userGagan commited on
Commit
f310a03
1 Parent(s): 6cfc5d9

Upload trainer_state.json

Browse files
Files changed (1) hide show
  1. checkpoint-240/trainer_state.json +1708 -0
checkpoint-240/trainer_state.json ADDED
@@ -0,0 +1,1708 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.15978053212165833,
3
+ "best_model_checkpoint": "segformer-b0-finetuned-segments-sidewalk-outputs/checkpoint-240",
4
+ "epoch": 7.5,
5
+ "global_step": 240,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.03,
12
+ "learning_rate": 5.98125e-05,
13
+ "loss": 0.0848,
14
+ "step": 1
15
+ },
16
+ {
17
+ "epoch": 0.06,
18
+ "learning_rate": 5.9625000000000006e-05,
19
+ "loss": 0.1057,
20
+ "step": 2
21
+ },
22
+ {
23
+ "epoch": 0.09,
24
+ "learning_rate": 5.94375e-05,
25
+ "loss": 0.1762,
26
+ "step": 3
27
+ },
28
+ {
29
+ "epoch": 0.12,
30
+ "learning_rate": 5.9250000000000004e-05,
31
+ "loss": 0.1069,
32
+ "step": 4
33
+ },
34
+ {
35
+ "epoch": 0.16,
36
+ "learning_rate": 5.90625e-05,
37
+ "loss": 0.4235,
38
+ "step": 5
39
+ },
40
+ {
41
+ "epoch": 0.19,
42
+ "learning_rate": 5.8875e-05,
43
+ "loss": 0.1128,
44
+ "step": 6
45
+ },
46
+ {
47
+ "epoch": 0.22,
48
+ "learning_rate": 5.8687500000000003e-05,
49
+ "loss": 0.0893,
50
+ "step": 7
51
+ },
52
+ {
53
+ "epoch": 0.25,
54
+ "learning_rate": 5.85e-05,
55
+ "loss": 0.0531,
56
+ "step": 8
57
+ },
58
+ {
59
+ "epoch": 0.28,
60
+ "learning_rate": 5.83125e-05,
61
+ "loss": 0.0611,
62
+ "step": 9
63
+ },
64
+ {
65
+ "epoch": 0.31,
66
+ "learning_rate": 5.8125e-05,
67
+ "loss": 0.0709,
68
+ "step": 10
69
+ },
70
+ {
71
+ "epoch": 0.34,
72
+ "learning_rate": 5.79375e-05,
73
+ "loss": 0.2623,
74
+ "step": 11
75
+ },
76
+ {
77
+ "epoch": 0.38,
78
+ "learning_rate": 5.775e-05,
79
+ "loss": 0.1398,
80
+ "step": 12
81
+ },
82
+ {
83
+ "epoch": 0.41,
84
+ "learning_rate": 5.75625e-05,
85
+ "loss": 0.2857,
86
+ "step": 13
87
+ },
88
+ {
89
+ "epoch": 0.44,
90
+ "learning_rate": 5.7375000000000005e-05,
91
+ "loss": 0.0934,
92
+ "step": 14
93
+ },
94
+ {
95
+ "epoch": 0.47,
96
+ "learning_rate": 5.71875e-05,
97
+ "loss": 0.1849,
98
+ "step": 15
99
+ },
100
+ {
101
+ "epoch": 0.5,
102
+ "learning_rate": 5.6999999999999996e-05,
103
+ "loss": 0.0929,
104
+ "step": 16
105
+ },
106
+ {
107
+ "epoch": 0.53,
108
+ "learning_rate": 5.6812500000000005e-05,
109
+ "loss": 0.1765,
110
+ "step": 17
111
+ },
112
+ {
113
+ "epoch": 0.56,
114
+ "learning_rate": 5.6625e-05,
115
+ "loss": 0.1466,
116
+ "step": 18
117
+ },
118
+ {
119
+ "epoch": 0.59,
120
+ "learning_rate": 5.64375e-05,
121
+ "loss": 0.1918,
122
+ "step": 19
123
+ },
124
+ {
125
+ "epoch": 0.62,
126
+ "learning_rate": 5.625e-05,
127
+ "loss": 0.239,
128
+ "step": 20
129
+ },
130
+ {
131
+ "epoch": 0.62,
132
+ "eval_loss": 0.41225576400756836,
133
+ "eval_mean_accuracy": 0.7466867022012313,
134
+ "eval_mean_iou": 0.49318473255214207,
135
+ "eval_overall_accuracy": 0.8299140242615486,
136
+ "eval_per_category_accuracy": [
137
+ 0.9102603868063844,
138
+ NaN,
139
+ 0.5831130175960781
140
+ ],
141
+ "eval_per_category_iou": [
142
+ 0.9025356636572145,
143
+ 0.0,
144
+ 0.5770185339992117
145
+ ],
146
+ "eval_runtime": 20.0948,
147
+ "eval_samples_per_second": 0.796,
148
+ "eval_steps_per_second": 0.398,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.66,
153
+ "learning_rate": 5.60625e-05,
154
+ "loss": 0.0813,
155
+ "step": 21
156
+ },
157
+ {
158
+ "epoch": 0.69,
159
+ "learning_rate": 5.5875e-05,
160
+ "loss": 0.3204,
161
+ "step": 22
162
+ },
163
+ {
164
+ "epoch": 0.72,
165
+ "learning_rate": 5.56875e-05,
166
+ "loss": 0.1948,
167
+ "step": 23
168
+ },
169
+ {
170
+ "epoch": 0.75,
171
+ "learning_rate": 5.550000000000001e-05,
172
+ "loss": 0.3204,
173
+ "step": 24
174
+ },
175
+ {
176
+ "epoch": 0.78,
177
+ "learning_rate": 5.53125e-05,
178
+ "loss": 0.2348,
179
+ "step": 25
180
+ },
181
+ {
182
+ "epoch": 0.81,
183
+ "learning_rate": 5.5125e-05,
184
+ "loss": 0.1096,
185
+ "step": 26
186
+ },
187
+ {
188
+ "epoch": 0.84,
189
+ "learning_rate": 5.49375e-05,
190
+ "loss": 0.1375,
191
+ "step": 27
192
+ },
193
+ {
194
+ "epoch": 0.88,
195
+ "learning_rate": 5.475e-05,
196
+ "loss": 0.1218,
197
+ "step": 28
198
+ },
199
+ {
200
+ "epoch": 0.91,
201
+ "learning_rate": 5.4562500000000005e-05,
202
+ "loss": 0.2683,
203
+ "step": 29
204
+ },
205
+ {
206
+ "epoch": 0.94,
207
+ "learning_rate": 5.4375e-05,
208
+ "loss": 0.0548,
209
+ "step": 30
210
+ },
211
+ {
212
+ "epoch": 0.97,
213
+ "learning_rate": 5.4187499999999996e-05,
