rishitunu commited on
Commit
00f074e
1 Parent(s): 1798bf1

End of training

Browse files
Files changed (5) hide show
  1. README.md +3 -1
  2. all_results.json +18 -0
  3. eval_results.json +14 -0
  4. train_results.json +7 -0
  5. trainer_state.json +973 -0
README.md CHANGED
@@ -2,6 +2,8 @@
2
  license: other
3
  base_model: nvidia/mit-b5
4
  tags:
 
 
5
  - generated_from_trainer
6
  model-index:
7
  - name: new_ecc_segformer
@@ -13,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
13
 
14
  # new_ecc_segformer
15
 
16
- This model is a fine-tuned version of [nvidia/mit-b5](https://huggingface.co/nvidia/mit-b5) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
  - Loss: 0.0663
19
  - Mean Iou: 0.1943
 
2
  license: other
3
  base_model: nvidia/mit-b5
4
  tags:
5
+ - image-segmentation
6
+ - vision
7
  - generated_from_trainer
8
  model-index:
9
  - name: new_ecc_segformer
 
15
 
16
  # new_ecc_segformer
17
 
18
+ This model is a fine-tuned version of [nvidia/mit-b5](https://huggingface.co/nvidia/mit-b5) on the rishitunu/ECC_crackdataset_withsplit dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.0663
21
  - Mean Iou: 0.1943
all_results.json ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 22.83,
3
+ "eval_accuracy_background": NaN,
4
+ "eval_accuracy_crack": 0.39152312949515705,
5
+ "eval_iou_background": 0.0,
6
+ "eval_iou_crack": 0.3886899926792824,
7
+ "eval_loss": 0.06628751754760742,
8
+ "eval_mean_accuracy": 0.39152312949515705,
9
+ "eval_mean_iou": 0.1943449963396412,
10
+ "eval_overall_accuracy": 0.39152312949515705,
11
+ "eval_runtime": 311.0742,
12
+ "eval_samples_per_second": 0.498,
13
+ "eval_steps_per_second": 0.251,
14
+ "train_loss": 0.04071027485132218,
15
+ "train_runtime": 169673.5427,
16
+ "train_samples_per_second": 0.118,
17
+ "train_steps_per_second": 0.059
18
+ }
eval_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 22.83,
3
+ "eval_accuracy_background": NaN,
4
+ "eval_accuracy_crack": 0.39152312949515705,
5
+ "eval_iou_background": 0.0,
6
+ "eval_iou_crack": 0.3886899926792824,
7
+ "eval_loss": 0.06628751754760742,
8
+ "eval_mean_accuracy": 0.39152312949515705,
9
+ "eval_mean_iou": 0.1943449963396412,
10
+ "eval_overall_accuracy": 0.39152312949515705,
11
+ "eval_runtime": 311.0742,
12
+ "eval_samples_per_second": 0.498,
13
+ "eval_steps_per_second": 0.251
14
+ }
train_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 22.83,
3
+ "train_loss": 0.04071027485132218,
4
+ "train_runtime": 169673.5427,
5
+ "train_samples_per_second": 0.118,
6
+ "train_steps_per_second": 0.059
7
+ }
trainer_state.json ADDED
@@ -0,0 +1,973 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 22.831050228310502,
5
+ "eval_steps": 500,
6
+ "global_step": 10000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.23,
13
+ "learning_rate": 5.9401e-05,
14
+ "loss": 0.0937,
15
+ "step": 100
16
+ },
17
+ {
18
+ "epoch": 0.46,
19
+ "learning_rate": 5.8802000000000004e-05,
20
+ "loss": 0.0673,
21
+ "step": 200
22
+ },
23
+ {
24
+ "epoch": 0.68,
25
+ "learning_rate": 5.8203e-05,
26
+ "loss": 0.0715,
27
+ "step": 300
28
+ },
29
+ {
30
+ "epoch": 0.91,
31
+ "learning_rate": 5.7604e-05,
32
+ "loss": 0.0489,
33
+ "step": 400
34
+ },
35
+ {
36
+ "epoch": 1.0,
37
+ "eval_accuracy_background": NaN,
38
+ "eval_accuracy_crack": 0.29331513143464055,
39
+ "eval_iou_background": 0.0,
40
+ "eval_iou_crack": 0.29288824717786444,
