90.5
769
+ },
770
+ {
771
+ "episode": 97,
772
+ "task_id": "hard",
773
+ "score": 0.4900000000000001,
774
+ "rolling_avg": 0.9949999999999999,
775
+ "loss": -0.08663400262594223,
776
+ "elapsed_min": 91.6
777
+ },
778
+ {
779
+ "episode": 98,
780
+ "task_id": "hard",
781
+ "score": 0.8399999999999999,
782
+ "rolling_avg": 0.9649999999999999,
783
+ "loss": -0.059657029807567596,
784
+ "elapsed_min": 92.7
785
+ },
786
+ {
787
+ "episode": 99,
788
+ "task_id": "hard",
789
+ "score": 0.6400000000000001,
790
+ "rolling_avg": 0.93,
791
+ "loss": -0.06711545586585999,
792
+ "elapsed_min": 93.8
793
+ },
794
+ {
795
+ "episode": 100,
796
+ "task_id": "hard",
797
+ "score": 0.44000000000000017,
798
+ "rolling_avg": 0.8399999999999999,
799
+ "loss": -0.02081288956105709,
800
+ "elapsed_min": 94.8
801
+ },
802
+ {
803
+ "episode": 101,
804
+ "task_id": "hard",
805
+ "score": 0.54,
806
+ "rolling_avg": 0.805,
807
+ "loss": -0.07743717730045319,
808
+ "elapsed_min": 95.9
809
+ },
810
+ {
811
+ "episode": 102,
812
+ "task_id": "hard",
813
+ "score": 0.44000000000000006,
814
+ "rolling_avg": 0.7000000000000001,
815
+ "loss": -0.018033726140856743,
816
+ "elapsed_min": 97.0
817
+ },
818
+ {
819
+ "episode": 103,
820
+ "task_id": "hard",
821
+ "score": 0.54,
822
+ "rolling_avg": 0.6450000000000001,
823
+ "loss": -0.07679533958435059,
824
+ "elapsed_min": 98.1
825
+ },
826
+ {
827
+ "episode": 104,
828
+ "task_id": "hard",
829
+ "score": 0.54,
830
+ "rolling_avg": 0.5850000000000001,
831
+ "loss": -0.06460466980934143,
832
+ "elapsed_min": 99.2
833
+ },
834
+ {
835
+ "episode": 105,
836
+ "task_id": "hard",
837
+ "score": 0.5400000000000001,
838
+ "rolling_avg": 0.5450000000000002,
839
+ "loss": -0.07421746850013733,
840
+ "elapsed_min": 100.2
841
+ },
842
+ {
843
+ "episode": 106,
844
+ "task_id": "hard",
845
+ "score": 0.4900000000000001,
846
+ "rolling_avg": 0.55,
847
+ "loss": -0.10693149268627167,
848
+ "elapsed_min": 101.3
849
+ },
850
+ {
851
+ "episode": 107,
852
+ "task_id": "hard",
853
+ "score": 0.39,
854
+ "rolling_avg": 0.54,
855
+ "loss": -0.041160814464092255,
856
+ "elapsed_min": 102.4
857
+ },
858
+ {
859
+ "episode": 108,
860
+ "task_id": "hard",
861
+ "score": 0.23999999999999996,
862
+ "rolling_avg": 0.4800000000000001,
863
+ "loss": 0.0243326835334301,
864
+ "elapsed_min": 103.5
865
+ },
866
+ {
867
+ "episode": 109,
868
+ "task_id": "hard",
869
+ "score": 0.34,
870
+ "rolling_avg": 0.45000000000000007,
871
+ "loss": -0.025716159492731094,
872
+ "elapsed_min": 104.5
873
+ },
874
+ {
875
+ "episode": 110,
876
+ "task_id": "hard",
877
+ "score": 0.5900000000000001,
878
+ "rolling_avg": 0.465,
879
+ "loss": -0.0469183623790741,
880
+ "elapsed_min": 105.6
881
+ },
882
+ {
883
+ "episode": 111,
884
+ "task_id": "hard",
885
+ "score": 0.29,
886
+ "rolling_avg": 0.44000000000000006,
887
+ "loss": -0.003872685134410858,
888
+ "elapsed_min": 106.7
889
+ },
890
+ {
891
+ "episode": 112,
892
+ "task_id": "hard",
893
+ "score": 0.43999999999999995,
894
+ "rolling_avg": 0.44000000000000006,
895
+ "loss": -0.01721161976456642,
896
+ "elapsed_min": 107.8
897
+ },
898
+ {
899
+ "episode": 113,
900
+ "task_id": "hard",
901
+ "score": 0.6900000000000002,
902
+ "rolling_avg": 0.45499999999999996,