214
+ "loss": 0.1978,
215
+ "step": 31
216
+ },
217
+ {
218
+ "epoch": 1.0,
219
+ "learning_rate": 5.4000000000000005e-05,
220
+ "loss": 0.1099,
221
+ "step": 32
222
+ },
223
+ {
224
+ "epoch": 1.03,
225
+ "learning_rate": 5.38125e-05,
226
+ "loss": 0.1159,
227
+ "step": 33
228
+ },
229
+ {
230
+ "epoch": 1.06,
231
+ "learning_rate": 5.3625e-05,
232
+ "loss": 0.1443,
233
+ "step": 34
234
+ },
235
+ {
236
+ "epoch": 1.09,
237
+ "learning_rate": 5.3437500000000005e-05,
238
+ "loss": 0.065,
239
+ "step": 35
240
+ },
241
+ {
242
+ "epoch": 1.12,
243
+ "learning_rate": 5.325e-05,
244
+ "loss": 0.953,
245
+ "step": 36
246
+ },
247
+ {
248
+ "epoch": 1.16,
249
+ "learning_rate": 5.30625e-05,
250
+ "loss": 0.1242,
251
+ "step": 37
252
+ },
253
+ {
254
+ "epoch": 1.19,
255
+ "learning_rate": 5.2875e-05,
256
+ "loss": 0.1423,
257
+ "step": 38
258
+ },
259
+ {
260
+ "epoch": 1.22,
261
+ "learning_rate": 5.268750000000001e-05,
262
+ "loss": 0.3064,
263
+ "step": 39
264
+ },
265
+ {
266
+ "epoch": 1.25,
267
+ "learning_rate": 5.25e-05,
268
+ "loss": 0.0384,
269
+ "step": 40
270
+ },
271
+ {
272
+ "epoch": 1.25,
273
+ "eval_loss": 0.21261490881443024,
274
+ "eval_mean_accuracy": 0.9342305073015774,
275
+ "eval_mean_iou": 0.6157699255432437,
276
+ "eval_overall_accuracy": 0.9372036331047264,
277
+ "eval_per_category_accuracy": [
278
+ 0.9400738425234326,
279
+ NaN,
280
+ 0.9283871720797222
281
+ ],
282
+ "eval_per_category_iou": [
283
+ 0.9324786248273802,
284
+ 0.0,
285
+ 0.914831151802351
286
+ ],
287
+ "eval_runtime": 20.0231,
288
+ "eval_samples_per_second": 0.799,
289
+ "eval_steps_per_second": 0.4,
290
+ "step": 40
291
+ },
292
+ {
293
+ "epoch": 1.28,
294
+ "learning_rate": 5.23125e-05,
295
+ "loss": 0.1001,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 1.31,
300
+ "learning_rate": 5.2125e-05,
301
+ "loss": 0.0732,
302
+ "step": 42
303
+ },
304
+ {
305
+ "epoch": 1.34,
306
+ "learning_rate": 5.19375e-05,
307
+ "loss": 0.0736,
308
+ "step": 43
309
+ },
310
+ {
311
+ "epoch": 1.38,
312
+ "learning_rate": 5.1750000000000004e-05,
313
+ "loss": 0.145,
314
+ "step": 44
315
+ },
316
+ {
317
+ "epoch": 1.41,
318
+ "learning_rate": 5.15625e-05,
319
+ "loss": 0.0704,
320
+ "step": 45
321
+ },
322
+ {
323
+ "epoch": 1.44,
324
+ "learning_rate": 5.1375e-05,
325
+ "loss": 0.0602,
326
+ "step": 46
327
+ },
328
+ {
329
+ "epoch": 1.47,
330
+ "learning_rate": 5.1187500000000004e-05,
331
+ "loss": 0.2405,
332
+ "step": 47
333
+ },
334
+ {
335
+ "epoch": 1.5,
336
+ "learning_rate": 5.1e-05,
337
+ "loss": 0.0717,
338
+ "step": 48
339
+ },
340
+ {
341
+ "epoch": 1.53,
342
+ "learning_rate": 5.08125e-05,
343
+ "loss": 0.1311,
344
+ "step": 49
345
+ },
346
+ {
347
+ "epoch": 1.56,
348
+ "learning_rate": 5.0625000000000004e-05,
349
+ "loss": 0.1064,
350
+ "step": 50
351
+ },
352
+ {
353
+ "epoch": 1.59,
354
+ "learning_rate": 5.04375e-05,
355
+ "loss": 0.1129,
356
+ "step": 51
357
+ },
358
+ {
359
+ "epoch": 1.62,
360
+ "learning_rate": 5.025e-05,
361
+ "loss": 0.1423,
362
+ "step": 52
363
+ },
364
+ {
365
+ "epoch": 1.66,
366
+ "learning_rate": 5.00625e-05,
367
+ "loss": 0.2291,
368
+ "step": 53
369
+ },
370
+ {
371
+ "epoch": 1.69,
372
+ "learning_rate": 4.9875000000000006e-05,
373
+ "loss": 0.191,
374
+ "step": 54
375
+ },
376
+ {
377
+ "epoch": 1.72,
378
+ "learning_rate": 4.96875e-05,
379
+ "loss": 0.1337,
380
+ "step": 55
381
+ },
382
+ {
383
+ "epoch": 1.75,
384
+ "learning_rate": 4.95e-05,
385
+ "loss": 0.1408,
386
+ "step": 56
387
+ },
388
+ {
389
+ "epoch": 1.78,
390
+ "learning_rate": 4.93125e-05,
391
+ "loss": 0.1698,
392
+ "step": 57
393
+ },
394
+ {
395
+ "epoch": 1.81,
396
+ "learning_rate": 4.9125e-05,
397
+ "loss": 0.1266,
398
+ "step": 58
399
+ },
400
+ {
401
+ "epoch": 1.84,
402
+ "learning_rate": 4.8937500000000004e-05,
403
+ "loss": 0.1028,
404
+ "step": 59
405
+ },
406
+ {
407
+ "epoch": 1.88,
408
+ "learning_rate": 4.875e-05,
409
+ "loss": 0.1145,
410
+ "step": 60
411
+ },
412
+ {
413
+ "epoch": 1.88,
414
+ "eval_loss": 0.2426888346672058,
415
+ "eval_mean_accuracy": 0.8655657176277471,
416
+ "eval_mean_iou": 0.5680959283423686,
417
+ "eval_overall_accuracy": 0.8885364845945652,
418
+ "eval_per_category_accuracy": [
419
+ 0.9107121058324004,
420
+ NaN,
421
+ 0.8204193294230938
422
+ ],
423
+ "eval_per_category_iou": [
424
+ 0.9087610217218509,
425
+ 0.0,
426
+ 0.7955267633052547
427
+ ],
428
+ "eval_runtime": 20.2645,
429
+ "eval_samples_per_second": 0.79,
430
+ "eval_steps_per_second": 0.395,