41
+ "eval_loss": 0.06341855227947235,
42
+ "eval_mean_accuracy": 0.29331513143464055,
43
+ "eval_mean_iou": 0.14644412358893222,
44
+ "eval_overall_accuracy": 0.29331513143464055,
45
+ "eval_runtime": 297.3421,
46
+ "eval_samples_per_second": 0.521,
47
+ "eval_steps_per_second": 0.262,
48
+ "step": 438
49
+ },
50
+ {
51
+ "epoch": 1.14,
52
+ "learning_rate": 5.7005e-05,
53
+ "loss": 0.0581,
54
+ "step": 500
55
+ },
56
+ {
57
+ "epoch": 1.37,
58
+ "learning_rate": 5.6406e-05,
59
+ "loss": 0.0521,
60
+ "step": 600
61
+ },
62
+ {
63
+ "epoch": 1.6,
64
+ "learning_rate": 5.5806999999999996e-05,
65
+ "loss": 0.0545,
66
+ "step": 700
67
+ },
68
+ {
69
+ "epoch": 1.83,
70
+ "learning_rate": 5.5208000000000004e-05,
71
+ "loss": 0.0542,
72
+ "step": 800
73
+ },
74
+ {
75
+ "epoch": 2.0,
76
+ "eval_accuracy_background": NaN,
77
+ "eval_accuracy_crack": 0.3916558917583496,
78
+ "eval_iou_background": 0.0,
79
+ "eval_iou_crack": 0.3911766359140811,
80
+ "eval_loss": 0.04389064386487007,
81
+ "eval_mean_accuracy": 0.3916558917583496,
82
+ "eval_mean_iou": 0.19558831795704054,
83
+ "eval_overall_accuracy": 0.3916558917583496,
84
+ "eval_runtime": 307.0821,
85
+ "eval_samples_per_second": 0.505,
86
+ "eval_steps_per_second": 0.254,
87
+ "step": 876
88
+ },
89
+ {
90
+ "epoch": 2.05,
91
+ "learning_rate": 5.4609000000000005e-05,
92
+ "loss": 0.0531,
93
+ "step": 900
94
+ },
95
+ {
96
+ "epoch": 2.28,
97
+ "learning_rate": 5.401e-05,
98
+ "loss": 0.0561,
99
+ "step": 1000
100
+ },
101
+ {
102
+ "epoch": 2.51,
103
+ "learning_rate": 5.3411e-05,
104
+ "loss": 0.0464,
105
+ "step": 1100
106
+ },
107
+ {
108
+ "epoch": 2.74,
109
+ "learning_rate": 5.2812e-05,
110
+ "loss": 0.06,
111
+ "step": 1200
112
+ },
113
+ {
114
+ "epoch": 2.97,
115
+ "learning_rate": 5.2213e-05,
116
+ "loss": 0.0484,
117
+ "step": 1300
118
+ },
119
+ {
120
+ "epoch": 3.0,
121
+ "eval_accuracy_background": NaN,
122
+ "eval_accuracy_crack": 0.3550813313168862,
123
+ "eval_iou_background": 0.0,
124
+ "eval_iou_crack": 0.3438652488711812,
125
+ "eval_loss": 0.04342465475201607,
126
+ "eval_mean_accuracy": 0.3550813313168862,
127
+ "eval_mean_iou": 0.1719326244355906,
128
+ "eval_overall_accuracy": 0.3550813313168862,
129
+ "eval_runtime": 307.3443,
130
+ "eval_samples_per_second": 0.504,
131
+ "eval_steps_per_second": 0.254,
132
+ "step": 1314
133
+ },
134
+ {
135
+ "epoch": 3.2,
136
+ "learning_rate": 5.1614000000000004e-05,
137
+ "loss": 0.0492,
138
+ "step": 1400
139
+ },
140
+ {
141
+ "epoch": 3.42,
142
+ "learning_rate": 5.1015e-05,
143
+ "loss": 0.0448,
144
+ "step": 1500
145
+ },
146
+ {
147
+ "epoch": 3.65,
148
+ "learning_rate": 5.0416e-05,
149
+ "loss": 0.0472,
150
+ "step": 1600
151
+ },
152
+ {
153
+ "epoch": 3.88,
154
+ "learning_rate": 4.9817e-05,
155
+ "loss": 0.0539,
156
+ "step": 1700
157
+ },
158
+ {
159
+ "epoch": 4.0,
160
+ "eval_accuracy_background": NaN,
161
+ "eval_accuracy_crack": 0.38202774154073493,
162
+ "eval_iou_background": 0.0,
163
+ "eval_iou_crack": 0.37412116510013244,
164
+ "eval_loss": 0.04474446550011635,
165
+ "eval_mean_accuracy": 0.38202774154073493,
166
+ "eval_mean_iou": 0.18706058255006622,
167
+ "eval_overall_accuracy": 0.38202774154073493,
168
+ "eval_runtime": 309.7837,
169
+ "eval_samples_per_second": 0.5,
170
+ "eval_steps_per_second": 0.252,
171
+ "step": 1752
172
+ },
173
+ {
174
+ "epoch": 4.11,
175