903
+ "loss": -0.10440249741077423,
904
+ "elapsed_min": 108.9
905
+ },
906
+ {
907
+ "episode": 114,
908
+ "task_id": "hard",
909
+ "score": 0.33999999999999997,
910
+ "rolling_avg": 0.43500000000000005,
911
+ "loss": -0.031179871410131454,
912
+ "elapsed_min": 109.9
913
+ },
914
+ {
915
+ "episode": 115,
916
+ "task_id": "hard",
917
+ "score": 0.49000000000000005,
918
+ "rolling_avg": 0.43000000000000005,
919
+ "loss": -0.05355419963598251,
920
+ "elapsed_min": 111.0
921
+ },
922
+ {
923
+ "episode": 116,
924
+ "task_id": "hard",
925
+ "score": 0.5900000000000001,
926
+ "rolling_avg": 0.44000000000000006,
927
+ "loss": -0.04942004010081291,
928
+ "elapsed_min": 112.1
929
+ },
930
+ {
931
+ "episode": 117,
932
+ "task_id": "hard",
933
+ "score": 0.49000000000000016,
934
+ "rolling_avg": 0.45,
935
+ "loss": -0.04643632099032402,
936
+ "elapsed_min": 113.2
937
+ },
938
+ {
939
+ "episode": 118,
940
+ "task_id": "hard",
941
+ "score": 0.8400000000000001,
942
+ "rolling_avg": 0.51,
943
+ "loss": -0.05764473229646683,
944
+ "elapsed_min": 114.3
945
+ },
946
+ {
947
+ "episode": 119,
948
+ "task_id": "hard",
949
+ "score": 1.19,
950
+ "rolling_avg": 0.5950000000000001,
951
+ "loss": -0.12574931979179382,
952
+ "elapsed_min": 115.4
953
+ },
954
+ {
955
+ "episode": 120,
956
+ "task_id": "hard",
957
+ "score": 0.99,
958
+ "rolling_avg": 0.6350000000000001,
959
+ "loss": -0.07021882385015488,
960
+ "elapsed_min": 116.4
961
+ },
962
+ {
963
+ "episode": 121,
964
+ "task_id": "bonus",
965
+ "score": 0.72,
966
+ "rolling_avg": 0.72,
967
+ "loss": -0.1563481241464615,
968
+ "elapsed_min": 117.6
969
+ },
970
+ {
971
+ "episode": 122,
972
+ "task_id": "bonus",
973
+ "score": 0.6800000000000002,
974
+ "rolling_avg": 0.7000000000000001,
975
+ "loss": -0.13133162260055542,
976
+ "elapsed_min": 118.7
977
+ },
978
+ {
979
+ "episode": 123,
980
+ "task_id": "bonus",
981
+ "score": 0.76,
982
+ "rolling_avg": 0.7200000000000001,
983
+ "loss": -0.11883395165205002,
984
+ "elapsed_min": 119.9
985
+ },
986
+ {
987
+ "episode": 124,
988
+ "task_id": "bonus",
989
+ "score": 0.62,
990
+ "rolling_avg": 0.6950000000000001,
991
+ "loss": -0.12456952035427094,
992
+ "elapsed_min": 121.0
993
+ },
994
+ {
995
+ "episode": 125,
996
+ "task_id": "bonus",
997
+ "score": 0.49000000000000016,
998
+ "rolling_avg": 0.6540000000000001,
999
+ "loss": -0.10143512487411499,
1000
+ "elapsed_min": 122.1
1001
+ },
1002
+ {
1003
+ "episode": 126,
1004
+ "task_id": "bonus",
1005
+ "score": 0.7000000000000001,
1006
+ "rolling_avg": 0.6616666666666667,
1007
+ "loss": -0.13113725185394287,
1008
+ "elapsed_min": 123.2
1009
+ },
1010
+ {
1011
+ "episode": 127,
1012
+ "task_id": "bonus",
1013
+ "score": 0.7700000000000002,
1014
+ "rolling_avg": 0.6771428571428573,
1015
+ "loss": -0.11259196698665619,
1016
+ "elapsed_min": 124.3
1017
+ },
1018
+ {
1019
+ "episode": 128,
1020
+ "task_id": "bonus",
1021
+ "score": 0.81,
1022
+ "rolling_avg": 0.6937500000000001,
1023
+ "loss": -0.16549530625343323,
1024
+ "elapsed_min": 125.4
1025
+ },
1026
+ {
1027
+ "episode": 129,
1028
+ "task_id": "bonus",
1029
+ "score": 0.6700000000000002,
1030
+ "rolling_avg": 0.6911111111111112,
1031
+ "loss": -0.08555784821510315,
1032