431
+ "step": 60
432
+ },
433
+ {
434
+ "epoch": 1.91,
435
+ "learning_rate": 4.85625e-05,
436
+ "loss": 0.0419,
437
+ "step": 61
438
+ },
439
+ {
440
+ "epoch": 1.94,
441
+ "learning_rate": 4.8375000000000004e-05,
442
+ "loss": 0.157,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 1.97,
447
+ "learning_rate": 4.81875e-05,
448
+ "loss": 0.707,
449
+ "step": 63
450
+ },
451
+ {
452
+ "epoch": 2.0,
453
+ "learning_rate": 4.8e-05,
454
+ "loss": 0.0843,
455
+ "step": 64
456
+ },
457
+ {
458
+ "epoch": 2.03,
459
+ "learning_rate": 4.7812500000000003e-05,
460
+ "loss": 0.1084,
461
+ "step": 65
462
+ },
463
+ {
464
+ "epoch": 2.06,
465
+ "learning_rate": 4.7625e-05,
466
+ "loss": 0.2629,
467
+ "step": 66
468
+ },
469
+ {
470
+ "epoch": 2.09,
471
+ "learning_rate": 4.74375e-05,
472
+ "loss": 0.0796,
473
+ "step": 67
474
+ },
475
+ {
476
+ "epoch": 2.12,
477
+ "learning_rate": 4.7249999999999997e-05,
478
+ "loss": 0.1253,
479
+ "step": 68
480
+ },
481
+ {
482
+ "epoch": 2.16,
483
+ "learning_rate": 4.7062500000000006e-05,
484
+ "loss": 0.0744,
485
+ "step": 69
486
+ },
487
+ {
488
+ "epoch": 2.19,
489
+ "learning_rate": 4.6875e-05,
490
+ "loss": 0.0775,
491
+ "step": 70
492
+ },
493
+ {
494
+ "epoch": 2.22,
495
+ "learning_rate": 4.6687499999999996e-05,
496
+ "loss": 0.194,
497
+ "step": 71
498
+ },
499
+ {
500
+ "epoch": 2.25,
501
+ "learning_rate": 4.6500000000000005e-05,
502
+ "loss": 0.4576,
503
+ "step": 72
504
+ },
505
+ {
506
+ "epoch": 2.28,
507
+ "learning_rate": 4.63125e-05,
508
+ "loss": 0.0502,
509
+ "step": 73
510
+ },
511
+ {
512
+ "epoch": 2.31,
513
+ "learning_rate": 4.6125e-05,
514
+ "loss": 0.0402,
515
+ "step": 74
516
+ },
517
+ {
518
+ "epoch": 2.34,
519
+ "learning_rate": 4.59375e-05,
520
+ "loss": 0.4612,
521
+ "step": 75
522
+ },
523
+ {
524
+ "epoch": 2.38,
525
+ "learning_rate": 4.575e-05,
526
+ "loss": 0.0382,
527
+ "step": 76
528
+ },
529
+ {
530
+ "epoch": 2.41,
531
+ "learning_rate": 4.55625e-05,
532
+ "loss": 0.3068,
533
+ "step": 77
534
+ },
535
+ {
536
+ "epoch": 2.44,
537
+ "learning_rate": 4.5375e-05,
538
+ "loss": 0.2821,
539
+ "step": 78
540
+ },
541
+ {
542
+ "epoch": 2.47,
543
+ "learning_rate": 4.51875e-05,
544
+ "loss": 0.175,
545
+ "step": 79
546
+ },
547
+ {
548
+ "epoch": 2.5,
549
+ "learning_rate": 4.5e-05,
550
+ "loss": 0.163,
551
+ "step": 80
552
+ },
553
+ {
554
+ "epoch": 2.5,
555
+ "eval_loss": 0.22554153203964233,
556
+ "eval_mean_accuracy": 0.8740183132328878,
557
+ "eval_mean_iou": 0.5773767943990825,
558
+ "eval_overall_accuracy": 0.9256960064443283,
559
+ "eval_per_category_accuracy": [
560
+ 0.9755848478903867,
561
+ NaN,
562
+ 0.7724517785753888
563
+ ],
564
+ "eval_per_category_iou": [
565
+ 0.9662459147578681,
566
+ 0.0,
567
+ 0.7658844684393793
568
+ ],
569
+ "eval_runtime": 20.1088,
570
+ "eval_samples_per_second": 0.796,
571
+ "eval_steps_per_second": 0.398,
572
+ "step": 80
573
+ },
574
+ {
575
+ "epoch": 2.53,
576
+ "learning_rate": 4.48125e-05,
577
+ "loss": 0.2007,
578
+ "step": 81
579
+ },
580
+ {
581
+ "epoch": 2.56,
582
+ "learning_rate": 4.4625e-05,
583
+ "loss": 0.1948,
584
+ "step": 82
585
+ },
586
+ {
587
+ "epoch": 2.59,
588
+ "learning_rate": 4.44375e-05,
589
+ "loss": 0.2429,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 2.62,
594
+ "learning_rate": 4.4250000000000005e-05,
595
+ "loss": 0.0465,
596
+ "step": 84
597
+ },
598
+ {
599
+ "epoch": 2.66,
600
+ "learning_rate": 4.40625e-05,
601
+ "loss": 0.3231,
602
+ "step": 85
603
+ },
604
+ {
605
+ "epoch": 2.69,
606
+ "learning_rate": 4.3874999999999996e-05,
607
+ "loss": 0.1687,
608
+ "step": 86
609
+ },
610
+ {
611
+ "epoch": 2.72,
612
+ "learning_rate": 4.3687500000000005e-05,
613
+ "loss": 0.103,
614
+ "step": 87
615
+ },
616
+ {
617
+ "epoch": 2.75,
618
+ "learning_rate": 4.35e-05,
619
+ "loss": 0.2989,
620
+ "step": 88
621
+ },
622
+ {
623
+ "epoch": 2.78,
624
+ "learning_rate": 4.33125e-05,
625
+ "loss": 0.9179,
626
+ "step": 89
627
+ },
628
+ {
629
+ "epoch": 2.81,
630
+ "learning_rate": 4.3125e-05,
631
+ "loss": 0.072,
632
+ "step": 90
633
+ },
634
+ {
635
+ "epoch": 2.84,
636
+ "learning_rate": 4.29375e-05,
637
+ "loss": 0.3553,
638
+ "step": 91
639
+ },
640
+ {
641
+ "epoch": 2.88,
642
+ "learning_rate": 4.275e-05,
643
+ "loss": 0.0849,
644
+ "step": 92
645
+ },
646
+ {
647
+ "epoch": 2.91,
648
+ "learning_rate": 4.25625e-05,
649
+ "loss": 0.1466,
650
+ "step": 93
651
+ },
652
+ {
653
+ "epoch": 2.94,
654
+ "learning_rate": 4.237500000000001e-05,
655