+ "learning_rate": 4.9218e-05,
176
+ "loss": 0.0521,
177
+ "step": 1800
178
+ },
179
+ {
180
+ "epoch": 4.34,
181
+ "learning_rate": 4.8619e-05,
182
+ "loss": 0.035,
183
+ "step": 1900
184
+ },
185
+ {
186
+ "epoch": 4.57,
187
+ "learning_rate": 4.8020000000000004e-05,
188
+ "loss": 0.0447,
189
+ "step": 2000
190
+ },
191
+ {
192
+ "epoch": 4.79,
193
+ "learning_rate": 4.7421000000000006e-05,
194
+ "loss": 0.0565,
195
+ "step": 2100
196
+ },
197
+ {
198
+ "epoch": 5.0,
199
+ "eval_accuracy_background": NaN,
200
+ "eval_accuracy_crack": 0.39371514990591194,
201
+ "eval_iou_background": 0.0,
202
+ "eval_iou_crack": 0.3776707590042732,
203
+ "eval_loss": 0.043493106961250305,
204
+ "eval_mean_accuracy": 0.39371514990591194,
205
+ "eval_mean_iou": 0.1888353795021366,
206
+ "eval_overall_accuracy": 0.39371514990591194,
207
+ "eval_runtime": 304.996,
208
+ "eval_samples_per_second": 0.508,
209
+ "eval_steps_per_second": 0.256,
210
+ "step": 2190
211
+ },
212
+ {
213
+ "epoch": 5.02,
214
+ "learning_rate": 4.6822e-05,
215
+ "loss": 0.0526,
216
+ "step": 2200
217
+ },
218
+ {
219
+ "epoch": 5.25,
220
+ "learning_rate": 4.6223e-05,
221
+ "loss": 0.05,
222
+ "step": 2300
223
+ },
224
+ {
225
+ "epoch": 5.48,
226
+ "learning_rate": 4.5624e-05,
227
+ "loss": 0.0421,
228
+ "step": 2400
229
+ },
230
+ {
231
+ "epoch": 5.71,
232
+ "learning_rate": 4.5025000000000003e-05,
233
+ "loss": 0.0489,
234
+ "step": 2500
235
+ },
236
+ {
237
+ "epoch": 5.94,
238
+ "learning_rate": 4.4426000000000005e-05,
239
+ "loss": 0.0544,
240
+ "step": 2600
241
+ },
242
+ {
243
+ "epoch": 6.0,
244
+ "eval_accuracy_background": NaN,
245
+ "eval_accuracy_crack": 0.39296475450525853,
246
+ "eval_iou_background": 0.0,
247
+ "eval_iou_crack": 0.38078564516760544,
248
+ "eval_loss": 0.044212136417627335,
249
+ "eval_mean_accuracy": 0.39296475450525853,
250
+ "eval_mean_iou": 0.19039282258380272,
251
+ "eval_overall_accuracy": 0.39296475450525853,
252
+ "eval_runtime": 307.5649,
253
+ "eval_samples_per_second": 0.504,
254
+ "eval_steps_per_second": 0.254,
255
+ "step": 2628
256
+ },
257
+ {
258
+ "epoch": 6.16,
259
+ "learning_rate": 4.3827e-05,
260
+ "loss": 0.0442,
261
+ "step": 2700
262
+ },
263
+ {
264
+ "epoch": 6.39,
265
+ "learning_rate": 4.3228e-05,
266
+ "loss": 0.0637,
267
+ "step": 2800
268
+ },
269
+ {
270
+ "epoch": 6.62,
271
+ "learning_rate": 4.2629e-05,
272
+ "loss": 0.0384,
273
+ "step": 2900
274
+ },
275
+ {
276
+ "epoch": 6.85,
277
+ "learning_rate": 4.203e-05,
278
+ "loss": 0.0421,
279
+ "step": 3000
280
+ },
281
+ {
282
+ "epoch": 7.0,
283
+ "eval_accuracy_background": NaN,
284
+ "eval_accuracy_crack": 0.4651369182992577,
285
+ "eval_iou_background": 0.0,
286
+ "eval_iou_crack": 0.45126808677451646,
287
+ "eval_loss": 0.0448518767952919,
288
+ "eval_mean_accuracy": 0.4651369182992577,
289
+ "eval_mean_iou": 0.22563404338725823,
290
+ "eval_overall_accuracy": 0.4651369182992577,
291
+ "eval_runtime": 305.9247,
292
+ "eval_samples_per_second": 0.507,
293
+ "eval_steps_per_second": 0.255,
294
+ "step": 3066
295
+ },
296
+ {
297
+ "epoch": 7.08,
298
+ "learning_rate": 4.1431e-05,
299
+ "loss": 0.033,
300
+ "step": 3100
301
+ },
302
+ {
303
+ "epoch": 7.31,
304
+ "learning_rate": 4.0832e-05,
305
+ "loss": 0.053,
306
+ "step": 3200
307
+ },
308
+ {
309
+ "epoch": 7.53,
310
+ "learning_rate": 4.0233e-05,
311
+ "loss": 0.0431,
312
+ "step": 3300
313
+ },
314
+ {