+ "elapsed_min": 126.6
1033
+ },
1034
+ {
1035
+ "episode": 130,
1036
+ "task_id": "bonus",
1037
+ "score": 0.7300000000000002,
1038
+ "rolling_avg": 0.6950000000000001,
1039
+ "loss": -0.1284562349319458,
1040
+ "elapsed_min": 127.7
1041
+ },
1042
+ {
1043
+ "episode": 131,
1044
+ "task_id": "bonus",
1045
+ "score": 0.75,
1046
+ "rolling_avg": 0.6980000000000001,
1047
+ "loss": -0.13779829442501068,
1048
+ "elapsed_min": 128.8
1049
+ },
1050
+ {
1051
+ "episode": 132,
1052
+ "task_id": "bonus",
1053
+ "score": 0.7500000000000001,
1054
+ "rolling_avg": 0.7050000000000001,
1055
+ "loss": -0.10122223943471909,
1056
+ "elapsed_min": 130.0
1057
+ },
1058
+ {
1059
+ "episode": 133,
1060
+ "task_id": "bonus",
1061
+ "score": 0.6200000000000001,
1062
+ "rolling_avg": 0.6910000000000001,
1063
+ "loss": -0.10923080146312714,
1064
+ "elapsed_min": 131.1
1065
+ },
1066
+ {
1067
+ "episode": 134,
1068
+ "task_id": "bonus",
1069
+ "score": 0.5200000000000001,
1070
+ "rolling_avg": 0.6810000000000002,
1071
+ "loss": -0.13451352715492249,
1072
+ "elapsed_min": 132.3
1073
+ },
1074
+ {
1075
+ "episode": 135,
1076
+ "task_id": "bonus",
1077
+ "score": 0.7300000000000002,
1078
+ "rolling_avg": 0.7050000000000002,
1079
+ "loss": -0.16815370321273804,
1080
+ "elapsed_min": 133.5
1081
+ },
1082
+ {
1083
+ "episode": 136,
1084
+ "task_id": "bonus",
1085
+ "score": 0.9600000000000002,
1086
+ "rolling_avg": 0.7310000000000001,
1087
+ "loss": -0.1660919487476349,
1088
+ "elapsed_min": 134.6
1089
+ },
1090
+ {
1091
+ "episode": 137,
1092
+ "task_id": "bonus",
1093
+ "score": 0.6900000000000002,
1094
+ "rolling_avg": 0.7230000000000001,
1095
+ "loss": -0.13483691215515137,
1096
+ "elapsed_min": 135.7
1097
+ },
1098
+ {
1099
+ "episode": 138,
1100
+ "task_id": "bonus",
1101
+ "score": 0.5000000000000001,
1102
+ "rolling_avg": 0.6920000000000002,
1103
+ "loss": -0.10621734708547592,
1104
+ "elapsed_min": 136.8
1105
+ },
1106
+ {
1107
+ "episode": 139,
1108
+ "task_id": "bonus",
1109
+ "score": 0.76,
1110
+ "rolling_avg": 0.7010000000000001,
1111
+ "loss": -0.14708703756332397,
1112
+ "elapsed_min": 138.0
1113
+ },
1114
+ {
1115
+ "episode": 140,
1116
+ "task_id": "bonus",
1117
+ "score": 0.52,
1118
+ "rolling_avg": 0.68,
1119
+ "loss": -0.058200687170028687,
1120
+ "elapsed_min": 139.1
1121
+ },
1122
+ {
1123
+ "episode": 141,
1124
+ "task_id": "bonus",
1125
+ "score": 0.6000000000000001,
1126
+ "rolling_avg": 0.665,
1127
+ "loss": -0.09626239538192749,
1128
+ "elapsed_min": 140.2
1129
+ },
1130
+ {
1131
+ "episode": 142,
1132
+ "task_id": "bonus",
1133
+ "score": 0.8800000000000001,
1134
+ "rolling_avg": 0.678,
1135
+ "loss": -0.19182077050209045,
1136
+ "elapsed_min": 141.4
1137
+ },
1138
+ {
1139
+ "episode": 143,
1140
+ "task_id": "bonus",
1141
+ "score": 0.6800000000000002,
1142
+ "rolling_avg": 0.6840000000000002,
1143
+ "loss": -0.15185901522636414,
1144
+ "elapsed_min": 142.4
1145
+ },
1146
+ {
1147
+ "episode": 144,
1148
+ "task_id": "bonus",
1149
+ "score": 0.6700000000000002,
1150
+ "rolling_avg": 0.6990000000000001,
1151
+ "loss": -0.13845515251159668,
1152
+ "elapsed_min": 143.5
1153
+ },
1154
+ {
1155
+ "episode": 145,
1156
+ "task_id": "bonus",
1157
+ "score": 0.4200000000000001,
1158