+ "loss": 0.1367,
656
+ "step": 94
657
+ },
658
+ {
659
+ "epoch": 2.97,
660
+ "learning_rate": 4.21875e-05,
661
+ "loss": 0.3345,
662
+ "step": 95
663
+ },
664
+ {
665
+ "epoch": 3.0,
666
+ "learning_rate": 4.2e-05,
667
+ "loss": 0.0762,
668
+ "step": 96
669
+ },
670
+ {
671
+ "epoch": 3.03,
672
+ "learning_rate": 4.18125e-05,
673
+ "loss": 0.2443,
674
+ "step": 97
675
+ },
676
+ {
677
+ "epoch": 3.06,
678
+ "learning_rate": 4.1625e-05,
679
+ "loss": 0.2698,
680
+ "step": 98
681
+ },
682
+ {
683
+ "epoch": 3.09,
684
+ "learning_rate": 4.1437500000000004e-05,
685
+ "loss": 0.3754,
686
+ "step": 99
687
+ },
688
+ {
689
+ "epoch": 3.12,
690
+ "learning_rate": 4.125e-05,
691
+ "loss": 0.0832,
692
+ "step": 100
693
+ },
694
+ {
695
+ "epoch": 3.12,
696
+ "eval_loss": 0.18825271725654602,
697
+ "eval_mean_accuracy": 0.9523870826679519,
698
+ "eval_mean_iou": 0.6270557614254478,
699
+ "eval_overall_accuracy": 0.9553877902482951,
700
+ "eval_per_category_accuracy": [
701
+ 0.958284626685812,
702
+ NaN,
703
+ 0.9464895386500919
704
+ ],
705
+ "eval_per_category_iou": [
706
+ 0.9550280003976388,
707
+ 0.0,
708
+ 0.9261392838787048
709
+ ],
710
+ "eval_runtime": 19.8632,
711
+ "eval_samples_per_second": 0.806,
712
+ "eval_steps_per_second": 0.403,
713
+ "step": 100
714
+ },
715
+ {
716
+ "epoch": 3.16,
717
+ "learning_rate": 4.1062499999999995e-05,
718
+ "loss": 0.0569,
719
+ "step": 101
720
+ },
721
+ {
722
+ "epoch": 3.19,
723
+ "learning_rate": 4.0875000000000004e-05,
724
+ "loss": 0.3998,
725
+ "step": 102
726
+ },
727
+ {
728
+ "epoch": 3.22,
729
+ "learning_rate": 4.06875e-05,
730
+ "loss": 0.1192,
731
+ "step": 103
732
+ },
733
+ {
734
+ "epoch": 3.25,
735
+ "learning_rate": 4.05e-05,
736
+ "loss": 0.0834,
737
+ "step": 104
738
+ },
739
+ {
740
+ "epoch": 3.28,
741
+ "learning_rate": 4.0312500000000004e-05,
742
+ "loss": 0.1567,
743
+ "step": 105
744
+ },
745
+ {
746
+ "epoch": 3.31,
747
+ "learning_rate": 4.0125e-05,
748
+ "loss": 0.1221,
749
+ "step": 106
750
+ },
751
+ {
752
+ "epoch": 3.34,
753
+ "learning_rate": 3.99375e-05,
754
+ "loss": 0.1699,
755
+ "step": 107
756
+ },
757
+ {
758
+ "epoch": 3.38,
759
+ "learning_rate": 3.975e-05,
760
+ "loss": 0.0803,
761
+ "step": 108
762
+ },
763
+ {
764
+ "epoch": 3.41,
765
+ "learning_rate": 3.9562500000000006e-05,
766
+ "loss": 0.1208,
767
+ "step": 109
768
+ },
769
+ {
770
+ "epoch": 3.44,
771
+ "learning_rate": 3.9375e-05,
772
+ "loss": 0.1283,
773
+ "step": 110
774
+ },
775
+ {
776
+ "epoch": 3.47,
777
+ "learning_rate": 3.91875e-05,
778
+ "loss": 0.4262,
779
+ "step": 111
780
+ },
781
+ {
782
+ "epoch": 3.5,
783
+ "learning_rate": 3.9e-05,
784
+ "loss": 0.0729,
785
+ "step": 112
786
+ },
787
+ {
788
+ "epoch": 3.53,
789
+ "learning_rate": 3.88125e-05,
790
+ "loss": 0.1058,
791
+ "step": 113
792
+ },
793
+ {
794
+ "epoch": 3.56,
795
+ "learning_rate": 3.8625000000000004e-05,
796
+ "loss": 0.1903,
797
+ "step": 114
798
+ },
799
+ {
800
+ "epoch": 3.59,
801
+ "learning_rate": 3.84375e-05,
802
+ "loss": 0.2635,
803
+ "step": 115
804
+ },
805
+ {
806
+ "epoch": 3.62,
807
+ "learning_rate": 3.825e-05,
808
+ "loss": 0.0744,
809
+ "step": 116
810
+ },
811
+ {
812
+ "epoch": 3.66,
813
+ "learning_rate": 3.8062500000000004e-05,
814
+ "loss": 0.2714,
815
+ "step": 117
816
+ },
817
+ {
818
+ "epoch": 3.69,
819
+ "learning_rate": 3.7875e-05,
820
+ "loss": 0.1265,
821
+ "step": 118
822
+ },
823
+ {
824
+ "epoch": 3.72,
825
+ "learning_rate": 3.76875e-05,
826
+ "loss": 0.2965,
827
+ "step": 119
828
+ },
829
+ {
830
+ "epoch": 3.75,
831
+ "learning_rate": 3.7500000000000003e-05,
832
+ "loss": 0.047,
833
+ "step": 120
834
+ },
835
+ {
836
+ "epoch": 3.75,
837
+ "eval_loss": 0.19484524428844452,
838
+ "eval_mean_accuracy": 0.9191032313533529,
839
+ "eval_mean_iou": 0.605987572284888,
840
+ "eval_overall_accuracy": 0.9443237844948401,
841
+ "eval_per_category_accuracy": [
842
+ 0.9686713143218681,
843
+ NaN,
844
+ 0.8695351483848376
845
+ ],
846
+ "eval_per_category_iou": [
847
+ 0.9642150125223343,
848
+ 0.0,
849
+ 0.8537477043323296
850
+ ],
851
+ "eval_runtime": 21.0542,
852
+ "eval_samples_per_second": 0.76,
853
+ "eval_steps_per_second": 0.38,
854
+ "step": 120
855
+ },
856
+ {
857
+ "epoch": 3.78,
858
+ "learning_rate": 3.73125e-05,
859
+ "loss": 0.2113,
860
+ "step": 121
861
+ },
862
+ {
863
+ "epoch": 3.81,
864
+ "learning_rate": 3.7125e-05,
865
+ "loss": 0.1938,
866
+ "step": 122
867
+ },
868
+ {
869
+ "epoch": 3.84,
870
+ "learning_rate": 3.6937499999999997e-05,