315
+ "epoch": 7.76,
316
+ "learning_rate": 3.9634e-05,
317
+ "loss": 0.0486,
318
+ "step": 3400
319
+ },
320
+ {
321
+ "epoch": 7.99,
322
+ "learning_rate": 3.9035e-05,
323
+ "loss": 0.0352,
324
+ "step": 3500
325
+ },
326
+ {
327
+ "epoch": 8.0,
328
+ "eval_accuracy_background": NaN,
329
+ "eval_accuracy_crack": 0.31645328500017317,
330
+ "eval_iou_background": 0.0,
331
+ "eval_iou_crack": 0.3137668997468883,
332
+ "eval_loss": 0.05873505398631096,
333
+ "eval_mean_accuracy": 0.31645328500017317,
334
+ "eval_mean_iou": 0.15688344987344416,
335
+ "eval_overall_accuracy": 0.31645328500017317,
336
+ "eval_runtime": 304.4089,
337
+ "eval_samples_per_second": 0.509,
338
+ "eval_steps_per_second": 0.256,
339
+ "step": 3504
340
+ },
341
+ {
342
+ "epoch": 8.22,
343
+ "learning_rate": 3.8436e-05,
344
+ "loss": 0.0404,
345
+ "step": 3600
346
+ },
347
+ {
348
+ "epoch": 8.45,
349
+ "learning_rate": 3.7837000000000004e-05,
350
+ "loss": 0.0422,
351
+ "step": 3700
352
+ },
353
+ {
354
+ "epoch": 8.68,
355
+ "learning_rate": 3.7238000000000005e-05,
356
+ "loss": 0.0455,
357
+ "step": 3800
358
+ },
359
+ {
360
+ "epoch": 8.9,
361
+ "learning_rate": 3.6639e-05,
362
+ "loss": 0.0394,
363
+ "step": 3900
364
+ },
365
+ {
366
+ "epoch": 9.0,
367
+ "eval_accuracy_background": NaN,
368
+ "eval_accuracy_crack": 0.371040221193475,
369
+ "eval_iou_background": 0.0,
370
+ "eval_iou_crack": 0.3683547749348156,
371
+ "eval_loss": 0.044187385588884354,
372
+ "eval_mean_accuracy": 0.371040221193475,
373
+ "eval_mean_iou": 0.1841773874674078,
374
+ "eval_overall_accuracy": 0.371040221193475,
375
+ "eval_runtime": 307.1965,
376
+ "eval_samples_per_second": 0.505,
377
+ "eval_steps_per_second": 0.254,
378
+ "step": 3942
379
+ },
380
+ {
381
+ "epoch": 9.13,
382
+ "learning_rate": 3.604e-05,
383
+ "loss": 0.0537,
384
+ "step": 4000
385
+ },
386
+ {
387
+ "epoch": 9.36,
388
+ "learning_rate": 3.544100000000001e-05,
389
+ "loss": 0.0358,
390
+ "step": 4100
391
+ },
392
+ {
393
+ "epoch": 9.59,
394
+ "learning_rate": 3.4842e-05,
395
+ "loss": 0.0484,
396
+ "step": 4200
397
+ },
398
+ {
399
+ "epoch": 9.82,
400
+ "learning_rate": 3.4243000000000004e-05,
401
+ "loss": 0.0445,
402
+ "step": 4300
403
+ },
404
+ {
405
+ "epoch": 10.0,
406
+ "eval_accuracy_background": NaN,
407
+ "eval_accuracy_crack": 0.4173194144606966,
408
+ "eval_iou_background": 0.0,
409
+ "eval_iou_crack": 0.2333501709040323,
410
+ "eval_loss": 0.06093364953994751,
411
+ "eval_mean_accuracy": 0.4173194144606966,
412
+ "eval_mean_iou": 0.11667508545201614,
413
+ "eval_overall_accuracy": 0.4173194144606966,
414
+ "eval_runtime": 311.1107,
415
+ "eval_samples_per_second": 0.498,
416
+ "eval_steps_per_second": 0.251,
417
+ "step": 4380
418
+ },
419
+ {
420
+ "epoch": 10.05,
421
+ "learning_rate": 3.3644000000000005e-05,
422
+ "loss": 0.0355,
423
+ "step": 4400
424
+ },
425
+ {
426
+ "epoch": 10.27,
427
+ "learning_rate": 3.3045000000000006e-05,
428
+ "loss": 0.0373,
429
+ "step": 4500
430
+ },
431
+ {
432
+ "epoch": 10.5,
433
+ "learning_rate": 3.2446e-05,
434
+ "loss": 0.0401,
435
+ "step": 4600
436
+ },
437
+ {
438
+ "epoch": 10.73,
439
+ "learning_rate": 3.1847e-05,
440
+ "loss": 0.0414,
441
+ "step": 4700
442
+ },
443
+ {
444
+ "epoch": 10.96,
445
+ "learning_rate": 3.1248e-05,
446
+ "loss": 0.0503,
447
+ "step": 4800
448
+ },
449
+ {
450
+ "epoch": 11.0,
451
+ "eval_accuracy_background": NaN,