+ "rolling_avg": 0.6679999999999999,
1159
+ "loss": -0.055254243314266205,
1160
+ "elapsed_min": 144.6
1161
+ },
1162
+ {
1163
+ "episode": 146,
1164
+ "task_id": "bonus",
1165
+ "score": 0.81,
1166
+ "rolling_avg": 0.6530000000000001,
1167
+ "loss": -0.0969042181968689,
1168
+ "elapsed_min": 145.9
1169
+ },
1170
+ {
1171
+ "episode": 147,
1172
+ "task_id": "bonus",
1173
+ "score": 0.6000000000000001,
1174
+ "rolling_avg": 0.6440000000000001,
1175
+ "loss": -0.07441800832748413,
1176
+ "elapsed_min": 147.0
1177
+ },
1178
+ {
1179
+ "episode": 148,
1180
+ "task_id": "bonus",
1181
+ "score": 0.8200000000000001,
1182
+ "rolling_avg": 0.6759999999999999,
1183
+ "loss": -0.18915283679962158,
1184
+ "elapsed_min": 148.1
1185
+ },
1186
+ {
1187
+ "episode": 149,
1188
+ "task_id": "bonus",
1189
+ "score": 0.5700000000000001,
1190
+ "rolling_avg": 0.657,
1191
+ "loss": -0.10145045816898346,
1192
+ "elapsed_min": 149.3
1193
+ },
1194
+ {
1195
+ "episode": 150,
1196
+ "task_id": "bonus",
1197
+ "score": 0.81,
1198
+ "rolling_avg": 0.6860000000000002,
1199
+ "loss": -0.1386324167251587,
1200
+ "elapsed_min": 150.4
1201
+ },
1202
+ {
1203
+ "episode": 151,
1204
+ "task_id": "bonus",
1205
+ "score": 0.6900000000000001,
1206
+ "rolling_avg": 0.6950000000000002,
1207
+ "loss": -0.12235265225172043,
1208
+ "elapsed_min": 151.5
1209
+ },
1210
+ {
1211
+ "episode": 152,
1212
+ "task_id": "bonus",
1213
+ "score": 0.5400000000000001,
1214
+ "rolling_avg": 0.6610000000000001,
1215
+ "loss": -0.15657515823841095,
1216
+ "elapsed_min": 152.6
1217
+ },
1218
+ {
1219
+ "episode": 153,
1220
+ "task_id": "bonus",
1221
+ "score": 0.5800000000000002,
1222
+ "rolling_avg": 0.6510000000000001,
1223
+ "loss": -0.1349593997001648,
1224
+ "elapsed_min": 153.8
1225
+ },
1226
+ {
1227
+ "episode": 154,
1228
+ "task_id": "bonus",
1229
+ "score": 0.6900000000000001,
1230
+ "rolling_avg": 0.6530000000000002,
1231
+ "loss": -0.11009057611227036,
1232
+ "elapsed_min": 154.9
1233
+ },
1234
+ {
1235
+ "episode": 155,
1236
+ "task_id": "bonus",
1237
+ "score": 0.6400000000000001,
1238
+ "rolling_avg": 0.6750000000000002,
1239
+ "loss": -0.13187479972839355,
1240
+ "elapsed_min": 156.0
1241
+ },
1242
+ {
1243
+ "episode": 156,
1244
+ "task_id": "bonus",
1245
+ "score": 0.9100000000000004,
1246
+ "rolling_avg": 0.6850000000000002,
1247
+ "loss": -0.2191038727760315,
1248
+ "elapsed_min": 157.2
1249
+ },
1250
+ {
1251
+ "episode": 157,
1252
+ "task_id": "bonus",
1253
+ "score": 0.7000000000000002,
1254
+ "rolling_avg": 0.6950000000000002,
1255
+ "loss": -0.14810220897197723,
1256
+ "elapsed_min": 158.3
1257
+ },
1258
+ {
1259
+ "episode": 158,
1260
+ "task_id": "bonus",
1261
+ "score": 0.6200000000000001,
1262
+ "rolling_avg": 0.675,
1263
+ "loss": -0.12571939826011658,
1264
+ "elapsed_min": 159.4
1265
+ },
1266
+ {
1267
+ "episode": 159,
1268
+ "task_id": "bonus",
1269
+ "score": 0.7000000000000002,
1270
+ "rolling_avg": 0.6880000000000001,
1271
+ "loss": -0.11064934730529785,
1272
+ "elapsed_min": 160.5
1273
+ },
1274
+ {
1275
+ "episode": 160,
1276
+ "task_id": "bonus",
1277
+ "score": 0.5900000000000001,
1278
+ "rolling_avg": 0.6660000000000001,
1279
+ "loss": -0.09303998947143555,
1280
+ "elapsed_min": 161.6
1281
+ }
1282
+ ]