871
+ "loss": 0.3208,
872
+ "step": 123
873
+ },
874
+ {
875
+ "epoch": 3.88,
876
+ "learning_rate": 3.6750000000000006e-05,
877
+ "loss": 0.0783,
878
+ "step": 124
879
+ },
880
+ {
881
+ "epoch": 3.91,
882
+ "learning_rate": 3.65625e-05,
883
+ "loss": 0.1378,
884
+ "step": 125
885
+ },
886
+ {
887
+ "epoch": 3.94,
888
+ "learning_rate": 3.6374999999999996e-05,
889
+ "loss": 0.1342,
890
+ "step": 126
891
+ },
892
+ {
893
+ "epoch": 3.97,
894
+ "learning_rate": 3.6187500000000005e-05,
895
+ "loss": 0.0532,
896
+ "step": 127
897
+ },
898
+ {
899
+ "epoch": 4.0,
900
+ "learning_rate": 3.6e-05,
901
+ "loss": 0.2819,
902
+ "step": 128
903
+ },
904
+ {
905
+ "epoch": 4.03,
906
+ "learning_rate": 3.58125e-05,
907
+ "loss": 0.1006,
908
+ "step": 129
909
+ },
910
+ {
911
+ "epoch": 4.06,
912
+ "learning_rate": 3.5625e-05,
913
+ "loss": 0.1431,
914
+ "step": 130
915
+ },
916
+ {
917
+ "epoch": 4.09,
918
+ "learning_rate": 3.54375e-05,
919
+ "loss": 0.197,
920
+ "step": 131
921
+ },
922
+ {
923
+ "epoch": 4.12,
924
+ "learning_rate": 3.525e-05,
925
+ "loss": 0.1781,
926
+ "step": 132
927
+ },
928
+ {
929
+ "epoch": 4.16,
930
+ "learning_rate": 3.50625e-05,
931
+ "loss": 0.17,
932
+ "step": 133
933
+ },
934
+ {
935
+ "epoch": 4.19,
936
+ "learning_rate": 3.4875e-05,
937
+ "loss": 0.2306,
938
+ "step": 134
939
+ },
940
+ {
941
+ "epoch": 4.22,
942
+ "learning_rate": 3.46875e-05,
943
+ "loss": 0.1139,
944
+ "step": 135
945
+ },
946
+ {
947
+ "epoch": 4.25,
948
+ "learning_rate": 3.45e-05,
949
+ "loss": 0.0544,
950
+ "step": 136
951
+ },
952
+ {
953
+ "epoch": 4.28,
954
+ "learning_rate": 3.43125e-05,
955
+ "loss": 0.1489,
956
+ "step": 137
957
+ },
958
+ {
959
+ "epoch": 4.31,
960
+ "learning_rate": 3.4125e-05,
961
+ "loss": 0.0706,
962
+ "step": 138
963
+ },
964
+ {
965
+ "epoch": 4.34,
966
+ "learning_rate": 3.3937500000000005e-05,
967
+ "loss": 0.0953,
968
+ "step": 139
969
+ },
970
+ {
971
+ "epoch": 4.38,
972
+ "learning_rate": 3.375e-05,
973
+ "loss": 0.1989,
974
+ "step": 140
975
+ },
976
+ {
977
+ "epoch": 4.38,
978
+ "eval_loss": 0.18885326385498047,
979
+ "eval_mean_accuracy": 0.9465842428924998,
980
+ "eval_mean_iou": 0.6241060965774974,
981
+ "eval_overall_accuracy": 0.9592237852473473,
982
+ "eval_per_category_accuracy": [
983
+ 0.9714258028937567,
984
+ NaN,
985
+ 0.9217426828912428
986
+ ],
987
+ "eval_per_category_iou": [
988
+ 0.9659579975137067,
989
+ 0.0,
990
+ 0.9063602922187853
991
+ ],
992
+ "eval_runtime": 19.9313,
993
+ "eval_samples_per_second": 0.803,
994
+ "eval_steps_per_second": 0.401,
995
+ "step": 140
996
+ },
997
+ {
998
+ "epoch": 4.41,
999
+ "learning_rate": 3.3562499999999996e-05,
1000
+ "loss": 0.0458,
1001
+ "step": 141
1002
+ },
1003
+ {
1004
+ "epoch": 4.44,
1005
+ "learning_rate": 3.3375000000000005e-05,
1006
+ "loss": 0.2325,
1007
+ "step": 142
1008
+ },
1009
+ {
1010
+ "epoch": 4.47,
1011
+ "learning_rate": 3.31875e-05,
1012
+ "loss": 0.2194,
1013
+ "step": 143
1014
+ },
1015
+ {
1016
+ "epoch": 4.5,
1017
+ "learning_rate": 3.3e-05,
1018
+ "loss": 0.0531,
1019
+ "step": 144
1020
+ },
1021
+ {
1022
+ "epoch": 4.53,
1023
+ "learning_rate": 3.28125e-05,
1024
+ "loss": 0.0904,
1025
+ "step": 145
1026
+ },
1027
+ {
1028
+ "epoch": 4.56,
1029
+ "learning_rate": 3.2625e-05,
1030
+ "loss": 0.1224,
1031
+ "step": 146
1032
+ },
1033
+ {
1034
+ "epoch": 4.59,
1035
+ "learning_rate": 3.24375e-05,
1036
+ "loss": 0.1236,
1037
+ "step": 147
1038
+ },
1039
+ {
1040
+ "epoch": 4.62,
1041
+ "learning_rate": 3.225e-05,
1042
+ "loss": 0.0396,
1043
+ "step": 148
1044
+ },
1045
+ {
1046
+ "epoch": 4.66,
1047
+ "learning_rate": 3.206250000000001e-05,
1048
+ "loss": 0.0329,
1049
+ "step": 149
1050
+ },
1051
+ {
1052
+ "epoch": 4.69,
1053
+ "learning_rate": 3.1875e-05,
1054
+ "loss": 0.0514,
1055
+ "step": 150
1056
+ },
1057
+ {
1058
+ "epoch": 4.72,
1059
+ "learning_rate": 3.16875e-05,
1060
+ "loss": 0.7069,
1061
+ "step": 151
1062
+ },
1063
+ {
1064
+ "epoch": 4.75,
1065
+ "learning_rate": 3.15e-05,
1066
+ "loss": 0.1664,
1067
+ "step": 152
1068
+ },
1069
+ {
1070
+ "epoch": 4.78,
1071
+ "learning_rate": 3.13125e-05,
1072
+ "loss": 0.1071,
1073
+ "step": 153
1074
+ },
1075
+ {
1076
+ "epoch": 4.81,
1077
+ "learning_rate": 3.1125000000000004e-05,
1078
+ "loss": 0.0525,
1079
+ "step": 154
1080
+ },
1081
+ {
1082
+ "epoch": 4.84,
1083
+ "learning_rate": 3.09375e-05,
1084
+ "loss": 0.1084,
1085
+ "step": 155
1086
+ },
1087
+ {
1088
+ "epoch": 4.88,
1089