452
+ "eval_accuracy_crack": 0.37144860945959984,
453
+ "eval_iou_background": 0.0,
454
+ "eval_iou_crack": 0.3403064320588908,
455
+ "eval_loss": 0.05043744295835495,
456
+ "eval_mean_accuracy": 0.37144860945959984,
457
+ "eval_mean_iou": 0.1701532160294454,
458
+ "eval_overall_accuracy": 0.37144860945959984,
459
+ "eval_runtime": 306.2997,
460
+ "eval_samples_per_second": 0.506,
461
+ "eval_steps_per_second": 0.255,
462
+ "step": 4818
463
+ },
464
+ {
465
+ "epoch": 11.19,
466
+ "learning_rate": 3.0649000000000004e-05,
467
+ "loss": 0.0396,
468
+ "step": 4900
469
+ },
470
+ {
471
+ "epoch": 11.42,
472
+ "learning_rate": 3.0050000000000002e-05,
473
+ "loss": 0.0376,
474
+ "step": 5000
475
+ },
476
+ {
477
+ "epoch": 11.64,
478
+ "learning_rate": 2.9451e-05,
479
+ "loss": 0.0462,
480
+ "step": 5100
481
+ },
482
+ {
483
+ "epoch": 11.87,
484
+ "learning_rate": 2.8851999999999998e-05,
485
+ "loss": 0.0379,
486
+ "step": 5200
487
+ },
488
+ {
489
+ "epoch": 12.0,
490
+ "eval_accuracy_background": NaN,
491
+ "eval_accuracy_crack": 0.38687645028341855,
492
+ "eval_iou_background": 0.0,
493
+ "eval_iou_crack": 0.3806706282436299,
494
+ "eval_loss": 0.046007703989744186,
495
+ "eval_mean_accuracy": 0.38687645028341855,
496
+ "eval_mean_iou": 0.19033531412181495,
497
+ "eval_overall_accuracy": 0.38687645028341855,
498
+ "eval_runtime": 303.4058,
499
+ "eval_samples_per_second": 0.511,
500
+ "eval_steps_per_second": 0.257,
501
+ "step": 5256
502
+ },
503
+ {
504
+ "epoch": 12.1,
505
+ "learning_rate": 2.8253e-05,
506
+ "loss": 0.0342,
507
+ "step": 5300
508
+ },
509
+ {
510
+ "epoch": 12.33,
511
+ "learning_rate": 2.7653999999999996e-05,
512
+ "loss": 0.0374,
513
+ "step": 5400
514
+ },
515
+ {
516
+ "epoch": 12.56,
517
+ "learning_rate": 2.7054999999999998e-05,
518
+ "loss": 0.0403,
519
+ "step": 5500
520
+ },
521
+ {
522
+ "epoch": 12.79,
523
+ "learning_rate": 2.6455999999999995e-05,
524
+ "loss": 0.0405,
525
+ "step": 5600
526
+ },
527
+ {
528
+ "epoch": 13.0,
529
+ "eval_accuracy_background": NaN,
530
+ "eval_accuracy_crack": 0.4084243428267972,
531
+ "eval_iou_background": 0.0,
532
+ "eval_iou_crack": 0.4034361743374538,
533
+ "eval_loss": 0.04515380412340164,
534
+ "eval_mean_accuracy": 0.4084243428267972,
535
+ "eval_mean_iou": 0.2017180871687269,
536
+ "eval_overall_accuracy": 0.4084243428267972,
537
+ "eval_runtime": 305.5019,
538
+ "eval_samples_per_second": 0.507,
539
+ "eval_steps_per_second": 0.255,
540
+ "step": 5694
541
+ },
542
+ {
543
+ "epoch": 13.01,
544
+ "learning_rate": 2.5857e-05,
545
+ "loss": 0.0329,
546
+ "step": 5700
547
+ },
548
+ {
549
+ "epoch": 13.24,
550
+ "learning_rate": 2.5258e-05,
551
+ "loss": 0.0345,
552
+ "step": 5800
553
+ },
554
+ {
555
+ "epoch": 13.47,
556
+ "learning_rate": 2.4659e-05,
557
+ "loss": 0.0402,
558
+ "step": 5900
559
+ },
560
+ {
561
+ "epoch": 13.7,
562
+ "learning_rate": 2.406e-05,
563
+ "loss": 0.0353,
564
+ "step": 6000
565
+ },
566
+ {
567
+ "epoch": 13.93,
568
+ "learning_rate": 2.3460999999999998e-05,
569
+ "loss": 0.0367,
570
+ "step": 6100
571
+ },
572
+ {
573
+ "epoch": 14.0,
574
+ "eval_accuracy_background": NaN,
575
+ "eval_accuracy_crack": 0.40603606515740986,
576
+ "eval_iou_background": 0.0,
577
+ "eval_iou_crack": 0.39904072521400735,
578
+ "eval_loss": 0.047693103551864624,
579
+ "eval_mean_accuracy": 0.40603606515740986,
580
+ "eval_mean_iou": 0.19952036260700368,
581
+ "eval_overall_accuracy": 0.40603606515740986,
582
+ "eval_runtime": 305.7852,
583
+ "eval_samples_per_second": 0.507,
584
+ "eval_steps_per_second": 0.255,
585
+ "step": 6132
586
+ },
587
+ {
588
+ "epoch": 14.16,
589
+ "learning_rate": 2.2862e-05,
590
+ "loss": 0.0398,
591
+ "step": 6200
592
+ },
593
+ {
594
+ "epoch": 14.38,
595
+ "learning_rate": 2.2263e-05,
596
+ "loss": 0.0369,
597
+ "step": 6300
598
+ },
599
+ {
600
+ "epoch": 14.61,
601
+ "learning_rate": 2.1663999999999998e-05,
602
+ "loss": 0.0318,
603
+ "step": 6400
604
+ },
605
+ {
606
+ "epoch": 14.84,
607
+ "learning_rate": 2.1065e-05,
608
+ "loss": 0.0315,
609
+ "step": 6500
610
+ },
611
+ {
612
+ "epoch": 15.0,
613
+ "eval_accuracy_background": NaN,
614
+ "eval_accuracy_crack": 0.42079720852910957,
615
+ "eval_iou_background": 0.0,
616
+ "eval_iou_crack": 0.41468731956915605,
617
+ "eval_loss": 0.049836643040180206,
618
+ "eval_mean_accuracy": 0.42079720852910957,
619
+ "eval_mean_iou": 0.20734365978457803,
620
+ "eval_overall_accuracy": 0.42079720852910957,
621
+ "eval_runtime": 305.8824,
622
+ "eval_samples_per_second": 0.507,
623
+ "eval_steps_per_second": 0.255,
624
+ "step": 6570
625
+ },
626
+ {
627
+ "epoch": 15.07,
628
+ "learning_rate": 2.0465999999999997e-05,
629
+ "loss": 0.0386,
630
+ "step": 6600
631
+ },
632
+ {
633
+ "epoch": 15.3,
634
+ "learning_rate": 1.9866999999999998e-05,
635
+ "loss": 0.0481,
636
+ "step": 6700
637
+ },
638
+ {
639
+ "epoch": 15.53,
640
+ "learning_rate": 1.9267999999999996e-05,
641
+ "loss": 0.0322,
642
+ "step": 6800
643
+ },
644
+ {
645
+ "epoch": 15.75,
646
+ "learning_rate": 1.8669e-05,
647
+ "loss": 0.0339,
648
+ "step": 6900
649
+ },
650
+ {
651
+ "epoch": 15.98,
652
+ "learning_rate": 1.807e-05,
653
+ "loss": 0.0244,
654
+ "step": 7000
655
+ },
656
+ {
657
+ "epoch": 16.0,
658
+ "eval_accuracy_background": NaN,
659
+ "eval_accuracy_crack": 0.40287863220235276,
660
+ "eval_iou_background": 0.0,
661
+ "eval_iou_crack": 0.3926174166302197,
662
+ "eval_loss": 0.04859909415245056,
663
+ "eval_mean_accuracy": 0.40287863220235276,
664
+ "eval_mean_iou": 0.19630870831510985,
665
+ "eval_overall_accuracy": 0.40287863220235276,
666
+ "eval_runtime": 306.6311,
667
+ "eval_samples_per_second": 0.505,
668
+ "eval_steps_per_second": 0.254,
669
+ "step": 7008
670
+ },
671
+ {
672
+ "epoch": 16.21,
673
+ "learning_rate": 1.7471e-05,
674
+ "loss": 0.0309,
675
+ "step": 7100
676
+ },
677
+ {
678
+ "epoch": 16.44,
679
+ "learning_rate": 1.6872e-05,
680
+ "loss": 0.0408,
681
+ "step": 7200
682
+ },
683
+ {
684
+ "epoch": 16.67,
685
+ "learning_rate": 1.6272999999999998e-05,
686
+ "loss": 0.0298,
687
+ "step": 7300
688
+ },
689
+ {
690
+ "epoch": 16.89,
691
+ "learning_rate": 1.5674e-05,
692
+ "loss": 0.031,
693
+ "step": 7400
694
+ },
695
+ {
696
+ "epoch": 17.0,
697
+ "eval_accuracy_background": NaN,
698
+ "eval_accuracy_crack": 0.38919257454889694,
699
+ "eval_iou_background": 0.0,
700
+ "eval_iou_crack": 0.3854711873515346,
701
+ "eval_loss": 0.05684218183159828,
702
+ "eval_mean_accuracy": 0.38919257454889694,
703
+ "eval_mean_iou": 0.1927355936757673,
704
+ "eval_overall_accuracy": 0.38919257454889694,
705
+ "eval_runtime": 307.265,
706
+ "eval_samples_per_second": 0.504,
707
+ "eval_steps_per_second": 0.254,
708
+ "step": 7446
709
+ },
710
+ {
711
+ "epoch": 17.12,
712
+ "learning_rate": 1.5075000000000002e-05,