+ "learning_rate": 3.0749999999999995e-05,
1090
+ "loss": 0.3014,
1091
+ "step": 156
1092
+ },
1093
+ {
1094
+ "epoch": 4.91,
1095
+ "learning_rate": 3.0562500000000004e-05,
1096
+ "loss": 0.0506,
1097
+ "step": 157
1098
+ },
1099
+ {
1100
+ "epoch": 4.94,
1101
+ "learning_rate": 3.0375e-05,
1102
+ "loss": 0.1682,
1103
+ "step": 158
1104
+ },
1105
+ {
1106
+ "epoch": 4.97,
1107
+ "learning_rate": 3.0187500000000002e-05,
1108
+ "loss": 0.1597,
1109
+ "step": 159
1110
+ },
1111
+ {
1112
+ "epoch": 5.0,
1113
+ "learning_rate": 3e-05,
1114
+ "loss": 0.1113,
1115
+ "step": 160
1116
+ },
1117
+ {
1118
+ "epoch": 5.0,
1119
+ "eval_loss": 0.28806212544441223,
1120
+ "eval_mean_accuracy": 0.8278887930983827,
1121
+ "eval_mean_iou": 0.5469872417793716,
1122
+ "eval_overall_accuracy": 0.8720594437681833,
1123
+ "eval_per_category_accuracy": [
1124
+ 0.9147011030779117,
1125
+ NaN,
1126
+ 0.7410764831188538
1127
+ ],
1128
+ "eval_per_category_iou": [
1129
+ 0.9109685952665123,
1130
+ 0.0,
1131
+ 0.7299931300716025
1132
+ ],
1133
+ "eval_runtime": 20.3787,
1134
+ "eval_samples_per_second": 0.785,
1135
+ "eval_steps_per_second": 0.393,
1136
+ "step": 160
1137
+ },
1138
+ {
1139
+ "epoch": 5.03,
1140
+ "learning_rate": 2.9812500000000003e-05,
1141
+ "loss": 0.0783,
1142
+ "step": 161
1143
+ },
1144
+ {
1145
+ "epoch": 5.06,
1146
+ "learning_rate": 2.9625000000000002e-05,
1147
+ "loss": 0.0648,
1148
+ "step": 162
1149
+ },
1150
+ {
1151
+ "epoch": 5.09,
1152
+ "learning_rate": 2.94375e-05,
1153
+ "loss": 0.0512,
1154
+ "step": 163
1155
+ },
1156
+ {
1157
+ "epoch": 5.12,
1158
+ "learning_rate": 2.925e-05,
1159
+ "loss": 0.1477,
1160
+ "step": 164
1161
+ },
1162
+ {
1163
+ "epoch": 5.16,
1164
+ "learning_rate": 2.90625e-05,
1165
+ "loss": 0.1665,
1166
+ "step": 165
1167
+ },
1168
+ {
1169
+ "epoch": 5.19,
1170
+ "learning_rate": 2.8875e-05,
1171
+ "loss": 0.1202,
1172
+ "step": 166
1173
+ },
1174
+ {
1175
+ "epoch": 5.22,
1176
+ "learning_rate": 2.8687500000000003e-05,
1177
+ "loss": 0.0873,
1178
+ "step": 167
1179
+ },
1180
+ {
1181
+ "epoch": 5.25,
1182
+ "learning_rate": 2.8499999999999998e-05,
1183
+ "loss": 0.0529,
1184
+ "step": 168
1185
+ },
1186
+ {
1187
+ "epoch": 5.28,
1188
+ "learning_rate": 2.83125e-05,
1189
+ "loss": 0.0848,
1190
+ "step": 169
1191
+ },
1192
+ {
1193
+ "epoch": 5.31,
1194
+ "learning_rate": 2.8125e-05,
1195
+ "loss": 0.1113,
1196
+ "step": 170
1197
+ },
1198
+ {
1199
+ "epoch": 5.34,
1200
+ "learning_rate": 2.79375e-05,
1201
+ "loss": 0.1652,
1202
+ "step": 171
1203
+ },
1204
+ {
1205
+ "epoch": 5.38,
1206
+ "learning_rate": 2.7750000000000004e-05,
1207
+ "loss": 0.0554,
1208
+ "step": 172
1209
+ },
1210
+ {
1211
+ "epoch": 5.41,
1212
+ "learning_rate": 2.75625e-05,
1213
+ "loss": 0.2955,
1214
+ "step": 173
1215
+ },
1216
+ {
1217
+ "epoch": 5.44,
1218
+ "learning_rate": 2.7375e-05,
1219
+ "loss": 0.1975,
1220
+ "step": 174
1221
+ },
1222
+ {
1223
+ "epoch": 5.47,
1224
+ "learning_rate": 2.71875e-05,
1225
+ "loss": 0.2093,
1226
+ "step": 175
1227
+ },
1228
+ {
1229
+ "epoch": 5.5,
1230
+ "learning_rate": 2.7000000000000002e-05,
1231
+ "loss": 0.0438,
1232
+ "step": 176
1233
+ },
1234
+ {
1235
+ "epoch": 5.53,
1236
+ "learning_rate": 2.68125e-05,
1237
+ "loss": 0.0997,
1238
+ "step": 177
1239
+ },
1240
+ {
1241
+ "epoch": 5.56,
1242
+ "learning_rate": 2.6625e-05,
1243
+ "loss": 0.0872,
1244
+ "step": 178
1245
+ },
1246
+ {
1247
+ "epoch": 5.59,
1248
+ "learning_rate": 2.64375e-05,
1249
+ "loss": 0.1928,
1250
+ "step": 179
1251
+ },
1252
+ {
1253
+ "epoch": 5.62,
1254
+ "learning_rate": 2.625e-05,
1255
+ "loss": 0.0877,
1256
+ "step": 180
1257
+ },
1258
+ {
1259
+ "epoch": 5.62,
1260
+ "eval_loss": 0.20857231318950653,
1261
+ "eval_mean_accuracy": 0.9132418593317307,
1262
+ "eval_mean_iou": 0.5999547882669239,
1263
+ "eval_overall_accuracy": 0.9410679367076951,
1264
+ "eval_per_category_accuracy": [
1265
+ 0.9679307991151817,
1266
+ NaN,
1267
+ 0.8585529195482798
1268
+ ],
1269
+ "eval_per_category_iou": [
1270
+ 0.9588284256593376,
1271
+ 0.0,
1272
+ 0.841035939141434
1273
+ ],
1274
+ "eval_runtime": 20.2448,
1275
+ "eval_samples_per_second": 0.79,
1276
+ "eval_steps_per_second": 0.395,
1277
+ "step": 180
1278
+ },
1279
+ {
1280
+ "epoch": 5.66,
1281
+ "learning_rate": 2.60625e-05,
1282
+ "loss": 0.5519,
1283
+ "step": 181
1284
+ },
1285
+ {
1286
+ "epoch": 5.69,
1287
+ "learning_rate": 2.5875000000000002e-05,
1288
+ "loss": 0.0631,
1289
+ "step": 182
1290
+ },
1291
+ {
1292
+ "epoch": 5.72,
1293