713
+ "loss": 0.0348,
714
+ "step": 7500
715
+ },
716
+ {
717
+ "epoch": 17.35,
718
+ "learning_rate": 1.4476e-05,
719
+ "loss": 0.0345,
720
+ "step": 7600
721
+ },
722
+ {
723
+ "epoch": 17.58,
724
+ "learning_rate": 1.3877e-05,
725
+ "loss": 0.0352,
726
+ "step": 7700
727
+ },
728
+ {
729
+ "epoch": 17.81,
730
+ "learning_rate": 1.3277999999999999e-05,
731
+ "loss": 0.0288,
732
+ "step": 7800
733
+ },
734
+ {
735
+ "epoch": 18.0,
736
+ "eval_accuracy_background": NaN,
737
+ "eval_accuracy_crack": 0.40924400549520323,
738
+ "eval_iou_background": 0.0,
739
+ "eval_iou_crack": 0.4066834831665307,
740
+ "eval_loss": 0.056042663753032684,
741
+ "eval_mean_accuracy": 0.40924400549520323,
742
+ "eval_mean_iou": 0.20334174158326535,
743
+ "eval_overall_accuracy": 0.40924400549520323,
744
+ "eval_runtime": 308.805,
745
+ "eval_samples_per_second": 0.502,
746
+ "eval_steps_per_second": 0.253,
747
+ "step": 7884
748
+ },
749
+ {
750
+ "epoch": 18.04,
751
+ "learning_rate": 1.2678999999999998e-05,
752
+ "loss": 0.0262,
753
+ "step": 7900
754
+ },
755
+ {
756
+ "epoch": 18.26,
757
+ "learning_rate": 1.2079999999999998e-05,
758
+ "loss": 0.0298,
759
+ "step": 8000
760
+ },
761
+ {
762
+ "epoch": 18.49,
763
+ "learning_rate": 1.1480999999999997e-05,
764
+ "loss": 0.0345,
765
+ "step": 8100
766
+ },
767
+ {
768
+ "epoch": 18.72,
769
+ "learning_rate": 1.0882000000000004e-05,
770
+ "loss": 0.0268,
771
+ "step": 8200
772
+ },
773
+ {
774
+ "epoch": 18.95,
775
+ "learning_rate": 1.0283000000000003e-05,
776
+ "loss": 0.0354,
777
+ "step": 8300
778
+ },
779
+ {
780
+ "epoch": 19.0,
781
+ "eval_accuracy_background": NaN,
782
+ "eval_accuracy_crack": 0.4055771695085487,
783
+ "eval_iou_background": 0.0,
784
+ "eval_iou_crack": 0.4013274197279475,
785
+ "eval_loss": 0.061260469257831573,
786
+ "eval_mean_accuracy": 0.4055771695085487,
787
+ "eval_mean_iou": 0.20066370986397375,
788
+ "eval_overall_accuracy": 0.4055771695085487,
789
+ "eval_runtime": 306.2991,
790
+ "eval_samples_per_second": 0.506,
791
+ "eval_steps_per_second": 0.255,
792
+ "step": 8322
793
+ },
794
+ {
795
+ "epoch": 19.18,
796
+ "learning_rate": 9.684000000000002e-06,
797
+ "loss": 0.0241,
798
+ "step": 8400
799
+ },
800
+ {
801
+ "epoch": 19.41,
802
+ "learning_rate": 9.085000000000002e-06,
803
+ "loss": 0.0175,
804
+ "step": 8500
805
+ },
806
+ {
807
+ "epoch": 19.63,
808
+ "learning_rate": 8.486000000000001e-06,
809
+ "loss": 0.0405,
810
+ "step": 8600
811
+ },
812
+ {
813
+ "epoch": 19.86,
814
+ "learning_rate": 7.887000000000001e-06,
815
+ "loss": 0.0315,
816
+ "step": 8700
817
+ },
818
+ {
819
+ "epoch": 20.0,
820
+ "eval_accuracy_background": NaN,
821
+ "eval_accuracy_crack": 0.37516018055667794,
822
+ "eval_iou_background": 0.0,
823
+ "eval_iou_crack": 0.3730927314267345,
824
+ "eval_loss": 0.0605228915810585,
825
+ "eval_mean_accuracy": 0.37516018055667794,
826
+ "eval_mean_iou": 0.18654636571336725,
827
+ "eval_overall_accuracy": 0.37516018055667794,
828
+ "eval_runtime": 306.4581,
829
+ "eval_samples_per_second": 0.506,
830
+ "eval_steps_per_second": 0.255,
831
+ "step": 8760
832
+ },
833
+ {
834
+ "epoch": 20.09,
835
+ "learning_rate": 7.2879999999999995e-06,
836
+ "loss": 0.0316,
837
+ "step": 8800
838
+ },
839
+ {
840
+ "epoch": 20.32,
841
+ "learning_rate": 6.688999999999999e-06,
842
+ "loss": 0.0305,
843
+ "step": 8900
844
+ },
845
+ {
846
+ "epoch": 20.55,