+ "learning_rate": 2.56875e-05,
1294
+ "loss": 0.1026,
1295
+ "step": 183
1296
+ },
1297
+ {
1298
+ "epoch": 5.75,
1299
+ "learning_rate": 2.55e-05,
1300
+ "loss": 0.1227,
1301
+ "step": 184
1302
+ },
1303
+ {
1304
+ "epoch": 5.78,
1305
+ "learning_rate": 2.5312500000000002e-05,
1306
+ "loss": 0.2057,
1307
+ "step": 185
1308
+ },
1309
+ {
1310
+ "epoch": 5.81,
1311
+ "learning_rate": 2.5125e-05,
1312
+ "loss": 0.1111,
1313
+ "step": 186
1314
+ },
1315
+ {
1316
+ "epoch": 5.84,
1317
+ "learning_rate": 2.4937500000000003e-05,
1318
+ "loss": 0.1093,
1319
+ "step": 187
1320
+ },
1321
+ {
1322
+ "epoch": 5.88,
1323
+ "learning_rate": 2.475e-05,
1324
+ "loss": 0.2486,
1325
+ "step": 188
1326
+ },
1327
+ {
1328
+ "epoch": 5.91,
1329
+ "learning_rate": 2.45625e-05,
1330
+ "loss": 0.3687,
1331
+ "step": 189
1332
+ },
1333
+ {
1334
+ "epoch": 5.94,
1335
+ "learning_rate": 2.4375e-05,
1336
+ "loss": 0.1365,
1337
+ "step": 190
1338
+ },
1339
+ {
1340
+ "epoch": 5.97,
1341
+ "learning_rate": 2.4187500000000002e-05,
1342
+ "loss": 0.1974,
1343
+ "step": 191
1344
+ },
1345
+ {
1346
+ "epoch": 6.0,
1347
+ "learning_rate": 2.4e-05,
1348
+ "loss": 0.0645,
1349
+ "step": 192
1350
+ },
1351
+ {
1352
+ "epoch": 6.03,
1353
+ "learning_rate": 2.38125e-05,
1354
+ "loss": 0.1426,
1355
+ "step": 193
1356
+ },
1357
+ {
1358
+ "epoch": 6.06,
1359
+ "learning_rate": 2.3624999999999998e-05,
1360
+ "loss": 0.1137,
1361
+ "step": 194
1362
+ },
1363
+ {
1364
+ "epoch": 6.09,
1365
+ "learning_rate": 2.34375e-05,
1366
+ "loss": 0.072,
1367
+ "step": 195
1368
+ },
1369
+ {
1370
+ "epoch": 6.12,
1371
+ "learning_rate": 2.3250000000000003e-05,
1372
+ "loss": 0.0737,
1373
+ "step": 196
1374
+ },
1375
+ {
1376
+ "epoch": 6.16,
1377
+ "learning_rate": 2.30625e-05,
1378
+ "loss": 0.1085,
1379
+ "step": 197
1380
+ },
1381
+ {
1382
+ "epoch": 6.19,
1383
+ "learning_rate": 2.2875e-05,
1384
+ "loss": 0.0563,
1385
+ "step": 198
1386
+ },
1387
+ {
1388
+ "epoch": 6.22,
1389
+ "learning_rate": 2.26875e-05,
1390
+ "loss": 0.0642,
1391
+ "step": 199
1392
+ },
1393
+ {
1394
+ "epoch": 6.25,
1395
+ "learning_rate": 2.25e-05,
1396
+ "loss": 0.0688,
1397
+ "step": 200
1398
+ },
1399
+ {
1400
+ "epoch": 6.25,
1401
+ "eval_loss": 0.2616099715232849,
1402
+ "eval_mean_accuracy": 0.881410709224524,
1403
+ "eval_mean_iou": 0.5805529011665603,
1404
+ "eval_overall_accuracy": 0.8979442578505313,
1405
+ "eval_per_category_accuracy": [
1406
+ 0.9139054885999133,
1407
+ NaN,
1408
+ 0.8489159298491348
1409
+ ],
1410
+ "eval_per_category_iou": [
1411
+ 0.9093434184699977,
1412
+ 0.0,
1413
+ 0.8323152850296831
1414
+ ],
1415
+ "eval_runtime": 20.1245,
1416
+ "eval_samples_per_second": 0.795,
1417
+ "eval_steps_per_second": 0.398,
1418
+ "step": 200
1419
+ },
1420
+ {
1421
+ "epoch": 6.28,
1422
+ "learning_rate": 2.23125e-05,
1423
+ "loss": 0.0385,
1424
+ "step": 201
1425
+ },
1426
+ {
1427
+ "epoch": 6.31,
1428
+ "learning_rate": 2.2125000000000002e-05,
1429
+ "loss": 0.0416,
1430
+ "step": 202
1431
+ },
1432
+ {
1433
+ "epoch": 6.34,
1434
+ "learning_rate": 2.1937499999999998e-05,
1435
+ "loss": 0.035,
1436
+ "step": 203
1437
+ },
1438
+ {
1439
+ "epoch": 6.38,
1440
+ "learning_rate": 2.175e-05,
1441
+ "loss": 0.1747,
1442
+ "step": 204
1443
+ },
1444
+ {
1445
+ "epoch": 6.41,
1446
+ "learning_rate": 2.15625e-05,
1447
+ "loss": 0.1221,
1448
+ "step": 205
1449
+ },
1450
+ {
1451
+ "epoch": 6.44,
1452
+ "learning_rate": 2.1375e-05,
1453
+ "loss": 0.1319,
1454
+ "step": 206
1455
+ },
1456
+ {
1457
+ "epoch": 6.47,
1458
+ "learning_rate": 2.1187500000000003e-05,
1459
+ "loss": 0.1493,
1460
+ "step": 207
1461
+ },
1462
+ {
1463
+ "epoch": 6.5,
1464
+ "learning_rate": 2.1e-05,
1465
+ "loss": 0.1611,
1466
+ "step": 208
1467
+ },
1468
+ {
1469
+ "epoch": 6.53,
1470
+ "learning_rate": 2.08125e-05,
1471
+ "loss": 0.1956,
1472
+ "step": 209
1473
+ },
1474
+ {
1475
+ "epoch": 6.56,
1476
+ "learning_rate": 2.0625e-05,
1477
+ "loss": 0.1174,
1478
+ "step": 210
1479
+ },
1480
+ {
1481
+ "epoch": 6.59,
1482
+ "learning_rate": 2.0437500000000002e-05,
1483
+ "loss": 0.0776,
1484
+ "step": 211
1485
+ },
1486
+ {
1487
+ "epoch": 6.62,
1488
+ "learning_rate": 2.025e-05,
1489
+ "loss": 0.2234,
1490
+ "step": 212
1491
+ },
1492
+ {
1493
+ "epoch": 6.66,
1494
+ "learning_rate": 2.00625e-05,
1495
+ "loss": 0.1132,
1496
+ "step": 213
1497
+ },
1498
+ {
1499
+ "epoch": 6.69,
1500
+ "learning_rate": 1.9875e-05,
1501
+ "loss": 0.1123,
1502
+ "step": 214
1503
+ },
1504
+ {
1505
+ "epoch": 6.72,
1506
+ "learning_rate": 1.96875e-05,
1507
+ "loss": 0.078,
1508
+ "step": 215
1509
+ },
1510
+ {
1511
+ "epoch": 6.75,
1512
+ "learning_rate": 1.95e-05,
1513
+ "loss": 0.1747,
1514
+ "step": 216
1515
+ },
1516
+ {
1517
+ "epoch": 6.78,
1518
+ "learning_rate": 1.9312500000000002e-05,
1519
+ "loss": 0.2602,
1520
+ "step": 217
1521
+ },
1522
+ {
1523
+ "epoch": 6.81,
1524
+ "learning_rate": 1.9125e-05,
1525
+ "loss": 0.1234,
1526
+ "step": 218
1527
+ },
1528
+ {
1529
+ "epoch": 6.84,
1530
+ "learning_rate": 1.89375e-05,
1531
+ "loss": 0.2718,
1532
+ "step": 219
1533
+ },
1534
+ {
1535
+ "epoch": 6.88,
1536
+ "learning_rate": 1.8750000000000002e-05,
1537
+ "loss": 0.0785,
1538
+ "step": 220
1539
+ },
1540
+ {
1541
+ "epoch": 6.88,
1542
+ "eval_loss": 0.18271270394325256,
1543
+ "eval_mean_accuracy": 0.950766931450973,
1544
+ "eval_mean_iou": 0.6214100814010263,
1545
+ "eval_overall_accuracy": 0.9585103367613023,
1546
+ "eval_per_category_accuracy": [
1547
+ 0.9659856998391196,
1548
+ NaN,
1549
+ 0.9355481630628265
1550
+ ],
1551
+ "eval_per_category_iou": [
1552
+ 0.949549266878738,
1553
+ 0.0,
1554
+ 0.9146809773243406
1555
+ ],
1556
+ "eval_runtime": 21.0499,
1557
+ "eval_samples_per_second": 0.76,
1558
+ "eval_steps_per_second": 0.38,
1559
+ "step": 220
1560
+ },
1561
+ {
1562
+ "epoch": 6.91,
1563
+ "learning_rate": 1.85625e-05,
1564
+ "loss": 0.0806,
1565
+ "step": 221
1566
+ },
1567
+ {
1568
+ "epoch": 6.94,
1569
+ "learning_rate": 1.8375000000000003e-05,
1570
+ "loss": 0.0847,
1571
+ "step": 222
1572
+ },
1573
+ {
1574
+ "epoch": 6.97,
1575
+ "learning_rate": 1.8187499999999998e-05,
1576
+ "loss": 0.259,
1577
+ "step": 223
1578
+ },
1579
+ {
1580
+ "epoch": 7.0,
1581
+ "learning_rate": 1.8e-05,
1582
+ "loss": 0.0936,
1583
+ "step": 224
1584
+ },
1585
+ {
1586
+ "epoch": 7.03,
1587
+ "learning_rate": 1.78125e-05,
1588
+ "loss": 0.0977,
1589
+ "step": 225
1590
+ },
1591
+ {
1592
+ "epoch": 7.06,
1593
+ "learning_rate": 1.7625e-05,
1594
+ "loss": 0.0597,
1595
+ "step": 226
1596
+ },
1597
+ {
1598
+ "epoch": 7.09,
1599
+ "learning_rate": 1.74375e-05,
1600
+ "loss": 0.0843,
1601
+ "step": 227
1602
+ },
1603
+ {
1604
+ "epoch": 7.12,
1605
+ "learning_rate": 1.725e-05,
1606
+ "loss": 0.0709,
1607
+ "step": 228
1608
+ },
1609
+ {
1610
+ "epoch": 7.16,
1611
+ "learning_rate": 1.70625e-05,
1612
+ "loss": 0.0542,
1613
+ "step": 229
1614
+ },
1615
+ {
1616
+ "epoch": 7.19,
1617
+ "learning_rate": 1.6875e-05,
1618
+ "loss": 0.2108,
1619
+ "step": 230
1620
+ },
1621
+ {
1622
+ "epoch": 7.22,
1623
+ "learning_rate": 1.6687500000000002e-05,
1624
+ "loss": 0.0454,
1625
+ "step": 231
1626
+ },
1627
+ {
1628
+ "epoch": 7.25,
1629
+ "learning_rate": 1.65e-05,
1630
+ "loss": 0.1047,
1631
+ "step": 232
1632
+ },
1633
+ {
1634
+ "epoch": 7.28,
1635
+ "learning_rate": 1.63125e-05,
1636
+ "loss": 0.03,
1637
+ "step": 233
1638
+ },
1639
+ {
1640
+ "epoch": 7.31,
1641
+ "learning_rate": 1.6125e-05,
1642
+ "loss": 0.204,
1643
+ "step": 234
1644
+ },
1645
+ {
1646
+ "epoch": 7.34,
1647
+ "learning_rate": 1.59375e-05,
1648
+ "loss": 0.1506,
1649
+ "step": 235
1650
+ },
1651
+ {
1652
+ "epoch": 7.38,
1653
+ "learning_rate": 1.575e-05,
1654
+ "loss": 0.07,
1655
+ "step": 236
1656
+ },
1657
+ {
1658
+ "epoch": 7.41,
1659
+ "learning_rate": 1.5562500000000002e-05,
1660
+ "loss": 0.0615,
1661
+ "step": 237
1662
+ },
1663
+ {
1664
+ "epoch": 7.44,
1665
+ "learning_rate": 1.5374999999999998e-05,
1666
+ "loss": 0.1664,
1667
+ "step": 238
1668
+ },
1669
+ {
1670
+ "epoch": 7.47,
1671
+ "learning_rate": 1.51875e-05,
1672
+ "loss": 0.0805,
1673
+ "step": 239
1674
+ },
1675
+ {
1676
+ "epoch": 7.5,
1677
+ "learning_rate": 1.5e-05,
1678
+ "loss": 0.1514,
1679
+ "step": 240
1680
+ },
1681
+ {
1682
+ "epoch": 7.5,
1683
+ "eval_loss": 0.15978053212165833,
1684
+ "eval_mean_accuracy": 0.9622730849481105,
1685
+ "eval_mean_iou": 0.6345740505797159,
1686
+ "eval_overall_accuracy": 0.9649951570787098,
1687
+ "eval_per_category_accuracy": [
1688
+ 0.9676230031857829,
1689
+ NaN,
1690
+ 0.956923166710438
1691
+ ],
1692
+ "eval_per_category_iou": [
1693
+ 0.9619821347955928,
1694
+ 0.0,
1695
+ 0.941740016943555
1696
+ ],
1697
+ "eval_runtime": 19.855,
1698
+ "eval_samples_per_second": 0.806,
1699
+ "eval_steps_per_second": 0.403,
1700
+ "step": 240
1701
+ }
1702
+ ],
1703
+ "max_steps": 320,
1704
+ "num_train_epochs": 10,
1705
+ "total_flos": 8414000735846400.0,
1706
+ "trial_name": null,
1707
+ "trial_params": null
1708
+ }