847
+ "learning_rate": 6.0899999999999984e-06,
848
+ "loss": 0.0232,
849
+ "step": 9000
850
+ },
851
+ {
852
+ "epoch": 20.78,
853
+ "learning_rate": 5.490999999999998e-06,
854
+ "loss": 0.0343,
855
+ "step": 9100
856
+ },
857
+ {
858
+ "epoch": 21.0,
859
+ "eval_accuracy_background": NaN,
860
+ "eval_accuracy_crack": 0.40193775181537966,
861
+ "eval_iou_background": 0.0,
862
+ "eval_iou_crack": 0.39810018752286863,
863
+ "eval_loss": 0.06526383757591248,
864
+ "eval_mean_accuracy": 0.40193775181537966,
865
+ "eval_mean_iou": 0.19905009376143432,
866
+ "eval_overall_accuracy": 0.40193775181537966,
867
+ "eval_runtime": 308.2848,
868
+ "eval_samples_per_second": 0.503,
869
+ "eval_steps_per_second": 0.253,
870
+ "step": 9198
871
+ },
872
+ {
873
+ "epoch": 21.0,
874
+ "learning_rate": 4.891999999999997e-06,
875
+ "loss": 0.0315,
876
+ "step": 9200
877
+ },
878
+ {
879
+ "epoch": 21.23,
880
+ "learning_rate": 4.292999999999997e-06,
881
+ "loss": 0.0274,
882
+ "step": 9300
883
+ },
884
+ {
885
+ "epoch": 21.46,
886
+ "learning_rate": 3.694000000000003e-06,
887
+ "loss": 0.0257,
888
+ "step": 9400
889
+ },
890
+ {
891
+ "epoch": 21.69,
892
+ "learning_rate": 3.0950000000000026e-06,
893
+ "loss": 0.0293,
894
+ "step": 9500
895
+ },
896
+ {
897
+ "epoch": 21.92,
898
+ "learning_rate": 2.496000000000002e-06,
899
+ "loss": 0.0327,
900
+ "step": 9600
901
+ },
902
+ {
903
+ "epoch": 22.0,
904
+ "eval_accuracy_background": NaN,
905
+ "eval_accuracy_crack": 0.3924438069290357,
906
+ "eval_iou_background": 0.0,
907
+ "eval_iou_crack": 0.38908561282725923,
908
+ "eval_loss": 0.06604981422424316,
909
+ "eval_mean_accuracy": 0.3924438069290357,
910
+ "eval_mean_iou": 0.19454280641362962,
911
+ "eval_overall_accuracy": 0.3924438069290357,
912
+ "eval_runtime": 307.3272,
913
+ "eval_samples_per_second": 0.504,
914
+ "eval_steps_per_second": 0.254,
915
+ "step": 9636
916
+ },
917
+ {
918
+ "epoch": 22.15,
919
+ "learning_rate": 1.8970000000000013e-06,
920
+ "loss": 0.0282,
921
+ "step": 9700
922
+ },
923
+ {
924
+ "epoch": 22.37,
925
+ "learning_rate": 1.298000000000001e-06,
926
+ "loss": 0.0296,
927
+ "step": 9800
928
+ },
929
+ {
930
+ "epoch": 22.6,
931
+ "learning_rate": 6.990000000000005e-07,
932
+ "loss": 0.0277,
933
+ "step": 9900
934
+ },
935
+ {
936
+ "epoch": 22.83,
937
+ "learning_rate": 1e-07,
938
+ "loss": 0.0252,
939
+ "step": 10000
940
+ },
941
+ {
942
+ "epoch": 22.83,
943
+ "eval_accuracy_background": NaN,
944
+ "eval_accuracy_crack": 0.39152312949515705,
945
+ "eval_iou_background": 0.0,
946
+ "eval_iou_crack": 0.3886899926792824,
947
+ "eval_loss": 0.06628751754760742,
948
+ "eval_mean_accuracy": 0.39152312949515705,
949
+ "eval_mean_iou": 0.1943449963396412,
950
+ "eval_overall_accuracy": 0.39152312949515705,
951
+ "eval_runtime": 306.5247,
952
+ "eval_samples_per_second": 0.506,
953
+ "eval_steps_per_second": 0.254,
954
+ "step": 10000
955
+ },
956
+ {
957
+ "epoch": 22.83,
958
+ "step": 10000,
959
+ "total_flos": 7.98337466892288e+18,
960
+ "train_loss": 0.04071027485132218,
961
+ "train_runtime": 169673.5427,
962
+ "train_samples_per_second": 0.118,
963
+ "train_steps_per_second": 0.059
964
+ }
965
+ ],
966
+ "logging_steps": 100,
967
+ "max_steps": 10000,
968
+ "num_train_epochs": 23,
969
+ "save_steps": 500,
970
+ "total_flos": 7.98337466892288e+18,
971
+ "trial_name": null,
972
+ "trial_params": null
973
+ }